Trim (computing)

A trim command (known as TRIM in the ATA command set, and UNMAP in the SCSI command set) allows an operating system to inform a solid-state drive (SSD) which blocks of data are no longer considered in use and can be wiped internally.[1]

Not to be confused with trimming in computer programming, which is about removing space characters.

Trim was introduced soon after SSDs were introduced. Because low-level operation of SSDs differs significantly from hard drives, the typical way in which operating systems handle operations like deletes and formats resulted in unanticipated progressive performance degradation of write operations on SSDs.[2] Trimming enables the SSD to more efficiently handle garbage collection, which would otherwise slow future write operations to the involved blocks.[3]

Although tools to "reset" some drives to a fresh state were already available before the introduction of trimming, they also delete all data on the drive, which makes them impractical to use for ongoing optimization.[4] By 2014, many SSDs had internal background garbage collection mechanisms that worked independently of trimming. Although this successfully maintained their performance even under operating systems that did not support trim, it had the associated drawbacks of increased write amplification and wear of the flash cells.[5]

Background

Because of the way that many file systems handle delete operations, by flagging data blocks as "not in use",[6][7] storage media (SSDs, but also traditional hard drives) generally do not know which sectors/pages are truly in use and which can be considered free space. Contrary to (for example) an overwrite operation, a delete will not involve a physical write to the sectors that contain the data. Since a common SSD has no knowledge of the file system structures, including the list of unused blocks/sectors, the storage medium remains unaware that the blocks have become available. While this often enables undelete tools to recover files from electromechanical hard disks,[7][8] despite the files being reported as "deleted" by the operating system, it also means that when the operating system later performs a write operation to one of the sectors, which it considers free space, it effectively becomes an overwrite operation from the point of view of the storage medium. For magnetic disks, an overwrite of existing data is no different from writing into an empty sector, but because of how some SSDs function at the lowest level, an overwrite produces significant overhead compared to writing data into an empty page, potentially crippling write performance.[7][9]

SSDs store data in flash memory cells that are grouped into pages typically of 4 to 16 kiB, grouped together into blocks of typically 128 to 512 pages. Example: 512 kiB blocks that group 128 pages of 4 kiB each.[6][10] NAND flash memory cells can be directly written to only when they are empty. If they happen to contain data, the contents must be erased before a write operation. An SSD write operation can be done on a single page but, due to hardware limitations, erase commands always affect entire blocks;[10] consequently, writing data to empty pages on an SSD is very fast, but slows down considerably once previously written pages need to be overwritten. Since an erase of the cells in the page is needed before it can be written again, but only entire blocks can be erased, an overwrite will initiate a read-erase-modify-write cycle:[6][11] the contents of the entire block are stored in cache, then the entire block is erased from the SSD, then the overwritten page is written to the cached block, and only then can the entire updated block be written to the flash medium. This phenomenon is known as write amplification.[12][13]

Operation

The TRIM command enables an operating system to notify the SSD of pages which no longer contain valid data. For a file deletion operation, the operating system will mark the file's sectors as free for new data, then send a TRIM command to the SSD. After trimming, the SSD will not preserve any contents of the block when writing new data to a page of flash memory, resulting in less write amplification (fewer writes), higher write throughput (no need for a read-erase-modify sequence), thus increasing drive life.

Different SSDs implement the command somewhat differently, so performance can vary.[3][8]

TRIM tells the SSD to mark a LBA region as invalid and subsequent reads on the region will not return any meaningful data. For a very brief time, the data could still reside on the flash internally. However, after the TRIM command is issued and garbage collection has taken place, it's highly unlikely that even a forensic scientist would be able to recover the data.[14]

Implementation

Operating system support

Trimming is only effective on operating systems which support it. The table below identifies each notable operating system and the first version supporting the command. Additionally, older solid-state drives designed before the addition of the TRIM command to the ATA standard will need firmware updates, otherwise the new command will be ignored. However, not every drive can be upgraded to support trimming.

The support for TRIM also varies by what the particular filesystem driver on the operating system is capable of, since only a program with an understanding of what parts of the disk is free space can safely issue the command, and on the system level this ability tends to lie in the filesystem driver itself.

Operating System Supported since Notes
DragonFly BSD 2011-05May 2011[15]
FreeBSD 2010-078.1 – July 2010[16] Support was added at the block device layer in 8.1. Filesystem support was added in FreeBSD 8.3 and FreeBSD 9, beginning with UFS.[17] ZFS trimming support was added in FreeBSD 9.2.[18][19] FreeBSD 10 supports trimming on software RAID configurations.[20]
NetBSD 2012-10October 2012[21]
Linux 2008-12-252.6.28–25 December 2008[22] Initial support for discard operations was added for FTL NAND flash devices in 2.6.28. Support for the ATA TRIM command was added in 2.6.33.[23]

Not all filesystems make use of trim. Among the filesystems that can issue trim requests automatically are Ext4,[24] Btrfs,[25] FAT, GFS2, JFS,[26], XFS[27], and NTFS-3G. However, in some distributions, this is disabled by default due to performance concerns,[28] in favor of scheduled trimming on supported SSDs.[29] Ext3, NILFS2 and OCFS2 offer ioctls to perform offline trimming. The TRIM specification calls for supporting a list of trim ranges, but as of kernel 3.0 trim is only invoked with a single range that is slower.[30]

macOS 2011-06-2310.6.8–23 June 2011[31] Although the AHCI block device driver gained the ability to display whether a device supports the TRIM operation in 10.6.6 (10J3210),[32] the functionality itself remained inaccessible until 10.6.8, when the TRIM operation was exposed via the IOStorageFamily and filesystem (HFS+) support was added. Until 10.10.4, Mac OS X natively enabled TRIM only for Apple-branded SSDs; third-party utilities are available to enable it for other brands. Old third party TRIM drivers stopped working as of the Yosemite update.[33] Updated drivers now exist that work with OS X Yosemite.[34][35] In Mac OS X update 10.10.4, Apple added a command line utility, trimforce, that can be used to enable TRIM on third-party SSDs.[36]
Microsoft Windows 2009–10Windows 7 and Windows Server 2008 R2 – October 2009[37][38] Windows 7 initially supported TRIM only for drives in the AT Attachment family including Parallel ATA and Serial ATA, and did not support this command for any other devices including Storport PCI-Express SSDs even if the device itself would accept the command.[39] It is confirmed that with native Microsoft drivers the TRIM command works on Windows 7 in AHCI and legacy IDE / ATA Mode.[40] Windows 8 and later Windows operating systems support trim for PCI Express SSDs based on NVMe, and the unmap command which is a full analog of the TRIM command from Serial ATA for devices that use the SCSI driver stack, including USB Attached SCSI Protocol (UASP). Microsoft has released an update for Windows 7, KB2990941, which when integrated into Windows 7 Setup using DISM, adds NVM Express support including TRIM for PCIe SSDs.

TRIM is known to be supported for ReFS and NTFS, both of which implement a DisableDeleteNotify switch for disabling it.[41] Sources disagree on whether TRIM support exists for other filesystems.

OpenSolaris 2010-07July 2010[42]
Android 2013-74.3[43] – 24 July 2013[44] Runs fstrim automatically up to once every 24 hours if the device has been idle for at least an hour and is at least 80% charged (30% if connected to a charger).[43]

RAID issues

As of January 2017, support for the TRIM command is not implemented in most hardware-based RAID technologies. However, software RAID implementations often do include support for TRIM.

Windows

Windows 10 offers support for TRIM in SSD RAID volumes using the "optimize drives" option when configuring a RAID volume.

macOS

The macOS RAID driver does not support TRIM. This is true for all versions of Mac OS X from 10.7 through macOS 10.12.x.

TRIM is supported for RAID (0,1,4,5 & 10) volumes when using the third-party SoftRAID® application, including TRIM support with non-Apple SSD devices. (Note: TRIM for non-Apple SSD devices must be specifically enabled using the terminal command "sudo trimforce enable")

Linux

TRIM is available with RAID volumes in post-January-2011 releases of the Linux kernel's dmraid, which implements BIOS-assisted "fake hardware RAID" support, and which now passes through any TRIM requests from the filesystem that sits on the RAID array.[45]

Not to be confused with dmraid, Linux's general-purpose software RAID system, mdraid, has experimental support for batch-based (rather than live, upon file deletion) TRIM on RAID 1 arrays when systems are configured to periodically run the mdtrim utility on filesystems (even those like ext3 without native TRIM support).[46] In later versions of Linux, e.g. Red Hat Enterprise Linux 6.5 and beyond, mdraid supports actually passing through TRIM commands in real-time, rather than just as a batch job.[47]

However, note that Red Hat recommends against using software RAID levels 1, 4, 5, and 6 on SSDs with most RAID technologies, because during initialization, most RAID management utilities (e.g. Linux's mdadm) write to all blocks on the devices to ensure that checksums (or drive-to-drive verifies, in the case of RAID 1 / 10) operate properly, causing the SSD to believe that all blocks other than in the spare area are in use, significantly degrading performance.[48]

On the other hand, Red Hat does recommend the use of RAID 1 or RAID 10 for LVM RAIDs on SSDs, as these levels support TRIM ("discard" in Linux terminology), and the LVM utilities do not write to all blocks when creating a RAID 1 or RAID 10 volume.[47]

For a short time in March 2010, users were led to believe that the Intel Rapid Storage Technology (RST) 9.6 drivers supported TRIM on RAID volumes, but Intel later clarified that TRIM was supported for the BIOS settings of AHCI mode and RAID mode, but not if the drive was part of a RAID volume.[49]

As of August 2012, Intel confirms that 7-series chipsets with Rapid Storage Technology (RST) 11.2 drivers support TRIM for RAID 0 in Microsoft Windows 7.[50] While Intel did not confirm support for 6-series chipsets, TRIM on RAID 0 volumes has been shown to work on Z68, P67, and X79 chipsets by hardware enthusiasts with a modified RAID option ROM.[51] It is speculated that the lack of official support for 6-series chipsets is due to validation costs[52] or an attempt to encourage consumers to upgrade,[53] rather than for technical reasons.

An exception to the need for a modified option ROM on motherboards with an X79 chipset is if the manufacturer has added a ROM switch; this entails both the RST and RST-E ROMs being inside the BIOS/UEFI. This allows the RST ROM to be used instead of the RST-E ROM, allowing TRIM to function.[54] Intel notes that best performance can be achieved by using a driver with same version as the ROM; for example, if the BIOS/UEFI has an 11.0.0.0m option ROM, an 11.x version driver should be used.[55]

Enabling unsupported filesystems

Where the filesystem does not automatically support TRIM, some utilities can send trimming commands manually. Usually they determine which blocks are free and then pass this list as a series of trimming commands to the drive. These utilities are available from various manufacturers (e.g. Intel,[56] G.Skill[57]), or as general utilities (e.g. Linux's hdparm "wiper" since v9.17,[58][59] or mdtrim, as mentioned above). Both hdparm and mdtrim finds free blocks by allocating a large file on the filesystem and resolving what physical location it was assigned to.

Hardware support

ATA

The TRIM command specification[60] has been standardized as part of the AT Attachment (ATA) interface standard, led by Technical Committee T13 of the International Committee for Information Technology Standards (INCITS).[61] TRIM is implemented under the DATA SET MANAGEMENT command (opcode 06h) of the draft ACS-2 specification.[62] The ATA standard is supported by both parallel (IDE, PATA) and serial (SATA) ATA hardware.

A drawback of the original ATA TRIM command is that it was defined as a non-queueable command and therefore could not easily be mixed with a normal workload of queued read and write operations. SATA 3.1 introduced a queued TRIM command to remedy this.[63]

There are different types of TRIM defined by SATA Words 69 and 169 returned from an ATA IDENTIFY DEVICE command:

  • Non-deterministic TRIM: Each read command to the Logical block address (LBA) after a TRIM may return different data.
  • Deterministic TRIM (DRAT): All read commands to the LBA after a TRIM shall return the same data, or become determinate.
  • Deterministic Read Zero after TRIM (RZAT): All read commands to the LBA after a TRIM shall return zero.

There is additional information in SATA Word 105 that describes the Maximum number of 512-byte blocks per DATA SET MANAGEMENT command that a drive can support. Typically this defaults to 8 (or 4 kB) but many drives reduce this to 1 to meet the Microsoft Windows Hardware Requirements for TRIM, that command completion time shall not exceed 20 ms or 8 ms × (number of LBA range entries), whichever is greater, and shall always be less than 600 ms.[64]

An individual LBA range is called an LBA Range Entry and is represented by eight bytes. The LBA is expressed by the LBA Range Entry's first six bytes and the Range Length is a zero-based counter (e.g., 0=0 and 1=1) represented by the remaining two bytes. If the two-byte range length is zero, then the LBA Range Entry shall be discarded as padding.[65] This means that for each 512-byte block of TRIM ranges that a device supports, the maximum is 64 ranges of 32 MB, or 2 GB. If a device supports SATA Word 105 at 8 then it should be able to trim 16 GB in a single TRIM (DATA SET MANAGEMENT) command.

SCSI

SCSI provides the UNMAP command (a full analog of TRIM), and the WRITE SAME command (10 and 16 variants) with the UNMAP flag set.[66]

SD/MMC

The MultiMediaCard and SD ERASE (CMD38) command provides similar functionality to the ATA TRIM command, although it requires that erased blocks be overwritten with either zeroes or ones. eMMC 4.5 further defines a "discard" sub-operation that more closely matches ATA TRIM in that the contents of discarded blocks can be considered indeterminate (i.e., "don't care").

NVM Express

The NVM Express command set has a generic Dataset Management command, for hinting the host's intent to the storage device on a set of block ranges. One of its operations, deallocate performs trim. It also has a Write Zeroes command that provides a deallocate hint and allows the disk to trim and return zeroes.

Disadvantages
  • When encryption is in use, using the TRIM command reveals information about which blocks are in use and which are not, reducing plausible deniability.[67]
  • The original version of the TRIM command has been defined as a non-queued command by the T13 subcommittee, and consequently can incur massive execution penalty if used carelessly, e.g., if sent after each filesystem delete command. The non-queued nature of the command requires the driver to first wait for all outstanding commands to be finished, issue the TRIM command, then resume normal commands. TRIM can take a lot of time to complete, depending on the firmware in the SSD, and may even trigger a garbage collection cycle. This penalty can be minimized in solutions that periodically do a batched TRIM, rather than trimming upon every file deletion, by scheduling such batch jobs for times when system utilization is minimal. This TRIM disadvantage has been overcome in Serial ATA revision 3.1 with the introduction of the Queued TRIM Command.[68][69]
  • Faulty drive firmware that misreports support for queued TRIM or has critical bugs in its implementation has been linked to serious data corruption in several devices, most notably Micron and Crucial's M500[70] and Samsung's 840 and 850 series.[71] The data corruption has been confirmed on the Linux operating system (the only OS with queued trim support as of 1 July 2015).[72]

These devices are blacklisted in the Linux kernel's libata-core.c to force sending non-queued TRIM commands (ATA_HORKAGE_NO_NCQ_TRIM) to these drives instead of queued TRIM commands:[73]

  • Micron/Crucial M500 using all firmware versions including factory recertified SSDs
  • Micron M510 using firmware version MU01
  • Micron/Crucial M550 using firmware version MU01
  • Crucial MX100 using firmware version MU01
  • Samsung 840 and 850 series SSDs using all firmware versions

This file also blacklists the SuperSSpeed S238 against TRIM in general due to causing the wrong blocks to lose data when TRIM is issued.[73][74]

libata-core.c also has a whitelist to list SSDs that are reliably known to the subsystem's maintainers to correctly implement the DRAT and RZAT flags (ATA_HORKAGE_ZERO_AFTER_TRIM), rather than ignoring them, as many drives do. The whitelisted drives are as follows:[73]

  • Crucial SSDs
  • Intel SSDs excluding the Intel SSD 510
  • Micron SSDs
  • Samsung SSDs
  • Seagate SSDs[75]
gollark: 3 dumb? 4 dumb?
gollark: That makes sense, since if something is a bomb you will obviously stay as close to it and use it as much as possible.
gollark: `((0 * 4) + +[(a * b), (b * c), (a + 3)])` you.
gollark: osmarkscalculator™ MUST simplify expressions, yet this makes it quite hard to.
gollark: No, this is actually bad.

See also

References
  1. "Intel High Performance Solid State Drive – Advantages of TRIM". Intel.com. Intel. 14 September 2010. Retrieved 21 February 2012.
  2. Shimpi, Anand Lal (18 March 2009). "The SSD Anthology". AnandTech.com. p. 4. Retrieved 19 June 2010.
  3. Shimpi, Anand Lal. (18 March 2009). p. 10.
  4. Shimpi, Anand Lal. (18 March 2009). p. 11.
  5. "Kingston SSDNow V Plus 100 Review". AnandTech.com. Anand Lal Shimpi. 11 November 2010. Retrieved 10 December 2010.
  6. Savill, John (21 April 2009). "I heard solid-state disks (SSDs) suffer from a decline in write performance as they're used. Why?". WindowsITPro. Archived from the original on 30 June 2012. Retrieved 19 June 2010.
  7. Shimpi, Anand Lal. (18 March 2009). p. 7.
  8. Savill, John (22 April 2009). "What is the TRIM function for solid-state disks (SSDs) and why is it important?". WindowsITPro. Archived from the original on 9 February 2013. Retrieved 19 June 2010.
  9. Malventano, Allyn (13 February 2009). "Long-term performance analysis of Intel Mainstream SSDs". PC Perspective. Retrieved 10 February 2012.
  10. Shimpi, Anand Lal. (18 March 2009). p. 5.
  11. Shimpi, Anand Lal. (18 March 2009). p. 8.
  12. "Write Amplification: Intel's Secret Sauce". ExtremeTech.com. Retrieved 6 November 2010.
  13. "Inside the X25-M Controller: Wear Leveling, Write Amplification Control". TomsHardware.com. 8 September 2008. Retrieved 6 November 2010.
  14. "Too TRIM? When SSD Data Recovery is Impossible". TechGage. TechGage. 5 March 2010. Retrieved 21 August 2018.
  15. "Preliminary TRIM support".
  16. "FreeBSD 8.1-RELEASE Release Notes: Disks and Storage". Retrieved 1 September 2010.
  17. "Add kernel side support for BIO_DELETE/TRIM on UFS". Retrieved 1 February 2011.
  18. "Added ZFS TRIM support which is enabled by default". Retrieved 5 June 2013.
  19. "FreeBSD 9.2-RELEASE Announcement". Retrieved 4 October 2013.
  20. "Base revision 242323 – Add basic BIO_DELETE support to GEOM RAID class for all RAID levels". Retrieved 30 October 2012.
  21. "TRIM support in NetBSD". 19 October 2012.
  22. "Block device changes in Linux 2.6.28".
  23. "Storage changes in Linux 2.6.33". Kernel Newbies. Retrieved 1 September 2010.
  24. "HOWTO: Configure Ext4 to Enable TRIM Support for SSDs on Ubuntu and Other Distributions – Forked by Nicolay".
  25. "Btrfs improvements in Linux 2.6.32".
  26. "JFS Filesystem can now handle trim". Retrieved 28 March 2012.
  27. "Filesystem changes in Linux 3.0".
  28. Freemyer, Greg (2010). "OpenSUSE forum – SSD detection when creating first time fstab".
  29. "How to enable TRIM?". Ask Ubuntu. Retrieved 11 November 2015.
  30. "SDB:SSD discard (trim) support".
  31. "Mac OS X 10.6.8 Brings TRIM Support for Apple SSDs, Graphics Improvements". 24 February 2012.
  32. "Macbook Pros ship with active SSD TRIM support". 4 March 2011.
  33. "Yosemite kills third-party SSD support".
  34. "Chameleon SSD Optimizer".
  35. "Cindori Software".
  36. Lee Hutchinson (30 June 2015). "Latest OS X update allows you to enable TRIM for third-party SSDs". Ars Technica. Condé Nast. Retrieved 1 July 2015.
  37. "Support and Q&A for Solid-State Drives". MSDN. 9 May 2009. Retrieved 12 August 2010.
  38. "Windows 7 Enhancements for Solid-State Drives" (PDF). Microsoft downloads. Microsoft Corporation. 12 November 2008. Retrieved 8 July 2009.
  39. Geoff Gasior (2012) OCZ's RevoDrive 3 X2 240GB solid-state drive
  40. Coles, Olin (15 April 2010). "SSD Benchmark Tests: SATA IDE vs AHCI Mode – AHCI vs IDE Final Thoughts". BenchmarkReviews. Retrieved 20 August 2013.
  41. "Fsutil behavior". Microsoft Documentation. Retrieved 17 November 2019.
  42. "SATA TRIM support in OpenSolaris". 29 July 2010.
  43. "Android 4.3 Update Brings TRIM to All Nexus Devices". 29 July 2013. Retrieved 30 July 2013.
  44. "Android 4.3 announced, rolling out to Nexus devices today".The Verge. 24 July 2013. Retrieved 24 July 2013.
  45. "Possible to get SSD TRIM (discard) working on ext4 + LVM + software RAID in Linux? – Server Fault". Retrieved 29 November 2011.
  46. "Cyberax/mdtrim – GitHub". Retrieved 29 November 2011.
  47. "Red Hat Enterprise Linux 6 Storage Administration Guide, Chapter 21. Solid-State Disk Deployment Guidelines". Retrieved 9 March 2016.
  48. "Red Hat Enterprise Linux 7 Storage Administration Guide". 11 November 2015. Retrieved 29 February 2016.
  49. "Is there TRIM support for RAID configurations?". Intel. 26 March 2010. Retrieved 6 November 2010.
  50. "Intel Confirms RAID0 TRIM Support on 7-Series Chipsets". 19 August 2012. Retrieved 14 January 2013.
  51. "RAID0 trim and SRT seems possible on 6-series and 5-Series". 18 November 2012. Retrieved 14 January 2013.
  52. "TRIM & RAID-0 SSD Arrays Work With Intel 6-Series Motherboards Too". 28 November 2012. Retrieved 14 January 2013.
  53. "Intel Brings TRIM to RAID-0 SSD Arrays on 7-Series Motherboards, We Test It". 16 August 2012. Retrieved 14 January 2013.
  54. Anand Lal Shimpi. "Intel Brings TRIM to RAID-0 SSD Arrays on 7-Series Motherboards, We Test It".
  55. Doug Crowthers. "TRIM Command Confirmed With RAID 0 on Intel 7 Series". Tom's Hardware.
  56. "Intel SSD Optimizer White Paper" (PDF). Intel Corporation. Archived from the original (PDF) on 30 June 2012. Retrieved 23 January 2010.
  57. "wiper.exe for Falcon Series". XtremeSystems. Retrieved 23 January 2010.
  58. "hdparm-9.17 released, with experimental trim/wiper scripts for SSDs". Retrieved 14 August 2010.
  59. "hdparm project page". sourceforge.net. Retrieved 14 August 2010.
  60. "Data Set Management Commands Proposal for ATA8-ACS2 (revision 6)". INCITS T13. INCITS. 12 December 2007. Retrieved 8 July 2009. (draft specification T13/e07154r6)
  61. "T13 documents referring to TRIM". INCITS T13. INCITS. Retrieved 8 July 2009.
  62. "T13 ATA8 Draft Spec 1697-D" (PDF). INCITS. 23 June 2010. Retrieved 14 August 2010.
  63. "SATA 6Gb/s". Serial ATA International Organization. Retrieved 26 September 2012.
  64. "Device.Storage Requirements (Windows)". Retrieved 29 February 2016.
  65. "T13/2161-D: Information technology – ATA/ATAPI Command Set – 3 (ACS-3)" (PDF). Retrieved 29 February 2016.
  66. "Note for SCSI UNMAP support in FreeBSD mailing list".
  67. "Milan Broz's blog: TRIM & dm-crypt ... problems?". Milan Broz. Retrieved 1 October 2014.
  68. http://www.sata-io.org/technology/6Gbdetails.asp
  69. "SATA 3.1 spec brings swap standard, improved TRIM". The Tech Report.
  70. "Bug 71371 – [PATCH]Crucial M500, broken "queued TRIM" support".
  71. "Bug #1449005 "trim does not work with Samsung 840 EVO after firm..." : Bugs : fstrim package : Ubuntu".
  72. torvalds. "linux/libata-core.c at e64f638483a21105c7ce330d543fa1f1c35b5bc7 · torvalds/linux · GitHub". GitHub.
  73. torvalds. "linux/libata-core.c at master · torvalds/linux · GitHub". GitHub.
  74. torvalds. "libata: force disable trim for SuperSSpeed S238 · torvalds/linux@cda57b1". GitHub.
  75. "[1/3] libata: Whitelist SSDs that are known to properly return zeroes after TRIM".
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.