The subject matter of this application is related to copending U.S. patent application Ser. No. 12/478,013, filed concurrently with this application, having common inventors with this application, assigned to the same assignee as this application, and the teachings of which are incorporated herein in its entirety by reference. The subject matter of this application is also related to U.S. patent application Ser. Nos. 12/436,227 filed May 6, 2009, 12/475,710 filed Jun. 1, 2009, 12/475,716 filed Jun. 1, 2009, 12/478,013 filed Jun. 4, 2009, 12/508,879 filed Jul. 24, 2009, 12/508,915 filed Jul. 24, 2009, 12/643,471 filed Dec. 21, 2009, 12/649,490 filed Dec. 30, 2009, 12/722,828 filed Mar. 12, 2010, 12/730,627 filed Mar. 24, 2010, 12/731,631 filed Mar. 25, 2010, 12/767,985 filed Apr. 27, 2010, 12/768,058 filed Apr. 27, 2010, 12/769,882 filed Apr. 29, 2010 and 12/769,910 filed Apr. 29, 2010.
The present invention relates to flash-memory generally, and, in particular, to flash-memory file system architectures or the like.
Disk-drive (or hard-drive) memory systems are being replaced with solid-state memory systems utilizing flash-memory technology. Compared to hard-drive systems, flash-memory systems offer the reliability of semiconductor-based memory along with less energy consumption and smaller size. While significant in-roads have been made in replacing hard-drives in consumer-based products such as in laptop computers, few of the hard-drives in enterprise-level systems have been replaced with solid-state drives for a variety of reasons. The most notable reason is the incompatibility of the file system structure in existing flash drive systems with the file system structure in enterprise-based hard-drives. This incompatibility is forced by the flash-memory architecture and by de facto hard-disk file structure system conventions.
Generally, flash-memory architecture requires the erasure of large blocks of memory but subsections, referred to as pages, may be written to as needed. Within each page, there are usually 2N bytes of memory (N is an integer and, at present, N ranges from 10 to 14 or more) for storing user data and an additional 100 to 500 or more bytes of memory for storing redundancy data (ECC) and file system information (e.g., metadata). The ECC is for detecting and correcting data stored in the corresponding user data in the page and the file system information is used for mapping virtual to physical addresses and vice-versa. As such, the additional bytes of memory are “hidden” from the user and are not available for storing data.
For consumer applications, hard-drive systems have data sectors that are generally arranged with data sized in powers of two, e.g., 28 or 210 bytes per sector. This works well with flash memories having similarly structured user data memory pages. However, for enterprise-based systems, the sectors are not sized by powers of two but larger, e.g., 520 or 528 bytes instead of 512 bytes (29). At present, forcing these larger sectors into existing flash-memory architectures results in inefficient designs with many unused bytes in each page, at least partially negating the advantages of flash-memory systems over hard-drive systems.
In one embodiment, the present invention is a flash-memory system having a plurality of blocks and a plurality of pages in each block, at least one page having 2N data memory locations and K spare memory locations. At least one of the pages is adapted to have 2M user data sectors for storing user data, each data sector having 2N-M+L memory locations therein, where M, N, L, and K are positive integers, N>M, M≧1, and 1≦L<2N-M.
In another embodiment, the present invention comprises a method of retaining user data in a memory system, the method comprising: providing a flash-memory system organized into a plurality of blocks and a plurality of pages in each block, at least one of the pages having 2N data locations and K spare locations, the at least one page having 2M user data sectors therein, each sector having 2N-M+L locations therein, where M, N, L, and K are positive integers, N>M, M≧1, 1≦L<2N-M; and storing user data in at least one of the 2M user data sectors.
The aspects, features, and advantages of the present invention will become more fully apparent from the following detailed description, the appended claims, and the accompanying drawings in which like reference numerals identify similar or identical elements.
Application Programming Interface (API): protocol or format used by an application to communicate with a different application.
Block: the number of memory locations that are erased at one time in a flash-memory.
ECC data: error correction code information. This is redundancy information relating to data written to the memory used to detect and, if possible, correct data read from the memory.
Inter-Processor Communications (IPC): communications protocol for communications between processors or systems.
Page: the minimum number of memory locations that are read or written at a time within a block.
Sector: the number of bytes that are grouped together within a page. There are multiple sectors in each page.
For purposes of this description and unless explicitly stated otherwise, each numerical value and range should be interpreted as being approximate as if the word “about” or “approximately” preceded the value of the value or range. Further, signals and corresponding nodes, ports, inputs, or outputs may be referred to by the same name and are interchangeable.
Additionally, reference herein to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments necessarily mutually exclusive of other embodiments. The same applies to the terms “implementation” and “example.”
Also for purposes of this description, the terms “couple,” “coupling,” “coupled,” “connect,” “connecting,” or “connected,” refer to any manner known in the art or later developed in which a signal is allowed to be transferred between two or more elements and the interposition of one or more additional elements is contemplated, although not required. Conversely, the terms “directly coupled,” “directly connected,” etc., imply the absence of such additional elements.
In
As is well known in the art, the flash-memory may be composed of a single chip or multiple chips. Operation of the flash-memory 108 (e.g., erase, write, and read operations) is well known; see, for example, “NAND Flash 101: An Introduction to NAND Flash and How to Design it into Your Next Product” by Micron Technology, Inc., TN-29-19, July, 2004, incorporated herein by reference in its entirety. Each memory chip is physically organized into blocks and typical memories have blocks of one megabyte or more. The blocks are in turn subdivided into pages, and each page further subdivided into sectors. Thus, and as will be discussed in more detail below, a file system utilizing a flash-memory writes data into a block on a page-by-page basis, each page having multiple sectors therein. Similarly, when reading data from the memory, a page at a time is read and all the sectors within the page are read. Data is organized into sectors to retain compatibility with hard-disk file systems.
Existing flash memories typically have a power of two pages per block (e.g., 28, 210, etc.); each page has power of two memory locations (e.g., 212 memory locations, each location storing a byte) allocated for data storage and a spare set of memory locations available for reading and writing along with the other bytes in the page (e.g., 128, 218, 224 bytes for a 212 byte page, and 376 or 520 bytes for a 213 byte page). The spare set of memory locations has heretofore been reserved for storing ECC data segments and other metadata (used by the FTL process, described below) for the data stored in the memory locations of the page. In a typical consumer flash-memory system, such as in an MP3 player, each page has multiple sectors, each sector having 29 memory locations (512 bytes). This results in the page storing eight sectors (212/29=8), along with the spare memory locations containing the necessary ECC data and metadata related to the sectors in the page. However, hard-disk enterprise (non-consumer) systems have 520 or 528 bytes in each sector and attempting to store eight of these larger sectors in a page will exceed the number of memory locations allocated for data storage.
As is well understood in the art, the FTL 208 translates the logical-to-physical addresses (and vice-versa) of data stored in the memory 108 by using, for example, the metadata stored in the flash-memory page being accessed. Further, the FTL 208 provides “garbage collection” (the identification of unused flash-memory pages for reuse), error recovery (adding redundancy information to data stored in the memory 108 and correcting erroneous data read from the memory), and wear leveling (spreading the erasing and writing of blocks of memory cells over the entire flash-memory to avoid repeatedly erasing and writing a subset of the memory blocks, thereby making the flash-memory last longer than would otherwise occur). Details on how an FTL operates and is organized are described in “A Superblock-based Flash Translation Layer for NAND Flash-memory” by Kang et al., presented at EMSOFT '06, 22-25 Oct. 2006 in Seoul, Korea, included herein by reference in its entirety.
In accordance with one embodiment of the invention,
In accordance with another embodiment of the invention,
Advantageously, by using the spare area (304, 404) that was previously off-limits to user data, enterprise-sized sectors can be efficiently stored in flash memories with little wasted memory, thereby making flash-memory systems compatible with existing hard-drive storage systems in enterprise system applications.
It is understood that the FTL process 208 (
It is further understood that various changes in the details, materials, and arrangements of the parts and processes which have been described and illustrated in order to explain the nature of this invention may be made by those skilled in the art without departing from the scope of the invention as expressed in the following claims.
The use of figure numbers and/or figure reference labels in the claims is intended to identify one or more possible embodiments of the claimed subject matter in order to facilitate the interpretation of the claims. Such use is not to be construed as necessarily limiting the scope of those claims to the embodiments shown in the corresponding figures.
Although the elements in the following method claims, if any, are recited in a particular sequence with corresponding labeling, unless the claim recitations otherwise imply a particular sequence for implementing some or all of those elements, those elements are not necessarily intended to be limited to being implemented in that particular sequence.
Number | Name | Date | Kind |
---|---|---|---|
4389715 | Eaton et al. | Jun 1983 | A |
4402046 | Cox et al. | Aug 1983 | A |
4584681 | Singh et al. | Apr 1986 | A |
5121480 | Bonke et al. | Jun 1992 | A |
5297029 | Nakai | Mar 1994 | A |
5353410 | Macon et al. | Oct 1994 | A |
5732409 | Ni | Mar 1998 | A |
5734821 | Chung et al. | Mar 1998 | A |
5974502 | DeKoning et al. | Oct 1999 | A |
6049838 | Miller et al. | Apr 2000 | A |
6081849 | Born et al. | Jun 2000 | A |
6145072 | Shams et al. | Nov 2000 | A |
6158004 | Mason et al. | Dec 2000 | A |
6212617 | Hardwick | Apr 2001 | B1 |
6247040 | Born et al. | Jun 2001 | B1 |
6324594 | Ellis et al. | Nov 2001 | B1 |
6363470 | Laurenti et al. | Mar 2002 | B1 |
6385683 | DeKoning et al. | May 2002 | B1 |
6449666 | Noeldner et al. | Sep 2002 | B2 |
6490635 | Holmes | Dec 2002 | B1 |
6567094 | Curry et al. | May 2003 | B1 |
6633942 | Balasubramanian | Oct 2003 | B1 |
6678785 | Lasser et al. | Jan 2004 | B2 |
6725329 | Ng et al. | Apr 2004 | B1 |
6751680 | Langerman et al. | Jun 2004 | B2 |
7069559 | Janssen et al. | Jun 2006 | B2 |
7286549 | Gaur | Oct 2007 | B2 |
7290066 | Voorhees et al. | Oct 2007 | B2 |
7408834 | Conley et al. | Aug 2008 | B2 |
7461183 | Ellis et al. | Dec 2008 | B2 |
7472331 | Kim | Dec 2008 | B2 |
7512847 | Bychkov et al. | Mar 2009 | B2 |
7590803 | Wintergerst | Sep 2009 | B2 |
7650449 | Lu | Jan 2010 | B2 |
7653778 | Merry et al. | Jan 2010 | B2 |
7925847 | Ellis et al. | Apr 2011 | B2 |
20030051078 | Yoshitake | Mar 2003 | A1 |
20030110325 | Roach et al. | Jun 2003 | A1 |
20030167395 | Chang et al. | Sep 2003 | A1 |
20040044873 | Wong et al. | Mar 2004 | A1 |
20040177212 | Chang et al. | Sep 2004 | A1 |
20050114729 | Nielsen et al. | May 2005 | A1 |
20050144516 | Gonzalez et al. | Jun 2005 | A1 |
20050203988 | Nollet et al. | Sep 2005 | A1 |
20060050693 | Bury et al. | Mar 2006 | A1 |
20060095611 | Winchester et al. | May 2006 | A1 |
20060123259 | Yokota et al. | Jun 2006 | A1 |
20060155920 | Smith et al. | Jul 2006 | A1 |
20070028040 | Sinclair | Feb 2007 | A1 |
20070109856 | Pellicone et al. | May 2007 | A1 |
20070255889 | Yogev et al. | Nov 2007 | A1 |
20070266200 | Gorobets et al. | Nov 2007 | A1 |
20070300130 | Gorobets | Dec 2007 | A1 |
20080034153 | Lee et al. | Feb 2008 | A1 |
20080052446 | Lasser et al. | Feb 2008 | A1 |
20080082726 | Elhamias | Apr 2008 | A1 |
20080120456 | Lee et al. | May 2008 | A1 |
20080140916 | Oh et al. | Jun 2008 | A1 |
20080155145 | Stenfort | Jun 2008 | A1 |
20080162079 | Astigarraga et al. | Jul 2008 | A1 |
20080224924 | Lethbridge | Sep 2008 | A1 |
20080263307 | Adachi | Oct 2008 | A1 |
20080279205 | Sgouros et al. | Nov 2008 | A1 |
20090138663 | Lee et al. | May 2009 | A1 |
20090172308 | Prins et al. | Jul 2009 | A1 |
20090271562 | Sinclair | Oct 2009 | A1 |
20090271796 | Kojima | Oct 2009 | A1 |
20090282301 | Flynn et al. | Nov 2009 | A1 |
20090285228 | Bagepalli et al. | Nov 2009 | A1 |
20090287859 | Bond et al. | Nov 2009 | A1 |
20090300277 | Jeddeloh | Dec 2009 | A1 |
20090313444 | Nakamura | Dec 2009 | A1 |
20100011260 | Nagadomi et al. | Jan 2010 | A1 |
20100023800 | Harari et al. | Jan 2010 | A1 |
20100122148 | Flynn et al. | May 2010 | A1 |
20100269015 | Borchers et al. | Oct 2010 | A1 |
20100325317 | Kimelman et al. | Dec 2010 | A1 |
20110041039 | Harari et al. | Feb 2011 | A1 |
20110055458 | Kuehne | Mar 2011 | A1 |
20110093766 | Murray et al. | Apr 2011 | A1 |
20110099355 | Tran | Apr 2011 | A1 |
Entry |
---|
Andrew Birrell & Michael Isard, et al., A Design For High-Performance Flash Disks, ACM SIGOPS Operating Systems Review, vol. 41, Issue 2, pp. 88-93, (Apr. 2007). |
Sun et al.; On the Use of Strong BCH Codes for Improving Multilevel NAND Flash Memory Storage Capacity; ECSE Department., Rensselaer Polytechnic Institute, Aug. 2006; USA. |
Kang et al.; A Superblock-based Flash Translation Layer for NAND Flash Memory: Computer Science Division; Korea Advanced Institute of Science and Technology (KAIST); EMSOFT '06 Seoul, Korea; pp. 161-170; Published by ACM, New York, NY, USA. |
Micron Technolog, Inc.; NAND Flash 101: An Introduction to NAND Flash and How to Design it In to Your Next Product; TN-29-19; 2006; pp. 1-28; Micron Technology, Inc., Boise, Idaho, USA. |
TCG Core Architecture Specification, Version 2.0, Trusted Computing Group, 2009 USA. |
TCG Storage Interface Interactions Specification, Version 1.0, Trusted Computing Group, 2009 USA. |
TCG Storage SSC: Enterprise, Version 1.0, Trusted Computing Group 2009 USA. |
TCG Storage SSC: Opal, Version 1.0, Trusted Computing Group 2009 USA. |
Specification for the Advanced Encryption Standard (AES), Federal Information Processing Standard (FIPS) Publication 197, 2001 USA. |
Specification for the Secure Hash Standard (SHS), FIPS Publication 180-3 (2008), National Institute of Standards and Technology (NIST) USA. |
Number | Date | Country | |
---|---|---|---|
20100313097 A1 | Dec 2010 | US |