Patent Grant
8995197
Flash memory device cells are used widely in the industry as memory storage. The flash memory device cell must be able to program, erase and retain charge on its floating gate. During program or erase the flash memory device cell is subject to high electric field to support oxide tunneling and hot electron injection. Those are the mechanisms that may cause reliability failure. Most of the available digital integrated circuits (IC) operate on rail-to-rail logic, flash memory device operation is distinctively analog. The digitally stored information induces analog distribution of the threshold voltage for read operations. In multi-level flash memory device cells, multiple bits are stored by precise placement of threshold voltage in each cell. The readout of the threshold voltage requires precise placement of the sense amplifier decision threshold.
The voltage programming and readout are statistical, and there is therefore a probability of error in each decision. The use of error correction codes (ECC) enables reduction of the error rate to an acceptable value, e.g. 10−15.
There are currently three common methods of programming flash memory device memory: channel electron ejection, Fowler-Nordhiem (F-N) tunneling from the source or drain, and F-N tunneling from the channel. There are also three common methods for erasing data from the flash memory device memory: F-N tunneling through the channel, F-N tunneling through the source or drain, and F-N tunneling to the floating gate. These techniques, however, require a high electric field in the oxide and may create traps and leakage current.
The result of performing program and erase (P/E) cycles with a fixed set of flash memory device parameters may be highly inefficient. Currently known flash memory device s may be using a single set of program and erase parameters to control the desired program distribution. A significant increase of cycle count for a given margin may be obtained by adapting program and erase parameters to the effective endurance state of the flash memory device.
Another effect of the P/E cycles is the increase of the threshold variance. As the number of cycles increases, the number of traps also increases. The average number of traps reduces the threshold window as discussed above. However, it also increases the threshold variance for every level in the program. The relation to the number of cycles is derived, for example, in an empirical model described in: Mielke, N. Belgal, H. Kalastirsky, I. Kalavade, P. Kurtz, A. Meng, Q. Righos, N. Wu, J. “flash memory device EEPROM Threshold Instabilities Due to Charge Trapping During Programming and/or erase Cycling”. IEEE Transactions on Device and Materials Reliability, Vol. 4, No. 3, September 2004, p 335-344, which is incorporated herein in its entirety by reference. The bit error rate of a given flash memory device memory may be related to the threshold voltage window and to the threshold voltage variance at the highest cycle count.
The probability of error may be approximated in different ways, as to allow analytical modelling of the flash memory device cells. One approximation for calculating the probability of error is a Gaussian distribution of the threshold voltage: Where:
The minimum threshold voltage window and the maximum threshold voltage variance may be measured at the maximum cycle counts of a given device (100,000 at Single Level Cell, 1,000 at Multi-Level Cell and 1,000 at 3 bits per cell). The device bit error rate (BER) may therefore not be constant with respect to the cycle count and the numbers of errors may constantly increase as the number of cycle increases. The flash memory device memory controller ECC is designed to correct the highest number of errors at the maximum number of cycles.
There is a growing need to provide computer readable media, methods and systems for increasing the lifespan of flash memory devices.
According to an embodiment of the invention a method may be provided and may include receiving or generating an estimate of a wear level of at least one group of flash memory cells of the flash memory device; and finding erase parameters and programming parameters to be applied on one or more groups of flash memory cells of the flash memory device in response to estimate of the wear level.
The finding may include evaluating a plurality of sets of evaluated parameters that differ from each other by a value of at least one parameter; wherein each one of the plurality of sets of evaluated parameters comprise at least one out of a set of erase parameters and a set of programming parameters.
The method may include evaluating sets of erase parameters by erasing different groups of flash memory blocks of the flash memory device that have substantially a same weal level.
The method may include evaluating sets of programming parameters by programming different groups of flash memory blocks of the flash memory device that have substantially a same weal level.
Values of the evaluated parameters of each of the different sets of evaluated parameters may be within sub-ranges of possible ranges of values of the evaluated parameters.
The method may include finding a set of evaluated parameters that minimize a formation of traps.
The method may include finding a set of erase parameters that minimize a wear of the group of flash memory cells resulting from erase processes.
The method may include finding a set of erase parameters that may include at least one of a minimal start erase voltage, minimal number of maximal allowable erase pulses, minimal erase pulse width, minimal erase voltage step, erase verify level, maximal number of non-erased cells and maximal erase voltage level.
The method may include finding a set of programming parameters that may include at least one of a maximal start programming voltage, minimal number of maximal allowable programming pulses, minimal programming voltage step, minimal programming pulse width, program verify levels, maximal number of non-verified cells, and minimal program bias voltage.
The method may include finding a selected set of evaluated parameters that has a smallest effect on the wear of the group of flash memory cells out of a plurality of sets of evaluated parameters that once applied introduce a tolerable deviation in a voltage threshold distribution.
The method may include evaluating sets of evaluated parameters wherein each set of erase parameters is sub-optimal in a context of voltage threshold distribution.
The method may include finding a set of erase parameters in response to duration attribute of erase operations to be obtained when applying the set of erase parameters.
The method may include ignoring sets of erase parameters that once applied result in an average duration of erase operations that exceeds an erase duration threshold.
The method may include altering a value of the erase duration threshold in response to values of the wear level.
The method may include finding a set of programming parameters in response to duration attribute of programming operations to be obtained when applying the set of programming parameters.
The method may include ignoring sets of programming parameters that once applied result in an average duration of programming operations that exceeds a programming duration threshold.
The method may include altering a value of the programming length threshold in response to values of the wear level.
The evaluating of the plurality of sets of evaluated parameters may include calculating a score per each set of evaluated parameter.
A score of a set of evaluated parameter is responsive to number of corrected read errors resulting from a read operation that follows a programming operation that was executed by applying programming parameters of the set of the evaluated parameter.
The method may include calculating a score of a set of evaluated parameters in response to at least one value of at least one parameter of the set of evaluated parameters.
The method may include calculating a score of a set of evaluated parameters in response to the wear level.
The method may include calculating a score of a set of evaluated parameters in response to statistical attribute of the voltage threshold distribution resulting from applying the set of evaluated parameters.
Possible values of the estimate of the wear level belong to a range of wear levels that may include multiple non-overlapping sub-ranges of wear levels; wherein the finding is executed at least once for each one for each sub-range of wear levels.
The method may include finding at a beginning of each sub-range of wear levels, erase parameters and programming parameters to be applied during the entire sub-range of wear levels.
Sub-ranges of higher order may be associated with higher wear levels; wherein the method may include modifying at least one parameter out of a number of maximal allowable programming pulses, start program voltage, programming voltage step, programming window, number of non-verified programmed cells, program pulse width and program bias voltage with an increase of an order to the sub-range.
Sub-ranges of higher order may be associated with higher wear levels; wherein the method may include modifying at least one parameter out of a start erase voltage, a number of maximal allowable erase pulses, an erase voltage step and erase pulse width with an increase of an order to the sub-range.
The method may include finding a set of evaluated parameters that once applied results in manageable level of bit error rates that exceeds a minimal allowable level of bit error rates.
The method may include selecting a set of evaluated parameters that once applied results in manageable level of bit error rates and has a smallest contribution to a wear level of the at least one group of flash memory cells.
The method may include selecting a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
The method may include selecting a set of evaluated parameters that once applied results in (a) a higher bit error rate and a lower programming duration at low wear levels and (b) a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
The method may include selecting a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and at a lower increment rate of bit error rates per wear level in relation to a reference set of parameters that is expected to be applied regardless of a wear level of the at least one group of flash memory cells.
Each set of evaluated parameters may include erase and programming parameters of values that once applied will not result in a corruption of data programmed to the flash memory device.
Each set of evaluated parameters may include programming parameters that may be applied during programming operations of information that is not related to the finding of the erase parameters and the programming parameters.
The finding may include repetitively programming same information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other.
The method may include repetitively programming the same information to multiple groups of flash memory cells is a probability of a programming operation failure exceeds a threshold.
The finding may include repetitively programming same pilot information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other.
The finding may include applying a steepest decent algorithm.
The finding may include changing a value of a single evaluated parameter between one set of evaluated parameter to another.
The method may include: evaluating a first set of evaluated parameters; calculating a first score; changing a single parameter of the first set to provide a second set of evaluated parameters that may include a single changed parameter; wherein the changing is performed along a first change direction; evaluating the second set of evaluated parameters; calculating a second score; determining, in response to the first and second scores, whether the first set of evaluated parameters achieves better results than the second set of evaluated parameters; if it is determined that the first set of evaluated parameters achieves better results than the second set of evaluated parameters then changing the single changed parameter along a second change direction that differs from the first change direction to provide a third set of evaluated parameters; if it is determined that the first set of evaluated parameters achieves worse results than the second set of evaluated parameters then changing the single changed parameter along the first direction to provide the third set of evaluated parameters.
The method may include evaluating a single set of evaluated parameters based upon an outcome of multiple read attempts.
The finding is triggered when a bit error rate, which is calculated during a read operation, is associated with information programmed to at least one group of flash memory cells reaches a predetermined value.
The method may include finding a set of erase parameters that once applied minimizes a wear level increment resulting from a programming operation.
Further embodiments of the invention include a computer readable medium that is non-transitory and may store instructions for performing the above-described methods and any steps thereof, including any combinations of same. For example, the computer readable medium may store instructions for receiving or generating an estimate of a wear level of at least one group of flash memory cells of the flash memory device; and finding erase parameters and programming parameters to be applied on one or more groups of flash memory cells of the flash memory device in response to estimate of the wear level.
The non-transitory computer readable medium may store instructions for finding that may include evaluating a plurality of sets of evaluated parameters that differ from each other by a value of at least one parameter; wherein each one of the plurality of sets of evaluated parameters comprise at least one out of a set of erase parameters and a set of programming parameters.
The non-transitory computer readable medium may store instructions for evaluating sets of erase parameters by erasing different groups of flash memory blocks of the flash memory device that have substantially a same weal level.
The non-transitory computer readable medium may store instructions for evaluating sets of programming parameters by programming different groups of flash memory blocks of the flash memory device that have substantially a same weal level.
Values of the evaluated parameters of each of the different sets of evaluated parameters may be within sub-ranges of possible ranges of values of the evaluated parameters.
The non-transitory computer readable medium may store instructions for finding a set of evaluated parameters that minimize a formation of traps.
The non-transitory computer readable medium may store instructions for finding a set of erase parameters that minimize a wear of the group of flash memory cells resulting from erase processes.
The non-transitory computer readable medium may store instructions for finding a set of erase parameters that may include at least one of a minimal start erase voltage, minimal number of maximal allowable erase pulses, minimal erase pulse width, minimal erase voltage step, erase verify level, maximal number of non-erased cells and maximal erase voltage level.
The non-transitory computer readable medium may store instructions for finding a set of programming parameters that may include at least one of a maximal start programming voltage, minimal number of maximal allowable programming pulses, minimal programming voltage step, minimal programming pulse width, program verify levels, maximal number of non-verified cells, and minimal program bias voltage.
The non-transitory computer readable medium may store instructions for finding a selected set of evaluated parameters that has a smallest effect on the wear of the group of flash memory cells out of a plurality of sets of evaluated parameters that once applied introduce a tolerable deviation in a voltage threshold distribution.
The non-transitory computer readable medium may store instructions for evaluating sets of evaluated parameters wherein each set of erase parameters is sub-optimal in a context of voltage threshold distribution.
The non-transitory computer readable medium may store instructions for finding a set of erase parameters in response to duration attribute of erase operations to be obtained when applying the set of erase parameters.
The non-transitory computer readable medium may store instructions for ignoring sets of erase parameters that once applied result in an average duration of erase operations that exceeds an erase duration threshold.
The non-transitory computer readable medium may store instructions for altering a value of the erase duration threshold in response to values of the wear level.
The non-transitory computer readable medium may store instructions for finding a set of programming parameters in response to duration attribute of programming operations to be obtained when applying the set of programming parameters.
The non-transitory computer readable medium may store instructions for ignoring sets of programming parameters that once applied result in an average duration of programming operations that exceeds a programming duration threshold.
The non-transitory computer readable medium may store instructions for altering a value of the programming length threshold in response to values of the wear level.
The evaluating of the plurality of sets of evaluated parameters may include calculating a score per each set of evaluated parameter.
A score of a set of evaluated parameter is responsive to number of corrected read errors resulting from a read operation that follows a programming operation that was executed by applying programming parameters of the set of the evaluated parameter.
The non-transitory computer readable medium may store instructions for calculating a score of a set of evaluated parameters in response to at least one value of at least one parameter of the set of evaluated parameters.
The non-transitory computer readable medium may store instructions for calculating a score of a set of evaluated parameters in response to the wear level.
The non-transitory computer readable medium may store instructions for calculating a score of a set of evaluated parameters in response to statistical attribute of the voltage threshold distribution resulting from applying the set of evaluated parameters.
Possible values of the estimate of the wear level belong to a range of wear levels that may include multiple non-overlapping sub-ranges of wear levels; wherein the finding is executed at least once for each one for each sub-range of wear levels.
The non-transitory computer readable medium may store instructions for finding at a beginning of each sub-range of wear levels, erase parameters and programming parameters to be applied during the entire sub-range of wear levels.
Sub-ranges of higher order may be associated with higher wear levels; wherein The non-transitory computer readable medium may store instructions for modifying at least one parameter out of a number of maximal allowable programming pulses, start program voltage, programming voltage step, programming window, number of non-verified programmed cells, program pulse width and program bias voltage with an increase of an order to the sub-range.
Sub-ranges of higher order may be associated with higher wear levels. The non-transitory computer readable medium may store instructions for modifying at least one parameter out of a start erase voltage, a number of maximal allowable erase pulses, an erase voltage step and erase pulse width with an increase of an order to the sub-range.
The non-transitory computer readable medium may store instructions for finding a set of evaluated parameters that once applied results in manageable level of bit error rates that exceeds a minimal allowable level of bit error rates.
The non-transitory computer readable medium may store instructions for selecting a set of evaluated parameters that once applied results in manageable level of bit error rates and has a smallest contribution to a wear level of the at least one group of flash memory cells.
The non-transitory computer readable medium may store instructions for selecting a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
The non-transitory computer readable medium may store instructions for selecting a set of evaluated parameters that once applied results in (a) a higher bit error rate and a lower programming duration at low wear levels and (b) a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
The non-transitory computer readable medium may store instructions for selecting a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and at a lower increment rate of bit error rates per wear level in relation to a reference set of parameters that is expected to be applied regardless of a wear level of the at least one group of flash memory cells.
each set of evaluated parameters may include erase and programming parameters of values that once applied will not result in a corruption of data programmed to the flash memory device.
Each set of evaluated parameters may include programming parameters that may be applied during programming operations of information that is not related to the finding of the erase parameters and the programming parameters.
The finding may include repetitively programming same information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other.
The non-transitory computer readable medium may store instructions for repetitively programming the same information to multiple groups of flash memory cells is a probability of a programming operation failure exceeds a threshold.
The finding may include repetitively programming same pilot information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other.
The finding may include applying a steepest decent algorithm.
The finding may include changing a value of a single evaluated parameter between one set of evaluated parameter to another.
The non-transitory computer readable medium may store instructions for evaluating a first set of evaluated parameters; calculating a first score; changing a single parameter of the first set to provide a second set of evaluated parameters that may include a single changed parameter; wherein the changing is performed along a first change direction; evaluating the second set of evaluated parameters; calculating a second score; determining, in response to the first and second scores, whether the first set of evaluated parameters achieves better results than the second set of evaluated parameters; if it is determined that the first set of evaluated parameters achieves better results than the second set of evaluated parameters then changing the single changed parameter along a second change direction that differs from the first change direction to provide a third set of evaluated parameters; if it is determined that the first set of evaluated parameters achieves worse results than the second set of evaluated parameters then changing the single changed parameter along the first direction to provide the third set of evaluated parameters.
The non-transitory computer readable medium may store instructions for evaluating a single set of evaluated parameters based upon an outcome of multiple read attempts.
The finding is triggered when a bit error rate, which is calculated during a read operation, is associated with information programmed to at least one group of flash memory cells reaches a predetermined value.
The non-transitory computer readable medium may store instructions for finding a set of erase parameters that once applied minimizes a wear level increment resulting from a programming operation.
Additional embodiments of the invention include a system arranged to execute any or all of the methods described above, including any stages and any combinations of same. For example, the system may include a flash memory that may be arranged to receive or generate an estimate of a wear level of at least one group of flash memory cells of the flash memory device; and find erase parameters and programming parameters to be applied on one or more groups of flash memory cells of the flash memory device in response to estimate of the wear level.
The finding may include evaluate a plurality of sets of evaluated parameters that differ from each other by a value of at least one parameter; wherein each one of the plurality of sets of evaluated parameters comprise at least one out of a set of erase parameters and a set of programming parameters.
The flash memory controller may be arranged to evaluate sets of erase parameters by erasing different groups of flash memory blocks of the flash memory device that have substantially a same weal level.
The flash memory controller may be arranged to evaluate sets of programming parameters by programming different groups of flash memory blocks of the flash memory device that have substantially a same weal level.
Values of the evaluated parameters of each of the different sets of evaluated parameters may be within sub-ranges of possible ranges of values of the evaluated parameters.
The flash memory controller may be arranged to find a set of evaluated parameters that minimize a formation of traps.
The flash memory controller may be arranged to find a set of erase parameters that minimize a wear of the group of flash memory cells resulting from erase processes.
The flash memory controller may be arranged to find a set of erase parameters that may include at least one of a minimal start erase voltage, minimal number of maximal allowable erase pulses, minimal erase pulse width, minimal erase voltage step, erase verify level, maximal number of non-erased cells and maximal erase voltage level.
The flash memory controller may be arranged to find a set of programming parameters that may include at least one of a maximal start programming voltage, minimal number of maximal allowable programming pulses, minimal programming voltage step, minimal programming pulse width, program verify levels, maximal number of non-verified cells, and minimal program bias voltage.
The flash memory controller may be arranged to find a selected set of evaluated parameters that has a smallest effect on the wear of the group of flash memory cells out of a plurality of sets of evaluated parameters that once applied introduce a tolerable deviation in a voltage threshold distribution.
The flash memory controller may be arranged to evaluate sets of evaluated parameters wherein each set of erase parameters is sub-optimal in a context of voltage threshold distribution.
The flash memory controller may be arranged to find a set of erase parameters in response to duration attribute of erase operations to be obtained when applying the set of erase parameters.
The flash memory controller may be arranged to ignore sets of erase parameters that once applied result in an average duration of erase operations that exceeds an erase duration threshold.
The flash memory controller may be arranged to alter a value of the erase duration threshold in response to values of the wear level.
The flash memory controller may be arranged to find a set of programming parameters in response to duration attribute of programming operations to be obtained when applying the set of programming parameters.
The flash memory controller may be arranged to ignore sets of programming parameters that once applied result in an average duration of programming operations that exceeds a programming duration threshold.
The flash memory controller may be arranged to alter a value of the programming length threshold in response to values of the wear level.
The evaluating of the plurality of sets of evaluated parameters may include calculate a score per each set of evaluated parameter.
A score of a set of evaluated parameter is responsive to number of corrected read errors resulting from a read operation that follows a programming operation that was executed by applying programming parameters of the set of the evaluated parameter.
The flash memory controller may be arranged to calculate a score of a set of evaluated parameters in response to at least one value of at least one parameter of the set of evaluated parameters.
The flash memory controller may be arranged to calculate a score of a set of evaluated parameters in response to the wear level.
The flash memory controller may be arranged to calculate a score of a set of evaluated parameters in response to statistical attribute of the voltage threshold distribution resulting from applying the set of evaluated parameters.
Possible values of the estimate of the wear level belong to a range of wear levels that may include multiple non-overlapping sub-ranges of wear levels; wherein the finding is executed at least once for each one for each sub-range of wear levels.
The flash memory controller may be arranged to find at a beginning of each sub-range of wear levels, erase parameters and programming parameters to be applied during the entire sub-range of wear levels.
Sub-ranges of higher order may be associated with higher wear levels; wherein The flash memory controller may be arranged to modify at least one parameter out of a number of maximal allowable programming pulses, start program voltage, programming voltage step, programming window, number of non-verified programmed cells, program pulse width and program bias voltage with an increase of an order to the sub-range.
Sub-ranges of higher order may be associated with higher wear levels; wherein The flash memory controller may be arranged to modify at least one parameter out of a start erase voltage, a number of maximal allowable erase pulses, an erase voltage step and erase pulse width with an increase of an order to the sub-range.
The flash memory controller may be arranged to find a set of evaluated parameters that once applied results in manageable level of bit error rates that exceeds a minimal allowable level of bit error rates.
The flash memory controller may be arranged to select a set of evaluated parameters that once applied results in manageable level of bit error rates and has a smallest contribution to a wear level of the at least one group of flash memory cells.
The flash memory controller may be arranged to select a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
The flash memory controller may be arranged to select a set of evaluated parameters that once applied results in (a) a higher bit error rate and a lower programming duration at low wear levels and (b) a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
The flash memory controller may be arranged to select a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and at a lower increment rate of bit error rates per wear level in relation to a reference set of parameters that is expected to be applied regardless of a wear level of the at least one group of flash memory cells.
Each set of evaluated parameters may include erase and programming parameters of values that once applied will not result in a corruption of data programmed to the flash memory device.
Each set of evaluated parameters may include programming parameters that may be applied during programming operations of information that is not related to the finding of the erase parameters and the programming parameters.
The flash memory controller may repetitively program same information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other.
The flash memory controller may be arranged to repetitively program the same information to multiple groups of flash memory cells is a probability of a programming operation failure exceeds a threshold.
The flash memory controller may be arranged to repetitively program the same pilot information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other.
The finding, by the flash memory controller, may include applying a steepest decent algorithm.
The finding by the flash memory controller may include changing a value of a single evaluated parameter between one set of evaluated parameter to another.
The flash memory controller may be arranged to evaluate a first set of evaluated parameters; calculate a first score; change a single parameter of the first set to provide a second set of evaluated parameters that may include a single changed parameter; wherein the change is performed along a first change direction; evaluate the second set of evaluated parameters; calculate a second score; determine, in response to the first and second scores, whether the first set of evaluated parameters achieves better results than the second set of evaluated parameters; if it is determined that the first set of evaluated parameters achieves better results than the second set of evaluated parameters then change the single changed parameter along a second change direction that differs from the first change direction to provide a third set of evaluated parameters; if it is determined that the first set of evaluated parameters achieves worse results than the second set of evaluated parameters then change the single changed parameter along the first direction to provide the third set of evaluated parameters.
The flash memory controller may be arranged to evaluate a single set of evaluated parameters based upon an outcome of multiple read attempts.
The flash memory controller may be arranged to trigger the finding when a bit error rate, which is calculated during a read operation, is associated with information programmed to at least one group of flash memory cells reaches a predetermined value.
The flash memory controller may be arranged to find a set of erase parameters that once applied minimizes a wear level increment resulting from a programming operation.
The subject matter regarded as the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention, however, both as to organization and method of operation, together with objects, features, and advantages thereof, may best be understood by reference to the following detailed description when read with the accompanying drawings in which:
It will be appreciated that for simplicity and clarity of illustration, elements shown in the figures have not necessarily been drawn to scale. For example, the dimensions of some of the elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements.
In the following detailed description, numerous specific details may be set forth in order to provide a thorough understanding of the invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, and components have not been described in detail so as not to obscure the present invention.
The subject matter regarded as the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention, however, both as to organization and method of operation, together with objects, features, and advantages thereof, may best be understood by reference to the following detailed description when read with the accompanying drawings.
It will be appreciated that for simplicity and clarity of illustration, elements shown in the figures have not necessarily been drawn to scale. For example, the dimensions of some of the elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements.
Because the illustrated embodiments of the present invention may for the most part, be implemented using electronic components and circuits known to those skilled in the art, details will not be explained in any greater extent than that considered necessary as illustrated above, for the understanding and appreciation of the underlying concepts of the present invention and in order not to obfuscate or distract from the teachings of the present invention.
The terms process and method are used interchangeably in this specification.
According to various embodiments of the invention there are provided methods, non-transitory computer readable media and systems that dynamically optimize erase and programming parameters of a flash memory device memory device as a function of the endurance of the flash memory device memory device. This may obtain a high reliability level and a high endurance of the flash memory device memory device.
According to an embodiment of the invention the optimization method of the erase parameters is aimed to reduce and even minimize the wear effect involved with an erase operation as function of the endurance.
According to an embodiment of the invention programming parameters are dynamically optimized in order to minimize induced bit error rate (BER) and the standard deviation (STD) of the voltage threshold distribution of the programming operation.
According to an embodiment of the invention the optimization method of the programming parameters can be done by using a steepest-decent search for minimum, without adding any system overhead.
According to an embodiment of the invention the optimization method rules are designed so that using non-optimal erase or programming parameters are harmless, and do not induce extra programming operations. The evaluation of the programming parameters can be done without performance penalties since the scores that are used for the evaluation process are obtained through normal (non-dedicated) read operations of the previously programmed content.
The optimization method may be executed as an ongoing process, or may be limited to a fixed number of parameters modifications for a single or multiple program erase (P/E) cycle ranges.
It is noted that erase optimization method and/or program optimization method may have a complimentary contribution to the endurance maximization over the flash memory device memory device life span. According to various embodiments of the invention each of the optimization method (erase optimization method and program optimization method) may be done separately, or jointly. For example, the erase optimization method may be tuned as to minimize the wear effect at the expense of high program induced BER, while the program optimization method may be further tuned to minimize the BER of the programming.
Either one of the optimization methods may be done under constraints of average program time and average erase time. These constraints may change as a function of endurance.
The optimization method can evaluate the outcome of using certain programming parameters and/or erase parameter by computing one or more scores. One of these scores can be an error-score that may be measured on a whole block or on a subset of pages within a block.
The program optimization method does not have to use dedicated data, but may be implemented during programming of operational data into the flash memory device memory device. However, applications that run the program optimization method may check the validity of the programming result, and decide to re-program the same data in case the programming with modified parameters failed.
The programming optimization method may benefit also from applying a virtual retention effect function, as illustrated in U.S. patent application Ser. No. 12/769,208 filing date Apr. 28, 2010, which is incorporated herein by reference.
The span of the programming parameters, subject for optimization, may be set such to exclude values which might have a destructive impact on the programming result. Thus, only a sub-rage of values may be tested. For example, different combination of programming parameter values or extreme valid values for different programming parameters may lead to a failure in the programming, meaning that the data programmed cannot be reliably read after programming. Such scenarios may be avoided by limiting the span of parameters change in advance, according to the embodiments of this invention.
Any one of the erase optimization method and/or program optimization method can be performed in a periodical manner (for example—for every P/E cycle range—corresponding to a weal level sub-range), or triggered by events such as changes in a score that is computed during read operations done by a host (without adding extra read operations), or changes in a score computed from dedicated read operations which examine the programming results.
According to various embodiments of the invention the set of programming and/or erase parameters may be either pre-determined per P/E cycle range, or may be the last set used following the previous optimization method.
The programming optimization method and/or the erase optimization method may include testing a different set of parameters for optimization. For example, it may be required that the programming verify levels remain unchanged during the optimization method, while parameters like the programming start voltage and the program voltage step will be part of the optimization method
According to various embodiments of the invention the score computation for evaluating the program and/erase optimization method can be (a) dependent on the programming/erase parameters, (b) dependent on the read errors or/and lobes' distribution, and (c) change as a function of the P/E cycle value (or effective wear levels changes).
In order to reach high endurance it is required to adapt the programming scheme on the fly, per device. To do this it is suggested to modify the programming and erase parameters as function of the endurance (represented by P/E cycle count) or another wear level. A method for calculating wear level (also referred to as effective P/E cycle count) is illustrated in U.S. patent application Ser. No. 13/342,946 which is incorporated herein by reference.
An equivalent P/E cycles count which is estimated according to the last programming result. The estimate can be based on the average programming time and/or on the number of errors following the programming operation. In some cases there may be a strong correspondence between the number of errors during the programming and the wear level.
Controlling the erase parameters as function of the wear level (effective P/E cycles) can assist in obtaining high endurance and reliability. The erase parameters can be controlled such that the traps accumulation during erase will be as small as possible. If done carefully, may reduce dramatically the wear out of the cell per erase operation.
For example, performing an erase operation with as small erase voltage and minimal number of pulses, and to a highest possible erase level, may minimize the inherent destructive effect of the erase operation.
Such strategy may result however in higher errors during the programming operation, or result in a wider distribution, i.e. high standard deviation per lobe. An example for erase adaptation to wear state is disclosed in U.S. patent application Ser. No. 13/238,983 which is incorporated herein by reference.
Naturally, this may seem to contradict the goal of programming optimization method which aims at minimizing the BER and STD corresponding to a programming operation.
In this application a method for controlling the erase and programming parameters is disclosed. According to the embodiments of this invention, a tradeoff between the two contradicting targets is found such that the overall endurance is maximized without sacrificing the reliability. Erase parameters may be set in view of the programming parameters.
The optimization method can result by having a programming operation that follows an erase operation, to end up with a relatively high BER, which is still reliably corrected following a read operation and error correction. The programming parameters are optimized as to minimize the BER and STD of the program induced distribution. This goal is achieved with different programming and/or erase parameters as the endurance changes.
In many flash memory device memory controllers, management software makes nearly equal wear to large sets of physical blocks. That is known as wear leveling. When using wear leveling, the overhead of the erase and programming optimization methods may be minimized since it may be sufficient to optimize parameters on sample blocks for every P/E cycle range. That is, it is not needed to apply such optimization method for every block and every programming operation. It is possible, for example, to perform this optimization method at every cycle P/E range, like demonstrated in the tables below.
Table 1 provides an example of a set of erase parameters used per cycle range, as initial conditions for further optimization method.
Table 2 illustrates an example of a set of programming parameters used per cycle range, as initial conditions for further optimization method.
The x-axis of graph 30 represents that number of P/E cycle counts that typical blocks in the flash memory device have undergone. The y-axis represents the average raw bit error rate (RBER) when reading from the flash memory device, and applying (if needed) a read operation (to be applied by a read circuit that may be equipped with a Digital Signal Processor) which allocates approximately optimal read thresholds.
Graph 30 includes a reliable error correction decoding threshold 35, curves 31 and 32 illustrating the behaviour (P/E count values versus RBER) of prior art schemes that are fixed (use the same erase and programming parameters) regardless the P/E cycle count—before and after retention.
Graph 30 also includes curves 33 and 34 illustrating the behaviour (P/E count values versus RBER) of applying a method according to an embodiment of the invention—before and after retention.
As may be noticed, the average RBER grows with the P/E cycle count, and further more with retention. Using prior art programming/erase methods, the RBER at low cycle counts is fairly small. However, it increases at a rather high slope. This is due to the accelerated wear associated with every program and erase operation which are fixed. According to embodiments of this invention, the RBER of low cycle counts is rather high. This is due to using programming parameters that allow fast programming, at the expense of program level distribution (i.e. wide lobe distribution). This is also due to programming on a smaller voltage window on initial cycles. The smaller programming voltage window contributes to the minimization of the device wear due to program and erase operations. The higher RBER at low P/E cycles can be tolerated by the ECC, and therefore has no negative impact on read performance.
Another benefit of the embodiments of this invention is that the RBER slope is smaller than the RBER slope of prior art. The smaller slope allows achieving more P/E cycles with reliable decoding after retention. For example, like shown in the figure below, the adaptive programming and erase after retention at 20,000 cycles reaches the ECC limit of 2 E-2, in the example. Comparing this to prior art, with the same ECC limit, it may be noticed that only a little above 1,000 cycles may be reached post retention. Thus the adaptive programming and erase may greatly improve reliability and increase the lifespan of a flash memory device.
At rather high cycle counts (e.g. above 1000 cycles), using the default programming parameters (fixed parameters), results in high standard deviation (STD) around every program level. This causes a high RBER during read with (near) optimal thresholds. If programming is done with adaptive programming, like disclosed in the specification, the STD around every program level is reduced as well as the RBER during read with (near) optimal thresholds.
The dashed distribution (lobes 11-14) refers to a distribution associated with prior art programming schemes, while the solid line distribution (lobes 41-44) is associated with the programming result when using optimized programming parameters. The upper most graphs are the programming distribution without retention. The lower graphs are the post-retention voltage threshold distribution. As may be noticed from the post-retention distribution, the retention effect is more severe in the prior art implementation evident from a larger shift of the programming levels, and a larger increase in STD per programming level. The post-retention distribution associated with adaptive program and adaptive erase exhibits a smaller shift in the average program levels, and a smaller increase in the STD of the voltage threshold distribution around every level. This results in lower RBER during read operations, which in total allows reliable readouts at higher cycle counts.
The optimization method can include adapting error parameters and/or programming parameters. The programming and/or erase parameter optimization may be done without adding pilot/training data to the programming operation.
The optimization method 100 can start by receiving (102) a new programming command from an application executed by a host computer.
The flash memory controller may check if adaptation of the erase and/or programming parameters is required (104).
The checking may include using the score of read bit error rates (BER) from the read of previous programming. The RBER can be combined with programming time and erase time to provide a common score. Furthermore, the RBER is measured on different pages. The controller may take an average value, and max BER value, over a subset of programmed pages. The score may be for example a weighted sum of the mean BER and the maximal BER, i.e.
score=β·ē+(1−β)·max(e)+M·1(Tprog>THp)+L·1(Terase>THe)
Where 0≦β≦1 is a weighting factor of the BER vector e, and ē is the average BER and max(e) is the highest measured readout BER. The parameters M, L are penalty factors, which may be added to the score in case of constraint violation of the programming time Tprog or the erase time Terase, respectively.
The score may be even more generic and include more elements having to do with the programming and/or erase parameters, e.g.
score=β·ē+(1−β)·max(e)+ƒp(Tprog−THp)+ƒe(Terase−THe)+ge(EraseVstart,EraseVoltageStep,NppErase)+gp(PgVstart,PgVoltageStep,NppPg)
Where ƒe and ƒp are functions of the program time result w.r.t. a threshold parameter, and may for example be an increasing function as the deviation from the threshold increases, e.g. ƒe=a·(Terase−THe)·1(Terase−THe) where a is a positive scalar, and 1(x) is a step function which is 1 if x>0, otherwise 0.
The functions ge and gp are cost functions which depend on erase and programming parameters, respectively. For example, the higher the erase start voltage, the lower the cost may be. The functions perform weighting of the different parameters according to their actual contribution to the programming and erase results.
In case it is required to perform an optimization method of programming and/or erase, then the controller modifies the erase/programming parameters, and sets them (106) to the flash memory device, according to embodiments of the invention, and erases/programs (108) the next block according to management and application requirements, and then re-computes the score (110) following the program operation.
It is emphasized that the optimization method may be an on-going operation done simultaneously with the normal program/read commands issued by the host, as to introduce no overhead on read and write performance.
According to another embodiment of this invention, the parameter optimization method is done separately and possibly even differently per flash memory device. This is explained as follows: a controller might have an array of flash memory device dies, which are jointly programmed, in order to improve system performance. The read and write operations may be done faster if interleaving of codewords is done over multiple devices. In addition, the programming and/or erase optimization method may end up with different parameters optimizing the score per device. Therefore, the controller may collect RBER information and tProg and tErase from every device separately. For computing RBER of a multi-die interleaved codeword, the controller shall count the number of errors detected in the portion of the codeword that belongs to a specific single die. This may enable computing a score(d), i.e. a score for every device, which corresponds to the last parameters used for erase/program.
The erase parameters optimization method 140 may include stages 142, 146, 148, 150, 152, 154, 156, 158, 160 and 162.
Method 140 starts by receiving an initial set of erase parameters and beginning the erase parameter adaptation (142).
This is followed by determining (146) a subset of the erase parameter set to be optimized and keeping the programming parameters fixed.
This is followed by changing, before next erase operation, one or more erase parameters of the sub-set that should be optimized (148) to reduce erase wear effect.
Non-limiting example changes may include at least one of: reducing the erase start voltage, reducing the erase voltage step up value, increasing the erase verify level, or increasing the threshold of number of unverified cells (which is the number of non-erased cells for which erase can be terminated).
This is followed by using the erase parameters to erase a block of flash memory cells and program to the erase block user information (150).
When reading content of blocks programmed after erasing with the modified parameters, the statistics of average BER and maximal BER and erase time and program time are collected and can be used to compute a score (152), possibly like the score computed in (1).
If the score is smaller than a lower threshold (154), i.e. score<TH1, this means that the erase parameters may be further changed to reduce wear effect, and on the next erase operation (148), these parameters may be changed again, according to same guidelines provided earlier.
If the score exceeds the first threshold it is compared to a second threshold (156).
If the score is within an desired range, i.e. TH2>score≧TH1, then the erase parameters optimization method may terminated (162) after the erase parameters may be saved (160) and used within the cycle range, or up to a new optimization method trigger.
If the score exceeds a high threshold, i.e. score≧TH2, which means that programming ended up with too many induced errors, then the valid erase parameters should be set (158) to those that were used in previous erase operation.
Optionally, a programming optimization method may be attempted, according to embodiments for programming optimization method disclosed in this invention.
Once programming optimization method successfully terminated the programming score is compared again with TH2, and if smaller the last erase parameters are used.
Otherwise, or if no program optimization method follows, then the erase parameters from previous erase parameters modification is set and further erase operations are carried out with the last erase parameters saved.
Method 120 may be targeted to minimize the wear effect involved in the erase operation, while allowing a sufficient quality of program results.
This optimization method serves to maximize the flash memory device supported endurance.
At the beginning (may be initialed periodically—122) of the erase optimization method an initial set of programming and erase parameters is received (124). These initial sets of programming and erase parameters can be different for every P/E cycle range, like demonstrated in Tables 1, and 2.
The process proceeds to a determination (126) whether there is a need to perform the optimization method. If the answer is positive then the programming parameters are modified (during a programming parameter optimization 128). This may be followed by checking if a programming adaptation value is indicative that the programming parameters optimization should end (130) and jumping to stage 126 of not. Else—the process continues to stage 132 of erase parameter optimization. This may be followed by end of process stage 134.
According to embodiments of the invention, the process may start periodically, for every cycle range, or may be triggered by a score which is computed following host read commands. That is, the score occasionally computed after every read operation. If the score is larger than a certain threshold, then a process of optimization method may take place.
According to some embodiments of the invention, if an optimization method is needed, the first process includes optimizing the programming parameters, such that the score becomes smaller. A smaller score expresses a more robust programming result, which may have less read errors after programming by obtaining a smaller STD of the voltage threshold distribution per level, or by having a greater spacing between the program levels.
The program adaptation may be done over a sequence of programming operations done on possibly multiple different blocks with approximately the same effective cycle count or wear level.
Once the programming optimization method is complete, the score which results from the optimization method is compared with a threshold. If the score is smaller than the threshold, than the current optimization method may terminate. Otherwise, an erase optimization method takes place.
During the erase optimization method, the erase parameters are changed that that the programming may end up with a smaller STD and or a smaller read RBER. This may include for example using more erase pulses, and/or increasing the erase start voltage, and other strategies according to embodiments of this invention.
The erase adaptation may be done over a sequence of erase operations done on possibly multiple different blocks with approximately the same effective cycle count or wear level.
The optimization method (170) begins by determining (172) to perform an optimization method of programming parameters on a flash memory device denoted d and receiving (174) a set of programming parameters of this device that include N programming parameters. The parameters of device d can be referred to as param(i,d) or simply param(i), index i ranges between 1 and N. The set of programming parameters may be either pre-determined per cycle range, or may be the last set used following the previous optimization method.
Stage 172 may also include determining a subset of parameters which will be subject for optimization. For example, it may be required that the programming verify levels remain unchanged during the optimization method, while parameters like the programming start voltage and the program voltage step will be part of the optimization method.
This may be followed by determining (176) a search area and a search grid for each parameter—this may include determining, for each parameter to be evaluated a step and range values. This may be either pre-configured, or dynamically set according to the sensitivity of the programming result to the change in a parameter. Usually, the range will be set such that the parameters will not be set to values which might have a destructive impact on the programming result. That is, there are different combinations of parameter values or extreme valid values for different parameters which may lead to a failure in the programming, meaning that the data programmed cannot be reliably read after programming. Such scenarios are avoided by limiting the span of parameters change in advance, according to the embodiments of this invention.
The parameters may be ordered (according to values of index i) for the optimization method. For every parameter (178) the initial search direction is set (180), and a single change step (182) is applied (the parameter may be changed with a subset of dependent parameters, or may be a single parameter). The modified parameters set is set to the flash memory device.
The next step is to program one or more blocks of the flash memory device (186) with similar cycle counts using this set of parameters. The flash memory device is programmed with operational data, without using any pilots. Then the programming results are evaluated by collecting (188) the RBER following programming of the block and using the tPROG and other statistics, a score is computed to evaluate the benefit of the step.
If (190) the score is bigger than a threshold, i.e. score(d)>TH1, and the number of tests did not exceed a maximum the new score (194 “score (d)<prevscore(d)”) is compared with a previous score. Else the process may terminate (192). If the new score is smaller, then the new parameters are saved (196—“prevscore(d)=score(d), param(i,d)=newparam(i,d)”), and another step of modification (198) of the parameters of the subset in the same direction will be done. In any case the test index is incremented (198) before performing a new optimization iteration. Otherwise, the parameter is not saved, and the optimization method can take place in the opposite direction, or a different parameters subset can be chosen, i.e. change the parameters election index i.
Once a sufficiently small score is obtained, or the number of tests exceeded a threshold, the optimization method is terminated (192) and a new optimization method is scheduled (193).
According to an embodiment of this invention, if the optimization method terminated with a score which is not smaller than TH1, then an addition optimization method with different initial conditions may be carried out, in the same lines.
According to an embodiment of this invention, the thresholds configuration may be changed according to the current endurance. That is, the higher the P/E cycles, or effective cycling, the higher we may want to set the score threshold.
According to an embodiment of this invention the score thresholds and functions may change as a function of the P/E cycles. For example, the weights attributed with different metrics used to calculate the score may change with the P/E cycles. E.g. with higher P/E cycles, lower weights to tProg may be used as it may be less important if it takes longer to program at high P/E cycles.
It is assumed that N program/erase cycle count ranges are defined or received (202). Stage may include defining or receiving N sub-ranges of wear levels.
These can be either following a counter of P/E cycles, or defined as effective cycle count ranges, where the effective number of cycles is estimated from programming results such as RBER following programming with known parameters, and/or by measuring its corresponding tPROG.
Every time the effective cycle count of a set of blocks reaches a new cycle range (204—current PIE cycle count reaches a first P/E cycle count value of a new P/E range), an optimization method of erase and programming parameters is carried out (206) and at least some evaluated parameters and updated (208).
During each program erase cycle the P/E counter is updated (210) and compared to boarders of a new P/E cycle count range (212) before jumping to stage 204.
This optimization method may be done separately per flash memory device.
According to an embodiment of the invention it may use as initial conditions for optimization, a set of parameters which may be either pre-determined per cycle range, or may be the last set obtained by the previous optimization method.
The optimization method (220) may begin with receiving (221) a set of programming related parameters. The set of parameters may be either pre-determined per cycle range, or may be the last set used following the previous execution of the optimization method.
The receiving may be followed by determining (222) the subset of parameters which will be subject for optimization. For example, it may be required that the programming verify levels remain unchanged during the optimization method while parameters like the programming start voltage and the program voltage step will be part of the optimization method.
For the set of parameters for optimization, a step and range values are determined (222) per parameter. This may be either pre-configured, or dynamically set according to the sensitivity of the programming result to the change in a parameter.
Usually, the range will be set such that the parameters will not be set to values which might have a destructive impact on the programming result. That is, there are different combinations of parameter values or extreme valid values for different parameters which may lead to a failure in the programming, meaning that the data programmed cannot be reliably read after programming. Such scenarios are avoided by limiting the span of parameters change in advance, according to the embodiments of this invention.
The parameters are ordered for the execution of the optimization method. For every parameter (226) the initial search direction is set {D(i)=1} (224), and a single change step is applied (228)—the parameter may be changed with a subset of dependent parameters, or may be a single parameter. The modified parameters set is to be applied to the flash memory device. The value of D(i) indicates the search direction for the i-th parameter.
The next step is to program (230) block(s) with similar cycle counts using this set of parameters. The flash memory device may be programmed with operational data, without using any pilots. Then the programming results are evaluated (230) by collecting the RBER following programming of the block and using the tPROG and other statistics, a score is computed to evaluate the benefit of the step. The test iteration counter is increased (box 232—“testldx=testldx+1”).
The method 220 then continues to text (234) if a minimum has been reached and whether the number of tests (test iterations) reached a maximum allowable value (234).
If the sequence of tests yielding a score per test reached a minimum, in the lines following a steepest-decent search for the minimum, or the number of evaluations exceeded a maximum then the optimization method may terminate (248).
Otherwise, if the number of tests did not exceed a maximum, and a minimum was not yet obtained, the new score is compared with a previous score (236). If the new score is smaller, then the new parameters are saved (238), and another step of modification to the parameter subset param(i) in the same direction will be done (226).
Otherwise, the parameter is not saved, and the param(i) optimization method, if Di=1, can take place in the opposite direction, else a different parameter subset can be chosen, i.e. change the parameters election index i: i=i+1 (248).
When D(i) is 1 in stage 240 it means that the search direction should be changed and then D(i) is set to 0. The next time we reach stage 240, D(i) is 0, and therefore the method will continue from stage 240 to stage 246 which increases the parameter pointer, i.e. optimization continues with another parameter.
Once a sufficiently small score is obtained, or the number of tests exceeded a threshold, the optimization method is terminated.
According to an embodiment of this invention, if the optimization method terminates with a score which is not smaller than TH1, then an addition optimization method with different initial conditions may be carried out, according to the same guidelines.
According to an embodiment of this invention, the thresholds may be changed according to the current endurance. That is, the higher the P/E cycles, or effective cycling, the higher we may want to change the thresholds.
Optimization method 250 differs from method 220 of
If the testing (234, 252) indicates that a minimum has not been reached and although the number of tests (test iterations) reached maximum allowable value, the method 250 continues to perform (256) a random/deterministic perturbation on the initial parameters. This perturbation may comply with the optimization search rules on the parameters set (for example—limiting the values in order to prevent a destructive programming process), as otherwise, these parameters might lead to destructive programming results.
Instead of this random perturbation (256), or in addition, an erase optimization method (not shown) may be carried out to provide a smaller effective score for the current parameters. Once, this is done, there is a better chance to succeed in the next programming optimization method.
Stage 256 may be followed by stage 252.
The programming and/or erase adaptation may be done without adding pilot/training data to the programming operation. As indicated in previous figures—on a new programming command from the application, the controller may check if adaptation of the erase/programming parameters is required. The check may include using a score.
The calculation of a score may start by collecting statistics on programming results (282) related to a group of flash memory cells. It is assumed, for brevity of explanation that the group of flash memory cells forms a block.
Stage 282 may be followed by counting (284) the number of corrected errors from a subset of pages of the block and computing (286) a threshold voltage distribution (for example—generating a histogram) of a single row or multiple rows of the block. The computing (286) can be followed by computing (288), based upon the histogram the standard deviation and average-voltage of each storage level of the flash memory ells. The outputs of the computing (286) and the counting (284) can be combined (290) with average programming operation duration and average erase operation duration to provide a score.
For example, a read operation generated bit error rate (BER) that can be gathered from a read operation that follows erase and programming operations can be obtained. The raw bit error rate (RBER) can be combined with programming time and erase time to provide a common score. The RBER is the bit error rate of the read from the Flash device before applying the error correction coding. It is expected that post ECC all errors will be corrected, and this is actually how also it is possible to compute the number or errors.
Furthermore, the RBER may be measured on different pages. The controller may take an average value, and max BER value, over a subset of programmed pages. The score may be for example a weighted sum of the mean BER and the maximal BER, i.e.
score=β·ē+(1−β)·max(e)+M·1(Tprog>THp)+M·1(Terase>THe)
Where 0≦β≦1 is a weighting factor of the BER vector e, and ē is the average BER and max(e) is the highest measured readout BER of the vector e.
The parameter M is a penalty factor which may be added to the score in case of constraint violation of the programming time Tprog or the erase time Terase.
In addition, the score may benefit from including, if available, a voltage threshold histogram processed data. For example, a histogram may be created during activation of a DSP for finding the optimal read thresholds. From such a histogram of the voltage threshold distribution it may be possible to estimate/approximate the STD of each program level. The average STD over all levels, and the maximal STD, can be combined in the resulting score in the following way
Where 0≦α≦1, 0≦β≦1, 0≦γ≦1 are weighting factor of the BER vector e, and ē is the average BER and max(e) is the highest measured readout BER, and the STD measured from the voltage threshold histogram. The parameter M is a penalty factor which may be added to the score in case of constraint violation of the programming time Tprog or the erase time Terase.
Method 500 is aimed for flash memory device parameter optimization.
Method 500 may start by stage 502 of determining to initiate a flash memory device parameter optimization method.
This determination can be responsive to an event, can be executed in a periodic manner, in a random or pseudo-random manner.
An event can be an occurrence of a certain change in a wear level of certain flash memory cells, reaching a certain wear level, entering a sub-range of wear levels, an occurrence of a programming failure, an occurrence of an erase failure, a detection of certain levels of bit error rates, and the like.
A non-limiting example of sub-ranges of wear levels are illustrated in tables 1 and 2—each sub-range of wear levels is associated with a range of P/E cycles counts—each row of these tables represents a different sub-range of wear levels. It is noted that the wear level can be evaluated in other manners and in this case the P/E cycle count does not define the sub-ranges.
Stage 502 can determine to initiate the optimization method once per one or more sub-ranges of wear levels, multiple times per one or more sub-ranges of wear level, skip the optimization per one or more sub-ranges of wear levels and the like.
According to an embodiment of the invention the optimization is executed one the wear level first enters (or proximate to the entering to) a new sub-range of wear levels. It is assumed that at this point the state of the state of the flash memory cells can be regarded as the best within the sub-range.
Stage 502 may be followed by stage 504 of receiving or generating an estimate of a wear level of certain flash memory cells of the flash memory device. These certain flash memory cells can be a page, multiple pages, an erase block, multiple erase block and the like.
Stage 504 may be followed by stage 510 of finding erase parameters and programming parameters to be applied on one or more groups of flash memory cells of the flash memory device in response to estimate of the wear level.
The finding process may include finding erase parameters and/or programming parameters. The finding process (510) may include stage 520 of evaluating a plurality of sets of evaluated parameters that differ from each other by a value of at least one parameter.
Each one of the plurality of sets of evaluated parameters may include comprise at least one out of a set of erase parameters and a set of programming parameters. Accordingly, stage 510 may include evaluating sets of programming parameters, programming sets of erase parameters or evaluating a combination thereof.
Stage 520 may include stages 530 and 560.
Stage 520 may also include stages such as stages 521-527 that can be shared between stages 530 and 560.
Any of the stages (521-527) is applicable to the finding of erase parameters, programming parameters or both erase and programming parameters.
Stage 521 may include finding a selected set of evaluated parameters that has a smallest effect on the wear of the group of flash memory cells out a plurality of sets of evaluated parameters that once applied introduce a tolerable deviation in a voltage threshold distribution.
Stage 522 may include finding a set of evaluated parameters that once applied results in manageable level of bit error rates that exceeds a minimal allowable level of bit error rates.
Stage 523 may include selecting a set of evaluated parameters that once applied results in manageable level of bit error rates and has a smallest contribution to a wear level of the at least one group of flash memory cells.
Stage 524 may include selecting a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and at a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
Stage 525 may include selecting a set of evaluated parameters that once applied results in a higher bit error rate at low wear levels and at a lower increment rate of bit error rates per wear level in relation to a reference set of parameters that is expected to be applied regardless of a wear level of the at least one group of flash memory cells.
Stage 526 may include calculating a score to a set of evaluated parameters. The score can be responsive to a set of erase parameters, to a set of programming parameters or to a set that includes programming and erase parameters.
Stage 527 may include selecting a set of evaluated parameters that once applied results in (a) a higher bit error rate and a lower programming duration at low wear levels and (b) a lower increment rate of bit error rates per wear level in relation to at least one other set of evaluated parameters.
Stage 530 includes evaluating sets of erase parameters.
Stage 530 may include stages 531-539.
Stage 531 may include evaluating sets of erase parameters by erasing different groups of flash memory cells of the flash memory device that have substantially a same weal level. Thus—instead repetitively erasing the same group of flash memory cells and thus performing erase operations that are dedicated to the evaluation of the erase parameter, the erasing can be executed on different groups of flash memory cells—especially those who were already scheduled to be erased—thus not erase overhead is required.
Nevertheless—even stage 531 can include performing one or more erase operations that are dedicated for the evaluation of the erase parameters.
Stage 553 may include finding a set of erase parameters that minimize a formation of traps.
Stage 534 may include finding a set of erase parameters that minimize a wear of the group of flash memory cells resulting from erase processes.
Stage 535 may include finding a set of erase parameters that includes at least one of a start erase voltage, a number of maximal allowable erase pulses, an erase pulse width, an erase voltage step and erase level. Especially, and as illustrated in table 1, stage 535 may include finding a set of erase parameters that includes at least one of a minimal start erase voltage, minimal number of maximal allowable erase pulses, minimal erase pulse width, minimal erase voltage step and maximal erase level.
Stage 536 may include evaluating sets of evaluated parameters wherein each set of erase parameters is sub-optimal in a context of voltage threshold distribution. Sub-optimal voltage threshold distribution can include lobes of width than exceeds a minimal obtainable width, lobes that are more close to each other than what could have been obtained with an optical voltage threshold distribution and the like.
Stage 537 may includes finding a set of erase parameters in response to a duration attribute of erase operations to be obtained when applying the set of erase parameters. The duration attribute can be an average duration of erase operation, a weighted sum of durations erase operations and the like.
Stage 537 may include ignoring sets of erase parameters that once applied result in an average duration of erase operations that exceeds an erase duration threshold.
Stage 537 may include changing a value of the erase duration threshold in response to values of the wear level. Higher wear levels may increase the erase duration threshold.
Stage 538 may include calculating a score per each set of erase parameter. The score is calculated during the evaluation and assist in selecting a selected set of erase parameters.
The score of a set of erase parameter may be responsive to read errors resulting from a read operation that follows a programming operation that in turn follows the erase operation.
The score of each set of erase parameters can be responsive to at least one value of at least one parameter of the set of erase parameters.
The score of each set of evaluated parameters can be response to the wear level.
Sub-ranges of higher order are associated with higher wear levels. According to an embodiment of the invention stage 539 may include increasing at least one parameter out of a start erase voltage, a number of maximal allowable erase pulses, an erase voltage step and erase pulse width with an increase of an order to the sub-range. A non-limiting example is provided in table 1.
Stage 540 may include finding a set of erase parameters that once applies minimizes a wear level increment resulting from an erase operation.
Stage 543 may include performing erase operations, during the evaluation of erase parameters, while using “normal” (not test-dedicated) erase parameters. Accordingly, stage 573 may include erase parameters that are applied during erase operations of information that is not related to the finding of the erase parameters and the erase parameters.
Stage 560) includes evaluating sets of programming parameters. Stage 560 may include stages 561-569.
Stage 560 may include stage 561 of programming different groups of flash memory cells of the flash memory device that have substantially a same weal level.
Thus—instead repetitively programming the same group of flash memory cells and thus performing programming operations that are dedicated to the evaluation of the programming parameter, the programming can be executed on different groups of flash memory cells—especially those who were already scheduled to be programmed—thus not programming overhead is required.
Nevertheless—even stage 561 can include performing one or more programming operations that are dedicated for the evaluation of the programming parameters.
In order not to corrupt programmed information (or at least reduce the chances of such a corruption) the values of the evaluated parameters (erase and/or programming parameters) can be limited—such as to belong to only within sub-ranges of possible ranges of values of the evaluated parameters. This is illustrated by stage 532 and 562 of limiting the values of the erase or programming parameters (respectively) to sub-ranges.
Stage 565 may include finding a set of programming parameters that includes at least one of a start programming voltage, a number of maximal allowable programming pulses, a programming voltage step, a programming pulse width and a program bias voltage. Especially, and as illustrated in table 2, stage 565 may include finding a set of programming parameters that includes at least one of a maximal start programming voltage, minimal number of maximal allowable programming pulses, minimal programming voltage step, minimal programming pulse width and minimal program bias voltage.
Stage 567 may includes finding a set of programming parameters in response to a duration attribute of programming operations to be obtained when applying the set of programming parameters. The duration attribute can be an average duration of programming operation, a weighted sum of durations programming operations and the like.
Stage 567 may include ignoring sets of programming parameters that once applied result in an average duration of programming operations that exceeds an programming duration threshold.
Stage 567 may include changing a value of the programming duration threshold in response to values of the wear level. Higher wear levels may increase the programming duration threshold.
Stage 568 may include calculating a score per each set of programming parameter. The score can be calculated during the evaluation and can assist in selecting a selected set of programming parameters.
The score of a set of programming parameter may be responsive to read errors resulting from a read operation that follows a programming operation that was executed by applying programming parameters of the set of the evaluated parameter.
The score of a set of programming parameters can be responsive to at least one value of at least one parameter of the set of evaluated parameters.
The score of a set of programming parameters can be responsive to the wear level.
According to an embodiment of the invention stage 569 may include increasing at least one parameter out of a number of maximal allowable programming pulses, programming voltage step, programming window, program pulse width and program bias voltage with an increase of an order to the sub-range. A non-limiting example is provided in table 2.
Stage 570 may include finding a set of programming parameters that once applied minimized a wear level increment resulting from a programming operation.
Stage 571 may include finding a set of programming parameters of values that once applied will not result in a corruption of data programmed to the flash memory device.
Stage 572 may include calculating a score of a set of programming parameters in response to statistical attribute of the voltage threshold distribution resulting from applying the set of programming parameters.
Stage 573 may include performing programming operations, during the evaluation of programming parameters, while using “normal” (not test-dedicated) programming parameters. Accordingly, stage 573 may include programming parameters that are applied during programming operations of information that is not related to the finding of the erase parameters and the programming parameters.
Stage 574 may include repetitively programming same information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other. These different groups may differ from each other by their wear level.
Stage 575 may include repetitively programming the same information to multiple groups of flash memory cells if a probability of a programming operation failure exceeds a threshold.
Stage 576 may include repetitively programming same pilot information to multiple groups of flash memory cells while applying different sets of programming parameters that differ from each other.
Stage 577 may include finding a set of programming parameters while applying a steepest decent algorithm. Non-limiting examples of such algorithm are illustrated in
Any of the methods mentioned above can be executed by a memory controller.
The sub-system 402 includes a flash memory controller 420 and a flash memory device 460. The flash memory device 460 is illustrated as having two flash memory dies 450(1) and 450(2), that include blocks 440(1)-440(4) respectively. Each block may include multiple pages such as flash memory pages 430(i,j). The flash memory controller 420 is coupled to host controller 410 and to the flash memory device 460. The number of pages, dies, and blocks may differ from those illustrated in
The flash memory controller 420 can include a read circuit 421, a write circuit 422, an erase circuit 423, and parameter finder 423 for finding erase parameters and programming parameters. The read circuit 421 can read from the flash memory device 460. The write circuit 422 can write (perform programming operations) to the flash memory device 460. The erase circuit 423 may be arranged to perform erase operations of erase blocks of the flash memory device. The parameter tinder 424 can find programming parameters, erase parameters or a combination of both. The parameter finder 424 may be arranged to execute all stages of any methods mentioned in the specification except read, write and erase operations that are executed by circuits 421, 422 and 423. The parameter finder 424 control the operations of circuits 421-423 but this is not necessarily so.
The invention may also be implemented in a computer program for running on a computer system, at least including code portions for performing steps of a method according to the invention when run on a programmable apparatus, such as a computer system or enabling a programmable apparatus to perform functions of a device or system according to the invention.
A computer program is a list of instructions such as a particular application program and/or an operating system. The computer program may for instance include one or more of: a subroutine, a function, a procedure, an object method, an object implementation, an executable application, an applet, a servlet, a source code, an object code, a shared library/dynamic load library and/or other sequence of instructions designed for execution on a computer system.
The computer program may be stored internally on a non-transitory computer readable medium. All or some of the computer program may be provided on computer readable media permanently, removably or remotely coupled to an information processing system. The computer readable media may include, for example and without limitation, any number of the following: magnetic storage media including disk and tape storage media; optical storage media such as compact disk media (e.g., CD-ROM, CD-R, etc.) and digital video disk storage media nonvolatile memory storage media including semiconductor-based memory units such as FLASH memory, EEPROM, EPROM. ROM; ferromagnetic digital memories; MRAM; volatile storage media including registers, buffers or caches, main memory, RAM, etc.
A computer process typically includes an executing (running) program or portion of a program, current program values and state information, and the resources used by the operating system to manage the execution of the process. An operating system (OS) is the software that manages the sharing of the resources of a computer and provides programmers with an interface used to access those resources. An operating system processes system data and user input, and responds by allocating and managing tasks and internal system resources as a service to users and programs of the system.
The computer system may for instance include at least one processing unit, associated memory and a number of input/output (I/O) devices. When executing the computer program, the computer system processes information according to the computer program and produces resultant output information via I/O devices.
In the foregoing specification, the invention has been described with reference to specific examples of embodiments of the invention. It will, however, be evident that various modifications and changes may be made therein without departing from the broader spirit and scope of the invention as set forth in the appended claims.
Moreover, the terms “front,” “back,” “top,” “bottom,” “over,” “under” and the like in the description and in the claims, if any, are used for descriptive purposes and not necessarily for describing permanent relative positions. It is understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the invention described herein are, for example, capable of operation in other orientations than those illustrated or otherwise described herein.
The connections as discussed herein may be any type of connection suitable to transfer signals from or to the respective nodes, units or devices, for example via intermediate devices. Accordingly, unless implied or stated otherwise, the connections may for example be direct connections or indirect connections. The connections may be illustrated or described in reference to being a single connection, a plurality of connections, unidirectional connections, or bidirectional connections. However, different embodiments may vary the implementation of the connections. For example, separate unidirectional connections may be used rather than bidirectional connections and vice versa. Also, plurality of connections may be replaced with a single connection that transfers multiple signals serially or in a time multiplexed manner. Likewise, single connections carrying multiple signals may be separated out into various different connections carrying subsets of these signals. Therefore, many options exist for transferring signals.
Although specific conductivity types or polarity of potentials have been described in the examples, it will appreciated that conductivity types and polarities of potentials may be reversed.
Each signal described herein may be designed as positive or negative logic. In the case of a negative logic signal, the signal is active low where the logically true state corresponds to a logic level zero. In the case of a positive logic signal, the signal is active high where the logically true state corresponds to a logic level one. Note that any of the signals described herein can be designed as either negative or positive logic signals. Therefore, in alternate embodiments, those signals described as positive logic signals may be implemented as negative logic signals, and those signals described as negative logic signals may be implemented as positive logic signals.
Furthermore, the terms “assert” or “set” and “negate” (or “deassert” or “clear”) are used herein when referring to the rendering of a signal, status bit, or similar apparatus into its logically true or logically false state, respectively. If the logically true state is a logic level one, the logically false state is a logic level zero. And if the logically true state is a logic level zero, the logically false state is a logic level one.
Those skilled in the art will recognize that the boundaries between logic blocks are merely illustrative and that alternative embodiments may merge logic blocks or circuit elements or impose an alternate decomposition of functionality upon various logic blocks or circuit elements. Thus, it is to be understood that the architectures depicted herein are merely exemplary, and that in fact many other architectures can be implemented which achieve the same functionality.
Any arrangement of components to achieve the same functionality is effectively “associated” such that the desired functionality is achieved. Hence, any two components herein combined to achieve a particular functionality can be seen as “associated with” each other such that the desired functionality is achieved, irrespective of architectures or intermedial components. Likewise, any two components so associated can also be viewed as being “operably connected,” or “operably coupled,” to each other to achieve the desired functionality.
Furthermore, those skilled in the art will recognize that boundaries between the above described operations merely illustrative. The multiple operations may be combined into a single operation, a single operation may be distributed in additional operations and operations may be executed at least partially overlapping in time. Moreover, alternative embodiments may include multiple instances of a particular operation, and the order of operations may be altered in various other embodiments.
Also for example, in one embodiment, the illustrated examples may be implemented as circuitry located on a single integrated circuit or within a same device. Alternatively, the examples may be implemented as any number of separate integrated circuits or separate devices interconnected with each other in a suitable manner.
Also for example, the examples, or portions thereof, may implemented as soft or code representations of physical circuitry or of logical representations convertible into physical circuitry, such as in a hardware description language of any appropriate type.
Also, the invention is not limited to physical devices or units implemented in non-programmable hardware but can also be applied in programmable devices or units able to perform the desired device functions by operating in accordance with suitable program code, such as mainframes, minicomputers, servers, workstations, personal computers, notepads, personal digital assistants, electronic games, automotive and other embedded systems, cell phones and various other wireless devices, commonly denoted in this application as ‘computer systems’.
However, other modifications, variations and alternatives are also possible. The specifications and drawings are, accordingly, to be regarded in an illustrative rather than in a restrictive sense.
In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word ‘comprising’ does not exclude the presence of other elements or steps then those listed in a claim. Furthermore, the terms “a” or “an,” as used herein, are defined as one or more than one. Also, the use of introductory phrases such as “at least one” and “one or more” in the claims should not be construed to imply that the introduction of another claim element by the indefinite articles “a” or “an” limits any particular claim containing such introduced claim element to inventions containing only one such element, even when the same claim includes the introductory phrases “one or more” or “at least one” and indefinite articles such as “a” or “an.” The same holds true for the use of definite articles. Unless stated otherwise, terms such as “first” and “second” are used to arbitrarily distinguish between the elements such terms describe. Thus, these terms are not necessarily intended to indicate temporal or other prioritization of such elements. The mere fact that certain measures are recited in mutually different claims does not indicate that a combination of these measures cannot be used to advantage.
While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes, and equivalents will now occur to those of ordinary skill in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.
This application is a continuation in part of U.S. patent application Ser. No. 12/769,208 having a filing date Apr. 28, 2010, which is incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4430701 | Christian et al. | Feb 1984 | A |
| 4463375 | Macovski | Jul 1984 | A |
| 4584686 | Fritze | Apr 1986 | A |
| 4589084 | Fling et al. | May 1986 | A |
| 4777589 | Boettner et al. | Oct 1988 | A |
| 4866716 | Weng | Sep 1989 | A |
| 5003597 | Merkle | Mar 1991 | A |
| 5077737 | Leger et al. | Dec 1991 | A |
| 5297153 | Baggen et al. | Mar 1994 | A |
| 5305276 | Uenoyama | Apr 1994 | A |
| 5592641 | Doyle et al. | Jan 1997 | A |
| 5623620 | Fandrich et al. | Apr 1997 | A |
| 5640529 | Hasbun | Jun 1997 | A |
| 5657332 | Auclair et al. | Aug 1997 | A |
| 5663901 | Harari et al. | Sep 1997 | A |
| 5724538 | Morris et al. | Mar 1998 | A |
| 5729490 | Calligaro et al. | Mar 1998 | A |
| 5740395 | Wells et al. | Apr 1998 | A |
| 5745418 | Ting-Chung Hu et al. | Apr 1998 | A |
| 5778430 | Ish et al. | Jul 1998 | A |
| 5793774 | Usui et al. | Aug 1998 | A |
| 5920578 | Zook et al. | Jul 1999 | A |
| 5926409 | Engh et al. | Jul 1999 | A |
| 5933368 | Ting-Chung Hu et al. | Aug 1999 | A |
| 5956268 | Lee | Sep 1999 | A |
| 5956473 | Ting-Chung Hu et al. | Sep 1999 | A |
| 5968198 | Balachandran | Oct 1999 | A |
| 5982659 | Irrinki et al. | Nov 1999 | A |
| 6011741 | Harari et al. | Jan 2000 | A |
| 6016275 | Han | Jan 2000 | A |
| 6038634 | Ji et al. | Mar 2000 | A |
| 6081878 | Estakhri et al. | Jun 2000 | A |
| 6094465 | Stein et al. | Jul 2000 | A |
| 6119245 | Hiratsuka | Sep 2000 | A |
| 6182261 | Haller et al. | Jan 2001 | B1 |
| 6192497 | Yang et al. | Feb 2001 | B1 |
| 6195287 | Hirano | Feb 2001 | B1 |
| 6199188 | Shen et al. | Mar 2001 | B1 |
| 6209114 | Wolf et al. | Mar 2001 | B1 |
| 6259627 | Wong | Jul 2001 | B1 |
| 6272052 | Miyauchi | Aug 2001 | B1 |
| 6278633 | Wong et al. | Aug 2001 | B1 |
| 6279133 | Vafai et al. | Aug 2001 | B1 |
| 6301151 | Engh et al. | Oct 2001 | B1 |
| 6370061 | Yachareni et al. | Apr 2002 | B1 |
| 6374383 | Weng | Apr 2002 | B1 |
| 6504891 | Chevallier | Jan 2003 | B1 |
| 6532169 | Mann et al. | Mar 2003 | B1 |
| 6532556 | Wong et al. | Mar 2003 | B1 |
| 6553533 | Demura et al. | Apr 2003 | B2 |
| 6560747 | Weng | May 2003 | B1 |
| 6637002 | Weng et al. | Oct 2003 | B1 |
| 6639865 | Kwon | Oct 2003 | B2 |
| 6674665 | Mann et al. | Jan 2004 | B1 |
| 6675281 | Oh et al. | Jan 2004 | B1 |
| 6704902 | Shinbashi et al. | Mar 2004 | B1 |
| 6751766 | Guterman et al. | Jun 2004 | B2 |
| 6772274 | Estakhri | Aug 2004 | B1 |
| 6781910 | Smith | Aug 2004 | B2 |
| 6792569 | Cox et al. | Sep 2004 | B2 |
| 6873543 | Smith et al. | Mar 2005 | B2 |
| 6891768 | Smith et al. | May 2005 | B2 |
| 6914809 | Hilton et al. | Jul 2005 | B2 |
| 6915477 | Gollamudi et al. | Jul 2005 | B2 |
| 6952365 | Gonzalez et al. | Oct 2005 | B2 |
| 6961890 | Smith | Nov 2005 | B2 |
| 6968421 | Conley | Nov 2005 | B2 |
| 6990012 | Smith et al. | Jan 2006 | B2 |
| 6996004 | Fastow et al. | Feb 2006 | B1 |
| 6999854 | Roth | Feb 2006 | B2 |
| 7010739 | Feng et al. | Mar 2006 | B1 |
| 7012835 | Gonzalez et al. | Mar 2006 | B2 |
| 7038950 | Hamilton et al. | May 2006 | B1 |
| 7068539 | Guterman et al. | Jun 2006 | B2 |
| 7079436 | Perner et al. | Jul 2006 | B2 |
| 7149950 | Spencer et al. | Dec 2006 | B2 |
| 7177977 | Chen et al. | Feb 2007 | B2 |
| 7188228 | Chang et al. | Mar 2007 | B1 |
| 7191379 | Adelmann et al. | Mar 2007 | B2 |
| 7196946 | Chen et al. | Mar 2007 | B2 |
| 7203874 | Roohparvar | Apr 2007 | B2 |
| 7212426 | Park et al | May 2007 | B2 |
| 7290203 | Emma et al. | Oct 2007 | B2 |
| 7292365 | Knox | Nov 2007 | B2 |
| 7301928 | Nakabayashi et al. | Nov 2007 | B2 |
| 7315916 | Bennett et al. | Jan 2008 | B2 |
| 7388781 | Litsyn et al. | Jun 2008 | B2 |
| 7395404 | Gorobets et al. | Jul 2008 | B2 |
| 7441067 | Gorobets et al. | Oct 2008 | B2 |
| 7443729 | Li et al. | Oct 2008 | B2 |
| 7450425 | Aritome | Nov 2008 | B2 |
| 7454670 | Kim et al. | Nov 2008 | B2 |
| 7466575 | Shalvi et al. | Dec 2008 | B2 |
| 7533328 | Alrod et al. | May 2009 | B2 |
| 7558109 | Brandman et al. | Jul 2009 | B2 |
| 7593263 | Sokolov et al. | Sep 2009 | B2 |
| 7610433 | Randell et al. | Oct 2009 | B2 |
| 7613043 | Cornwell et al. | Nov 2009 | B2 |
| 7619922 | Li et al. | Nov 2009 | B2 |
| 7697326 | Sommer et al. | Apr 2010 | B2 |
| 7706182 | Shalvi et al. | Apr 2010 | B2 |
| 7716538 | Gonzalez et al. | May 2010 | B2 |
| 7804718 | Kim | Sep 2010 | B2 |
| 7805663 | Brandman et al. | Sep 2010 | B2 |
| 7805664 | Yang et al. | Sep 2010 | B1 |
| 7844877 | Litsyn et al. | Nov 2010 | B2 |
| 7911848 | Eun et al. | Mar 2011 | B2 |
| 7961797 | Yang et al. | Jun 2011 | B1 |
| 7975192 | Sommer et al. | Jul 2011 | B2 |
| 8020073 | Emma et al. | Sep 2011 | B2 |
| 8108590 | Chow et al. | Jan 2012 | B2 |
| 8122328 | Liu et al. | Feb 2012 | B2 |
| 8159881 | Yang | Apr 2012 | B2 |
| 8169825 | Shalvi et al. | May 2012 | B1 |
| 8190961 | Yang et al. | May 2012 | B1 |
| 8250324 | Haas et al. | Aug 2012 | B2 |
| 8264890 | Mokhlesi et al. | Sep 2012 | B2 |
| 8300823 | Bojinov et al. | Oct 2012 | B2 |
| 8305812 | Levy et al. | Nov 2012 | B2 |
| 8327246 | Weingarten et al. | Dec 2012 | B2 |
| 8407560 | Ordentlich et al. | Mar 2013 | B2 |
| 8417893 | Khmelnitsky et al. | Apr 2013 | B2 |
| 20010034815 | Dugan et al. | Oct 2001 | A1 |
| 20020063774 | Hillis et al. | May 2002 | A1 |
| 20020085419 | Kwon et al. | Jul 2002 | A1 |
| 20020154769 | Petersen et al. | Oct 2002 | A1 |
| 20020156988 | Toyama et al. | Oct 2002 | A1 |
| 20020174156 | Birru et al. | Nov 2002 | A1 |
| 20030014582 | Nakanishi | Jan 2003 | A1 |
| 20030065876 | Lasser | Apr 2003 | A1 |
| 20030101404 | Zhao et al. | May 2003 | A1 |
| 20030105620 | Bowen | Jun 2003 | A1 |
| 20030177300 | Lee et al. | Sep 2003 | A1 |
| 20030192007 | Miller et al. | Oct 2003 | A1 |
| 20040015771 | Lasser et al. | Jan 2004 | A1 |
| 20040030971 | Tanaka et al. | Feb 2004 | A1 |
| 20040059768 | Denk et al. | Mar 2004 | A1 |
| 20040080985 | Chang et al. | Apr 2004 | A1 |
| 20040153722 | Lee | Aug 2004 | A1 |
| 20040153817 | Norman et al. | Aug 2004 | A1 |
| 20040181735 | Xin | Sep 2004 | A1 |
| 20040203591 | Lee | Oct 2004 | A1 |
| 20040210706 | In et al. | Oct 2004 | A1 |
| 20050013165 | Ban | Jan 2005 | A1 |
| 20050018482 | Cemea et al. | Jan 2005 | A1 |
| 20050083735 | Chen et al. | Apr 2005 | A1 |
| 20050117401 | Chen et al. | Jun 2005 | A1 |
| 20050120265 | Pline et al. | Jun 2005 | A1 |
| 20050128811 | Kato et al. | Jun 2005 | A1 |
| 20050138533 | Le-Bars et al. | Jun 2005 | A1 |
| 20050144213 | Simkins et al. | Jun 2005 | A1 |
| 20050144368 | Chung et al. | Jun 2005 | A1 |
| 20050169057 | Shibata et al. | Aug 2005 | A1 |
| 20050172179 | Brandenberger et al. | Aug 2005 | A1 |
| 20050213393 | Lasser | Sep 2005 | A1 |
| 20050243626 | Ronen | Nov 2005 | A1 |
| 20060059406 | Micheloni et al. | Mar 2006 | A1 |
| 20060059409 | Lee | Mar 2006 | A1 |
| 20060064537 | Oshima | Mar 2006 | A1 |
| 20060101193 | Murin | May 2006 | A1 |
| 20060106972 | Gorobets et al. | May 2006 | A1 |
| 20060195651 | Estakhri et al. | Aug 2006 | A1 |
| 20060203587 | Li et al. | Sep 2006 | A1 |
| 20060221692 | Chen | Oct 2006 | A1 |
| 20060248434 | Radke et al. | Nov 2006 | A1 |
| 20060268608 | Noguchi et al. | Nov 2006 | A1 |
| 20060282411 | Fagin et al. | Dec 2006 | A1 |
| 20060284244 | Forbes et al. | Dec 2006 | A1 |
| 20060294312 | Walmsley | Dec 2006 | A1 |
| 20070025157 | Wan et al. | Feb 2007 | A1 |
| 20070063180 | Asano et al. | Mar 2007 | A1 |
| 20070081388 | Joo | Apr 2007 | A1 |
| 20070098069 | Gordon | May 2007 | A1 |
| 20070103992 | Sakui et al. | May 2007 | A1 |
| 20070104004 | So et al. | May 2007 | A1 |
| 20070109858 | Conley et al. | May 2007 | A1 |
| 20070124652 | Litsyn et al. | May 2007 | A1 |
| 20070140006 | Chen et al. | Jun 2007 | A1 |
| 20070143561 | Gorobets | Jun 2007 | A1 |
| 20070150694 | Chang et al. | Jun 2007 | A1 |
| 20070168625 | Cornwell et al. | Jul 2007 | A1 |
| 20070171714 | Wu et al. | Jul 2007 | A1 |
| 20070171730 | Ramamoorthy et al. | Jul 2007 | A1 |
| 20070180346 | Murin | Aug 2007 | A1 |
| 20070223277 | Tanaka et al. | Sep 2007 | A1 |
| 20070226434 | Guterman et al. | Sep 2007 | A1 |
| 20070226582 | Tang et al. | Sep 2007 | A1 |
| 20070226592 | Radke | Sep 2007 | A1 |
| 20070228449 | Takano et al. | Oct 2007 | A1 |
| 20070253249 | Kang et al. | Nov 2007 | A1 |
| 20070253250 | Shibata et al. | Nov 2007 | A1 |
| 20070263439 | Cornwell et al. | Nov 2007 | A1 |
| 20070266291 | Toda et al. | Nov 2007 | A1 |
| 20070271494 | Gorobets | Nov 2007 | A1 |
| 20070297226 | Mokhlesi | Dec 2007 | A1 |
| 20080010581 | Alrod et al. | Jan 2008 | A1 |
| 20080028014 | Hilt et al. | Jan 2008 | A1 |
| 20080049497 | Mo | Feb 2008 | A1 |
| 20080055989 | Lee et al. | Mar 2008 | A1 |
| 20080082897 | Brandman et al. | Apr 2008 | A1 |
| 20080092026 | Brandman et al. | Apr 2008 | A1 |
| 20080104309 | Cheon et al. | May 2008 | A1 |
| 20080112238 | Kim et al. | May 2008 | A1 |
| 20080116509 | Harari et al. | May 2008 | A1 |
| 20080126686 | Sokolov et al. | May 2008 | A1 |
| 20080127104 | Li et al. | May 2008 | A1 |
| 20080128790 | Jung | Jun 2008 | A1 |
| 20080130341 | Shalvi et al. | Jun 2008 | A1 |
| 20080137413 | Kong et al. | Jun 2008 | A1 |
| 20080137414 | Park et al. | Jun 2008 | A1 |
| 20080141043 | Flynn et al. | Jun 2008 | A1 |
| 20080148115 | Sokolov et al. | Jun 2008 | A1 |
| 20080158958 | Sokolov et al. | Jul 2008 | A1 |
| 20080159059 | Moyer | Jul 2008 | A1 |
| 20080162079 | Astigarraga et al. | Jul 2008 | A1 |
| 20080168216 | Lee | Jul 2008 | A1 |
| 20080168320 | Cassuto et al. | Jul 2008 | A1 |
| 20080181001 | Shalvi | Jul 2008 | A1 |
| 20080198650 | Shalvi et al. | Aug 2008 | A1 |
| 20080198652 | Shalvi et al. | Aug 2008 | A1 |
| 20080201620 | Gollub | Aug 2008 | A1 |
| 20080209114 | Chow et al. | Aug 2008 | A1 |
| 20080219050 | Shalvi et al. | Sep 2008 | A1 |
| 20080225599 | Chae | Sep 2008 | A1 |
| 20080250195 | Chow et al. | Oct 2008 | A1 |
| 20080263262 | Sokolov et al. | Oct 2008 | A1 |
| 20080282106 | Shalvi et al. | Nov 2008 | A1 |
| 20080285351 | Shlick et al. | Nov 2008 | A1 |
| 20080294814 | Gorobets | Nov 2008 | A1 |
| 20080301532 | Uchikawa et al. | Dec 2008 | A1 |
| 20090024905 | Shalvi et al. | Jan 2009 | A1 |
| 20090027961 | Park et al. | Jan 2009 | A1 |
| 20090043951 | Shalvi et al. | Feb 2009 | A1 |
| 20090046507 | Aritome | Feb 2009 | A1 |
| 20090072303 | Prall et al. | Mar 2009 | A9 |
| 20090091979 | Shalvi | Apr 2009 | A1 |
| 20090103358 | Sommer et al. | Apr 2009 | A1 |
| 20090103371 | Goda et al. | Apr 2009 | A1 |
| 20090106485 | Anholt | Apr 2009 | A1 |
| 20090113275 | Chen et al. | Apr 2009 | A1 |
| 20090125671 | Flynn et al. | May 2009 | A1 |
| 20090132755 | Radke | May 2009 | A1 |
| 20090144598 | Yoon et al. | Jun 2009 | A1 |
| 20090144600 | Perlmutter et al. | Jun 2009 | A1 |
| 20090150599 | Bennett | Jun 2009 | A1 |
| 20090150748 | Egner et al. | Jun 2009 | A1 |
| 20090157964 | Kasorla et al. | Jun 2009 | A1 |
| 20090158126 | Perlmutter et al. | Jun 2009 | A1 |
| 20090168524 | Golov et al. | Jul 2009 | A1 |
| 20090187803 | Anholt et al. | Jul 2009 | A1 |
| 20090199074 | Sommer | Aug 2009 | A1 |
| 20090213653 | Perlmutter et al. | Aug 2009 | A1 |
| 20090213654 | Perlmutter et al. | Aug 2009 | A1 |
| 20090228761 | Perlmutter et al. | Sep 2009 | A1 |
| 20090240872 | Perlmutter et al. | Sep 2009 | A1 |
| 20090282185 | Van Cauwenbergh | Nov 2009 | A1 |
| 20090282186 | Mokhlesi et al. | Nov 2009 | A1 |
| 20090287930 | Nagaraja | Nov 2009 | A1 |
| 20090300269 | Radke et al. | Dec 2009 | A1 |
| 20090323942 | Sharon et al. | Dec 2009 | A1 |
| 20100005270 | Jiang | Jan 2010 | A1 |
| 20100025811 | Bronner et al. | Feb 2010 | A1 |
| 20100030944 | Hinz | Feb 2010 | A1 |
| 20100058146 | Weingarten et al. | Mar 2010 | A1 |
| 20100064096 | Weingarten et al. | Mar 2010 | A1 |
| 20100088557 | Weingarten et al. | Apr 2010 | A1 |
| 20100091535 | Sommer et al. | Apr 2010 | A1 |
| 20100095186 | Weingarten | Apr 2010 | A1 |
| 20100110787 | Shalvi et al. | May 2010 | A1 |
| 20100115376 | Shalvi et al. | May 2010 | A1 |
| 20100122113 | Weingarten et al. | May 2010 | A1 |
| 20100124088 | Shalvi et al. | May 2010 | A1 |
| 20100131580 | Kanter et al. | May 2010 | A1 |
| 20100131806 | Weingarten et al. | May 2010 | A1 |
| 20100131809 | Katz | May 2010 | A1 |
| 20100131826 | Shalvi et al. | May 2010 | A1 |
| 20100131827 | Sokolov et al. | May 2010 | A1 |
| 20100131831 | Weingarten et al. | May 2010 | A1 |
| 20100146191 | Katz | Jun 2010 | A1 |
| 20100146192 | Weingarten et al. | Jun 2010 | A1 |
| 20100149881 | Lee et al. | Jun 2010 | A1 |
| 20100172179 | Gorobets et al. | Jul 2010 | A1 |
| 20100174845 | Gorobets et al. | Jul 2010 | A1 |
| 20100174853 | Lee et al. | Jul 2010 | A1 |
| 20100180073 | Weingarten et al. | Jul 2010 | A1 |
| 20100199149 | Weingarten et al. | Aug 2010 | A1 |
| 20100202199 | Guterman et al. | Aug 2010 | A1 |
| 20100211724 | Weingarten | Aug 2010 | A1 |
| 20100211833 | Weingarten | Aug 2010 | A1 |
| 20100211856 | Weingarten | Aug 2010 | A1 |
| 20100241793 | Sugimoto et al. | Sep 2010 | A1 |
| 20100246265 | Moschiano et al. | Sep 2010 | A1 |
| 20100251066 | Radke | Sep 2010 | A1 |
| 20100253555 | Weingarten et al. | Oct 2010 | A1 |
| 20100257309 | Barsky et al. | Oct 2010 | A1 |
| 20100269008 | Leggette et al. | Oct 2010 | A1 |
| 20100293321 | Weingarten | Nov 2010 | A1 |
| 20100299350 | Konig et al. | Nov 2010 | A1 |
| 20100318724 | Yeh | Dec 2010 | A1 |
| 20110051521 | Levy et al. | Mar 2011 | A1 |
| 20110055461 | Steiner et al. | Mar 2011 | A1 |
| 20110093650 | Kwon et al. | Apr 2011 | A1 |
| 20110096612 | Steiner et al. | Apr 2011 | A1 |
| 20110099460 | Dusija et al. | Apr 2011 | A1 |
| 20110119562 | Steiner et al. | May 2011 | A1 |
| 20110153919 | Sabbag | Jun 2011 | A1 |
| 20110161775 | Weingarten | Jun 2011 | A1 |
| 20110194353 | Hwang et al. | Aug 2011 | A1 |
| 20110209028 | Post et al. | Aug 2011 | A1 |
| 20110214029 | Steiner et al. | Sep 2011 | A1 |
| 20110214039 | Steiner et al. | Sep 2011 | A1 |
| 20110246792 | Weingarten | Oct 2011 | A1 |
| 20110246852 | Sabbag | Oct 2011 | A1 |
| 20110252187 | Segal et al. | Oct 2011 | A1 |
| 20110252188 | Weingarten | Oct 2011 | A1 |
| 20110271043 | Segal et al. | Nov 2011 | A1 |
| 20110302428 | Weingarten | Dec 2011 | A1 |
| 20120001778 | Steiner et al. | Jan 2012 | A1 |
| 20120005554 | Steiner et al. | Jan 2012 | A1 |
| 20120005558 | Steiner et al. | Jan 2012 | A1 |
| 20120005560 | Steiner et al. | Jan 2012 | A1 |
| 20120008401 | Katz et al. | Jan 2012 | A1 |
| 20120008414 | Katz et al. | Jan 2012 | A1 |
| 20120017136 | Ordentlich et al. | Jan 2012 | A1 |
| 20120051144 | Weingarten et al. | Mar 2012 | A1 |
| 20120063227 | Weingarten et al. | Mar 2012 | A1 |
| 20120066441 | Weingarten | Mar 2012 | A1 |
| 20120110250 | Sabbag et al. | May 2012 | A1 |
| 20120124273 | Goss et al. | May 2012 | A1 |
| 20120246391 | Meir et al. | Sep 2012 | A1 |
| 20140156965 | Yang et al. | Jun 2014 | A1 |
| Entry |
|---|
| Search Report of PCT Patent Application WO 2009/118720 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/095902 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/078006 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/074979 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/074978 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/072105 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/072104 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/072103 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/072102 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/072101 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/072100 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/053963 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/053962 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/053961 A3, Mar. 4, 2010. |
| Search Report of PCT Patent Application WO 2009/037697 A3, Mar. 4, 2010. |
| Yani Chen, Kcshab K. Parhi, “Small Area Parallel Chien Search Architectures for Long BCH Codes”, Ieee Transactions on Very Large Scale Integration(VLSI) Systems, vol. 12, No. 5, May 2004. |
| Yuejian Wu, “Low Power Decoding of BCH Codes”, Nortel Networks, Ottawa, Ont., Canada, in Circuits and systems, 2004. ISCAS '04. Proceeding of the 2004 International Symposium on Circuits and Systems, published May 23-26, 2004, vol. 2, pp. II-369-72 vol. 2. |
| Michael Purser, “Introduction to Error Correcting Codes”, Artech House Inc., 1995. |
| Ron M. Roth, “Introduction to Coding Theory”, Cambridge University Press, 2006. |
| Akash Kumar, Sergei Sawitzki, “High-Throughput and Low Power Architectures for Reed Solomon Decoder”, (a.kumar at tue.nl, Eindhoven University of Technology and sergei.sawitzki at philips.com), Oct. 2005. |
| Todd K.Moon, “Error Correction Coding Mathematical Methods and Algorithms”, A John Wiley & Sons, Inc., 2005. |
| Richard E. Blahut, “Algebraic Codes for Data Transmission”, Cambridge University Press, 2003. |
| David Esseni, Bruno Ricco, “Trading-Off Programming Speed and Current Absorption in Flash Memories with the Ramped-Gate Programming Technique”, Ieee Transactions on Electron Devices, vol. 47, No. 4, Apr. 2000. |
| Giovanni Campardo, Rino Micheloni, David Novosel, “VLSI-Design of Non-Volatile Memories”, Springer Berlin Heidelberg New York, 2005. |
| John G. Proakis, “Digital Communications”, 3rd ed., New York: McGraw-Hill, 1995. |
| J.M. Portal, H. Aziza, D. Nee, “EEPROM Memory: Threshold Voltage Built in Self Diagnosis”, ITC International Test Conference, Paper 2.1, Feb. 2005. |
| J.M. Portal, H. Aziza, D. Nee, “EEPROM Diagnosis Based on Threshold Voltage Embedded Measurement”, Journal of Electronic Testing: Theory and Applications 21, 33-42, 2005. |
| G. Tao, A. Scarpa, J. Dijkstra, W. Stidl, F. Kuper, “Data retention prediction for modern floating gate non-volatile memories”, Microelectronics Reliability 40 (2000), 1561-1566. |
| T. Hirncno, N. Matsukawa, H. Hazama, K. Sakui, M. Oshikiri, K. Masuda, K. Kanda, Y. Itoh, J. Miyamoto, “A New Technique for Measuring Threshold Voltage Distribution in Flash EEPROM Devices”, Proc. IEEE 1995 Int. Conference on Microelectronics Test Structures, vol. 8, Mar. 1995. |
| Boaz Eitan, Guy Cohen, Assaf Shappir, Eli Lusky, Amichai Givant, Meir Janai, Ilan Bloom, Yan Polansky, Oleg Dadashev, Avi Lavan, Ran Sahar, Eduardo Maayan, “4-bit per Cell NROM Reliability”, Appears on the website of Saifun.com , 2005. |
| Paulo Cappelletti, Clara Golla, Piero Olivo, Enrico Zanoni, “Flash Memories”, Kluwer Academic Publishers, 1999. |
| JEDEC Standard, “Stress-Test-Driven Qualification of Integrated Circuits”, JEDEC Solid State Technology Association. JEDEC Standard No. 47F pp. 1-26. Dec. 2007. |
| Dempster, et al., “Maximum Likelihood from Incomplete Data via the EM Algorithm”, Journal of the Royal Statistical Society. Series B (Methodological), vol. 39, No. 1 (1997), pp. 1-38. |
| Mielke, et al., “Flash EEPROM Threshold Instabilities due to Charge Trapping During Program/Erase Cycling”, IEEE Transactions on Device and Materials Reliability, vol. 4, No. 3, Sep. 2004, pp. 335-344. |
| Daneshbeh, “Bit Serial Systolic Architectures for Multiplicative Inversion and Division over GF (2)”, A thesis presented to the University of Waterloo, Ontario, Canada, 2005, pp. 1-118. |
| Chen, Formulas for the solutions of Quadratic Equations over GF (2), IEEE Trans. Inform. Theory, vol. IT- 28, No. 5, Sep. 1982, pp. 792-794. |
| Berlekamp et al., “On the Solution of Algebraic Equations over Finite Fields”, Inform. Cont. 10, Oct. 1967, pp. 553-564. |
| Number | Date | Country | |
|---|---|---|---|
| 61236911 | Aug 2009 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 12769208 | Apr 2010 | US |
| Child | 13595782 | US |
