Trim (computing)

Template:Other uses2 A Trim (commonly TRIM) command 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.

Trim was introduced soon after SSDs started to become an affordable alternative to traditional hard disks. 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. Trimming enables the SSD to handle garbage collection overhead, which would otherwise significantly slow down future write operations to the involved blocks, in advance.

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. More recent SSDs will often contain internal idle/background garbage collection mechanisms that work independently of trimming; although this successfully maintains their performance even under operating systems that do not (yet) support Trim, it has the associated drawbacks of increased write amplification and wear of the flash cells.

Flash drive specific issues

Because of the way that file systems typically handle delete operations, 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. Delete operations are typically limited to flagging data blocks as "not in use" in the file system. Contrary to, for example, an overwrite operation, a delete will therefore not involve a physical write to the sectors that contain the data. Since a common SSD has no access to 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 traditional hard disks, 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 traditional hard disks this is no different than writing 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.

SSDs store data in flash memory cells that are grouped into pages, with the pages (typically 4 kB each) grouped together into blocks (typically 128 pages per block, totaling 512 kB). NAND flash memory cells can only be directly written to when they are empty. If they are considered to contain data, the contents first need to be erased before a write operation can be performed reliably. In SSDs, a write operation can be done on the page-level, but due to hardware limitations, erase commands always affect entire blocks. As a result, writing data to SSD media is very fast as long as empty pages can be used, 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: the contents of the entire block have to be stored in cache before it is effectively erased on the flash medium, then the overwritten page is modified in the cache so the cached block is up to date, and only then is the entire block (with updated page) written to the flash medium. This phenomenon is known as write amplification.

Operation

The Trim command is designed to enable the operating system to notify the SSD which pages no longer contain valid data due to erases either by the user or operating system itself. During a delete operation, the OS will both mark the sectors as free for new data and send a Trim command to the SSD to be marked as no longer valid. After that the SSD knows not to relocate data from the affected LBAs during garbage collection. This results in fewer writes to the flash, reducing write amplification and increasing drive life.

Different SSDs will act on the Trim command somewhat differently so the final performance can also be different between different SSDs.

Trim irreversibly deletes the data it affects.

Shortcomings

• When software-based disk encryption is in use, using the Trim command reveals information about which blocks are in use and which are not.
• Trim has been defined as a non-queued command by the T13 subcommittee, and consequently incurs 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 finish any operation, 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 shortcoming has been overcome in Serial ATA revision 3.1 with the introduction of the Queued Trim Command.

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.

Operating System	Supported since	Notes
DragonFly BSD	2011-05May 2011
FreeBSD	2010-078.1 - July 2010	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. ZFS obtained support with FreeBSD 9.1-STABLE. FreeBSD 10 will support trimming on ZFS as well as on software RAID configurations.
Linux	2008-12-252.6.28–25 December 2008	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. Not all filesystems make use of Trim. Among the filesystems that can issue Trim requests automatically are Ext4, Btrfs, FAT, GFS2 and XFS. However, this is disabled by default due to performance concerns, but can be enabled by setting the "discard" mount option. 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.
Mac OS X	2011-06-2310.6.8–23 June 2011	Although the AHCI block device driver gained the ability to display whether a device supports the Trim operation in 10.6.6 (10J3210), the functionality itself remained inaccessible until 10.6.8, when the Trim operation was exposed via the IOStorageFamily and filesystem (HFS+) support was added.
Microsoft Windows	2009-10NT 6.1 (Windows 7 and Windows Server 2008 R2) - October 2009	Windows 7 only supports trim for ordinary (SATA) drives and does not support this command for PCI-Express SSDs that are different type of device, even if the device itself would accept the command. It is confirmed that with native Microsoft drivers the Trim command works in AHCI and legacy IDE / ATA Mode.
OpenSolaris	2010-07July 2010
Android	2013-74.3 - 24 July 2013

RAID issues

The RST (Rapid Storage Technology Option Rom ) and drivers are only allowing Trim to pass to the controller onto the drive in Intel 7 series chipsets using driver versions post 11.2.0.0. For Intel chipsets such as the 6 series and the newest X79, a modded ( Option Rom ) must be used on these chipsets to allow Trim to be received at the drive. The exception to X79 chipset is if the Motherboard MFG has added Rom switch, this entails both the RST and RST-E Roms are inside the BIOS/UEFI, this allows the RST Rom to be used, allowing Trim to function, as Trim does not work on the RST-E Rom. Official response from Intel, Best performance is to use a driver with same version as Rom, if BIOS/UEFI has a 11.0.0.0m Option Rom, then use a 11. version driver.

As of January 2012, support for the Trim command has not been implemented in most RAID technologies. One case where it has been implemented is in post-January-2011 releases of the Linux kernel's dmraid, which implements BIOS-assisted "fake hardware RAID" support, and now passes through any Trim requests from the filesystem that sits on the RAID array. 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). For a short time in March 2010, users were led to believe that the Intel Rapid Storage Technology (RST) 9.6 drivers supported Trim in 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.

As of November 2011, Intel has indicated in the release notes for RST 11.5 Alpha that they intend to add support for Trim on RAID 0 volumes in the next version of RST. Red Hat has also recommended against using software RAID levels 1, 4, 5, and 6 on SSDs, 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) operate properly, causing the SSD to believe that all blocks other than in the spare area are in use, significantly degrading performance.

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

Enabling unsupported operating systems

Where Trim is not automatically supported by the filesystem, there are utilities which 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 (Intel, G.Skill) or as general utilities (hdparm since v9.17).

Hardware support

ATA

The TRIM command specification is being standardized as part of the AT Attachment (ATA) interface standard, led by Technical Committee T13 of the International Committee for Information Technology Standards (INCITS). TRIM is implemented under the DATA SET MANAGEMENT command (opcode 06h) the draft ACS-2 specification.

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.

There are different types of Trim defined by SATA Words returned from an ATA IDENTIFY DEVICE command.

Non-deterministic Trim: each read command to the LBA after a Trim may return different data. (SATA Word 169 bit 0)

Deterministic Trim (DRAT): all read commands to the LBA after a Trim shall return the same data, or become determinate. (SATA Word 69 bit 14)

Deterministic Read Zero after Trim (DZAT): all read commands to the LBA after a Trim shall return zero. (SATA Word 69 bit 5)

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 4k) but many drives reduce this to one to meet the Microsoft Windows Hardware Requirements for Trim, Command completion time shall not exceed 20ms or 8ms * (number of LBA range entries), whichever is greater, and shall always be less than 600ms. http://msdn.microsoft.com/en-us/library/windows/hardware/jj134356.aspx 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 number (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.(need reference to T13 ATA/ATAPI Command Set - 3 (ACS-3)) This means that for each 512 byte block of Trim ranges that a device supports, the maximum is 64 ranges of 32MB, or 2GB. If a device supports SATA Word 105 at 8 then it should be able to trim 16GB in a single Trim (DATA SET MANAGEMENT) command.

SCSI

SCSI provides UNMAP command (full analog of TRIM) and WRITE SAME (10,16) commands with unmap flag.

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").

External links

• From write() down to flash chips – an explanation on how the Trim command lets SSDs erase data not used by the filesystem
• TRIM Command White Paper - a white paper explaining the Trim command's purpose and actions