Patent Application
20140047177
This invention relates to data storage entities arranged in mirrored pairs.
Data storage systems may be arranged to provide data storage with varying degrees of data redundancy. RAID (redundant array of independent disks) is a term applied to pluralities of data storage physical entities, such as disk drives or hard disk drives, arranged to add a redundancy to the data so that the data can be reconstructed even if there is a limited loss of data. A RAID system typically comprises more than one data storage physical entity and the limited loss of data can be up to a catastrophic loss of data on one or more of the data storage physical entities. Numbers are applied to various versions of RAID, some of which copy or “mirror” the data, and others provide “parity” RAID, where data is stored on more than one data storage drive, the data is summed and the parity (which makes the sum of the data, and the parity, equal to all the same bit) is stored separately from the data.
An example of a mirrored RAID system is a RAID-1 which comprises an even number of data storage drives, and where data stored on one data storage drive is copied (mirrored) on another data storage drive, forming a mirrored pair. Thus, if one data storage drive fails, the data is still available on the other data storage drive. Typically, all of the data storage drives to be used in a RAID are selected randomly to be very similar to each other.
Mirrored data storage systems, mirrored arrangements of data storage physical entities (drives), methods and computer program products are provided for assigning mirrored pairing of drives.
One embodiment of a mirrored data storage system has a RAID control system; and a plurality of data storage physical entities (drives) arranged in mirrored pairs in accordance with reliability weightings assigned to each of the plurality of drives. Each mirrored pair comprises one of the drives with at least a median and greater reliability weighting, and one of the drives with at least a median and lesser reliability weighting.
In a further embodiment, the plurality of drives comprises an even number, and any drive having a median reliability weighting is arranged at the side of a mirrored pair required to equalize the number of drives at each side of the arrangement.
In another embodiment, the mirrored pairs are arranged as a RAID 01 data storage system having two groups of the drives, a first group of the drives of median and greater reliability weightings of the mirrored pairs, and the second group of the drives of median and lesser reliability weightings, to form the mirrored pairs.
In still another embodiment, the paired mirrored sets are arranged as a RAID 10 data storage system having a plurality of mirrored pairs of drives.
In yet another embodiment, the reliability weighting comprises an assessment of probability of operation of the physical entity without permanent loss of data.
In a further embodiment, the probability of operation is related to a probability of length of time without permanent loss of data.
In a still further embodiment, the probability of operation is related to static information provided with respect to the data storage physical entity.
In another embodiment, the probability of operation is related to dynamic information derived from previous operation of the data storage physical entity.
In still another embodiment, a computer-implemented method of assigning data storage physical entity mirrored pairings, comprises the steps of:
assigning a reliability weighting to each of a plurality of drives;
sorting the assigned reliability weightings into descending order;
assigning the physical entities having weightings in the upper half of the sorted order to a first set of the drives with greater reliability weighting, and assigning the physical entities having weightings in the lower half of the sorted order to a second set of drives with lesser reliability weighting; and
selecting, for each mirrored pair, one data storage physical entity from the first set and one data storage physical entity from the second set.
For a fuller understanding of the present invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings.
This invention is described in preferred embodiments in the following description with reference to the Figures, in which like numbers represent the same or similar elements. While this invention is described in terms of the best mode for achieving this invention's objectives, it will be appreciated by those skilled in the art that variations may be accomplished in view of these teachings without deviating from the spirit or scope of the invention.
Referring to
The storage of data on multiple data storage entities 15 is conducted by a control 20. The control comprises at least one computer processor 22 which operates in accordance with computer-usable program code stored in a computer-usable storage medium 23 having non-transient computer-usable program code embodied therein. The computer processor arranges the data storage entities, and stores data to be written to, or that has been read from, the data storage entities in a data memory 24. The control directs the data to locations of the data storage entities so that the data is mirrored, such that two copies of the data are stored by the data storage entities. Alternatively, the data storage functions may be conducted by a host system or server system arranged, inter alia, as control 20 and the data storage entities may comprise individual controls.
The data storage entities 15, for example, comprises a plurality of hard disk drives 30, 31, 32, 33 and 34, or may comprise an electronic memory 37 such as an SSD (solid state drive) as a substitute for one or more hard disk drives. One or more of the data storage entities may be held as a spare.
Referring to
RAID data storage is provided, for example, as the data storage attached to a DS8000™ Enterprise Storage Server of International Business Machines Corp. (IBM®). The DS8000™ is a high performance, high capacity storage server providing data storage, which may include RAID, that is designed to support continuous operations and implement virtualization of data storage, and is presented herein only by way of embodiment examples and is not intended to be limiting. Thus, the data storage system 10 of
In
Referring to
Thus, in
Mirror_Pair1: <Drive1 61,71, Drive3 63,73>
Mirror_Pair2: <Drive2 62,72, Drive4 64,74>
Reliability of RAID array=Min (Lifetime of Mirror_Pair1, Lifetime of Mirror_Pair2)
Lifetime of Mirror_Pair1=Max (Lifetime of Drive1 61,71, Lifetime of Drive3 63,73)
Lifetime of Mirror_Pair2=Max (Lifetime of Drive2 62,72, Lifetime of Drive4 64,74)
The assignment of the mirrored pairing of drives comprises arranging the drives in mirrored pairs in accordance with reliability weightings assigned to each of the plurality of drives. Each mirrored pair comprises one of the drives with at least a median and greater reliability weighting, and one of the drives with at least a median and lesser reliability weighting.
Thus, each mirrored pair has at least one drive with at least a median and greater reliability weighting, assuring that each mirrored pair will have at least a median and greater reliability and therefore a likely longer lifetime than if the drives were assigned on another basis.
For example, during the setup phase of the RAID system, such as by the RAID control 20, a system or application associated with the control, collects information about the health statistics of all the drives 61, 62, 63, 64, 71, 72, 73, 74 participating in the formation of the RAID array 50, or 52.
In the example of magnetic disk drives, examples of health information may include QoS (Quality of Service) information provided by the disk drive manufacturer for the type and model of disk drive. This information may be called “static” information. Other information may be derived from S.M.A.R.T. (Self Monitoring, Analysis and Reporting Technology) measured by the disk drive itself as defined by the disk drive manufacturer. This information changes as the drive is being used and may be called “dynamic” information.
Referring to
Head Flying Height—A downward trend in flying height will often presage a head crash.
Number of Remapped Sectors—If the drive is remapping many sectors due to internally detected errors, this can mean that the drive is approaching failure.
ECC (Error Correction Code) and Error Counts—The number of errors encountered by the drive, even if corrected internally, often signals problems developing with the drive. The trend is in some cases more important than the actual count.
Spin-Up Time—Changes in spin-up time can reflect problems with the spindle motor.
Temperature—Increases in drive temperature often signal spindle motor problems.
Data Throughput—Reduction in the transfer rate of the drive can signal various internal problems.
The dynamic information 86 is updated periodically and when the device is added to the storage system.
The gathered information may be stored and updated, for example, in a table as illustrated in
In
An example of the calculation of weighting formula comprises:
Weighting of the entity=β*StaticParameterValue+d*DynamicParameterValue
where, in an example of magnetic disk drives:
StaticParameterValue=β1*MTBF+β2*ReadPerformance+β3*other QoS+ . . .
DynamicParameterValue=d1*SMART1+d2*SMART2+d3*SMART3+ . . .
The parameters utilized in the collection and the values to accomplish the weighting are designed to create a reliability weighting that comprises an assessment of the probability of operation of the physical entity without permanent loss of data, and the probability of operation is related to a probability of length of time without permanent loss of data.
The values of β, d, β1, β2, β3, . . . , d1, d2, d3, . . . are defined by the user and/or the system, and may be based on preference or requirements. The values are established to give balance and preference to the various parameters, e.g. MTBF may be a ratio with respect to 100 years, read performance may be a ratio with respect to 2 MB/s (megabytes per second), etc.
Referring to
Step 97 determines whether all of the entities participating in the RAID array have been weighted and added to the sorted list 95. If not, the process proceeds back to step 93 to calculate and assign the reliability weighting to the next entity and add the listing of the entity to the sorted list 95. When step 97 indicates that all of the entities have been added to the sorted list, step 99 indicates that the sort provided in list 95 is complete for all of the entities participating in the RAID array.
These reliability weightings provide an assessment of probability of length of time of operation of the physical entity without permanent loss of data. Thus, in the example of
Referring to
To provide mirrored pairs, it is necessary that the data storage physical entities participating in the RAID array comprises an even number. Thus, the division of the list 95 into two sets of equal size comprising the upper 106 and lower 107 halves of the list is an example of a way to equalize the number of data storage physical entities at each side of the RAID mirrors. Should the lowest ranked physical entity of the set 106 comprising the upper half of the sorted list have the same reliability weighting as the highest ranked physical entity of the set 107 comprising the lower half of the sorted list, the entities are assigned to the first or second set of entities by some other means, such as by random selection or by drive number sequence. Another way of approaching the division of the list is that the set 106 comprising the upper half of the list comprises physical entities having at least a median and greater reliability weighting, and the set 107 comprising the lower half of the list comprises data storage physical entities having at least a median and lesser reliability weighting. Any drive having a median reliability rating is arranged at the side of a mirrored pair required to equalize the number of drives at each side of the arrangement. Thus, if two drives have the same median reliability weighting, one is assigned to the upper half set, and the other is assigned to the lower half set.
The selection of the mirrored pairs is accomplished in steps 108 and 109. For each mirrored pair, in step 108, one entity is selected from the first set 106 of physical entities and one entity is selected from the second set 107 of physical entities. In step 109, the selected physical entities are designated as forming a RAID mirrored pair. For example, in step 108, drive 2 is selected from set 106 and drive 4 is selected from set 107, and step 109 forms a mirrored pair of drive 2 and drive 4, as shown in
The result is that the RAID array, by virtue of the placement of one of the data storage physical entities with at least a median and greater said reliability weighting in each mirrored pair, assures that each pair in the array has an entity with the longest potential lifetime, and that the RAID array has the potential lifetime dictated by an entity from the set of drives having the minimum of the longest potential lifetimes.
To put the statement in perspective, using the algorithm discussed above:
In
Mirror_Pair1: <Drive1 61,71, Drive3 63,73>
Mirror_Pair2: <Drive2 62,72, Drive4 64,74>
Lifetime of Mirror_Pair1=Max (Lifetime of Drive1 61,71, Lifetime of Drive3 63,73)
Lifetime of Mirror_Pair2=Max (Lifetime of Drive2 62,72, Lifetime of Drive4 64,74)
Reliability of RAID array=Min (Lifetime of Mirror_Pair1, Lifetime of Mirror_Pair2)
Inserting the lifetimes of the reliability weightings:
Thus, in
Mirror_Pair1: <Drive1 (3.4), Drive3 (7.7)>
Mirror_Pair2: <Drive2 (8.0), Drive4 (4.5)>
Lifetime of Mirror_Pair1=Max (Lifetime of Drive1 (3.4), Lifetime of Drive3 (7.7)=7.7
Lifetime of Mirror_Pair2=Max (Lifetime of Drive2 (8.0), Lifetime of Drive4 (4.5)=8.0
Reliability of RAID array=Min (Lifetime of Mirror_Pair1=7.7), Lifetime of Mirror_Pair2=8.0)=7.7
This results in a potential lifetime of 7.7.
The process may be implemented with various numbers (even numbers) of physical data storage entities to form various numbers of mirrored pairs within a RAID array.
A person of ordinary skill in the art will appreciate that the embodiments of the present invention, disclosed herein, including the computer-implemented system 10 for storage of data in RAID arrays of
Any combination of one or more non-transient computer readable medium(s) may be utilized. The computer readable medium may be a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for embodiments of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).
Embodiments of the present invention are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer readable medium that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer readable medium produce an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
Those of skill in the art will understand that changes may be made with respect to the methods discussed above, including changes to the ordering of the steps. Further, those of skill in the art will understand that differing specific component arrangements may be employed than those illustrated herein.
While the preferred embodiments of the present invention have been illustrated in detail, it should be apparent that modifications and adaptations to those embodiments may occur to one skilled in the art without departing from the scope of the present invention as set forth in the following claims.
