Solid-state drive

An SSD in standard 2.5-inch (64 mm) form-factor

DDR SDRAM based SSD

PCI attached IO Accelerator SSD

PCI-E / DRAM / NAND based SSD

A solid-state drive (SSD) is a data storage device that uses solid-state memory to store persistent data. SSDs are distinguished from traditional hard disk drives (HDDs), which are electromechanical devices containing spinning disks and movable read/write heads. SSDs, in contrast, use microchips, and contain no moving parts. Compared to traditional HDDs, SSDs are typically less susceptible to physical shock, quieter, and have lower access time and latency. SSDs use the same interface as hard disk drives, thus easily replacing them in most applications.^[1]

As of 2010, most SSDs use NAND-based flash memory, which retains memory even without power. SSDs using volatile random-access memory (RAM) also exist for situations which require even faster access, but do not necessarily need data persistence after power loss, or use batteries to back up the data after power is removed.^[1]

A hybrid drive combines the features of an HDD and an SSD in one unit.

Development and history

Early SSDs using RAM and similar technology

The origins of SSDs came from the 1950s using two similar technologies, core memory and Card Capacitor Read Only Store (CCROS).^[2][3] These auxiliary memory units, as they were called at the time, emerged during the era of vacuum tube computers. But with the introduction of cheaper drum storage units, their use was discontinued.^[4] Later, in the 1970s and 1980s, SSDs were implemented in semiconductor memory for early supercomputers of IBM, Amdahl and Cray;^[5] however, the prohibitively high price of the built-to-order SSDs made them quite seldom used.

In 1978, Texas Memory Systems introduced a 16 kilobyte (KB) RAM solid-state drive to be used by oil companies for seismic data acquisition.^[6] The following year, StorageTek developed the first modern type of solid-state drive.^[7] The Sharp PC-5000, introduced in 1983, used 128 kilobyte solid-state storage cartridges, containing bubble memory.^[8] In September 1986, Santa Clara Systems introduced BatRam, 4 megabyte (MB) mass storage system expandable to 20 MBs using 4 MB memory modules. The package included a rechargeable battery to preserve the memory chip contents when the array was not powered.^[9] 1987 saw the entry of EMC Corporation into the SSD market, with drives introduced for the mini-computer market. However, EMC exited the business soon after.^[10]

Flash-based SSDs

In 1995, M-Systems introduced flash-based solid-state drives.^[11] Since then, SSDs have been used successfully as HDD replacements by the military and aerospace industries, as well as other mission-critical applications. These applications require the exceptional mean time between failures (MTBF) rates that solid-state drives achieve, by virtue of their ability to withstand extreme shock, vibration and temperature ranges.^[12]

BiTMICRO made a number of introductions and announcements in 1999 around flash-based SSDs including an 18 gigabyte 3.5" SSD.^[13] Fusion-io announced a PCIe-based SSD with 100,000 Input/Output Operations Per Second (IOPS) of performance in a single card with capacities up to 320 gigabytes in 2007.^[14] At Cebit 2009, OCZ demonstrated a 1 terabyte (TB) flash SSD using a PCI Express x8 interface. It achieves a maximum write speed of 654 megabytes per second (MB/s) and maximum read speed of 712 MB/s.^[15] In December 2009, Micron Technology announced the world's first SSD using a 6 gigabits per second (Gbps) SATA interface.^[16]

Enterprise Flash Drives

Enterprise Flash Drives (EFDs) are designed for applications requiring high I/O performance (IOPS), reliability, and energy efficiency. In most cases an EFD is an SSD with a higher set of specifications compared to SSDs which would typically be used in notebook computers. The term was first used by EMC in January 2008, to help them identify SSD manufacturers who would provide products meeting these higher standards.^[17] There are no standards bodies who control the definition of EFDs, so any SSD manufacturer may claim to produce EFDs when they may not actually meet the requirements. Likewise there may be other SSD manufacturers that meet the EFD requirements without being called EFDs.^[18]

Architecture and function

The primary storage component in an SSD has been DRAM volatile memory since they were first developed, but since 2009 it is more commonly NAND flash non-volatile memory.^[1][19]

Flash drives

Most SSD manufacturers use non-volatile flash memory to create more rugged and compact devices for the consumer market. These flash memory-based SSDs, also known as flash drives, do not require batteries.^[20] They are often packaged in standard disk drive form factors (1.8-, 2.5-, and 3.5-inch). In addition, non-volatility allows flash SSDs to retain memory even during sudden power outages, ensuring data persistence. Flash memory SSDs are slower than DRAM SSDs and some designs are slower than even traditional HDDs on large files, but flash SSDs have no moving parts and thus seek times and other delays inherent in conventional electro-mechanical disks are negligible.

SSD Components:

• Controller: Includes the electronics that bridge the NAND memory components to the SSD I/O interfaces. The controller is an embedded processor that executes firmware-level software and is one of the most important factors of SSD performance.^[21]
• Cache: A flash-based SSD uses a small amount of DRAM as a cache, similar to the cache in Hard disk drives. A directory of block placement and wear leveling data is also kept in the cache while the drive is operating.
• Energy storage: Another component in higher performing SSDs is a capacitor or some form of batteries. These are necessary to maintain data integrity such that the data in the cache can be flushed to the drive when power is dropped; some may even hold power long enough to maintain data in the cache until power is resumed.

The performance of the SSD can scale with the number of parallel NAND flash chips used in the device. A single NAND chip is relatively slow, due to narrow (8/16 bit) asynchronous IO interface, and additional high latency of basic IO operations (typical for SLC NAND - ~25 μs to fetch a 4K page from the array to the IO buffer on a read, ~250 μs to commit a 4K page from the IO buffer to the array on a write, ~2 ms to erase a 256 KB block). When multiple NAND devices operate in parallel inside an SSD, the bandwidth scales, and the high latencies can be hidden, as long as enough outstanding operations are pending and the load is evenly distributed between devices.

Micron/Intel SSD made faster flash drives by implementing data striping (similar to RAID 0) and interleaving. This allowed creation of ultra-fast SSDs with 250 MB/s effective read/write.^[22]

SLC versus MLC

Lower priced drives usually use multi-level cell (MLC) flash memory, which is slower and less reliable than single-level cell (SLC) flash memory.^[23][24] This can be mitigated or even reversed by the internal design structure of the SSD, such as interleaving, changes to writing algorithms^[24], and more excess capacity for the wear-leveling algorithms to work with.

DRAM based drive

SSDs based on volatile memory such as DRAM are characterized by ultrafast data access, generally less than 10 microseconds, and are used primarily to accelerate applications that would otherwise be held back by the latency of Flash SSDs or traditional HDDs. DRAM-based SSDs usually incorporate either an internal battery or an external AC/DC adapter and backup storage systems to ensure data persistence while no power is being supplied to the drive from external sources. If power is lost, the battery provides power while all information is copied from random access memory (RAM) to back-up storage. When the power is restored, the information is copied back to the RAM from the back-up storage, and the SSD resumes normal operation (similar to the hibernate function used in modern operating systems).^[20][25]

These types of SSD are usually fitted with the same type of DRAM modules used in regular PCs and servers, allowing them to be swapped out and replaced with larger modules.

A Remote Indirect Memory Access Disk (RIndMA Disk) uses a secondary computer with a fast network or (direct) Infiniband connection to act like a RAM-based SSD, but the new faster Flash memory based SSDs already available in 2009 are making this option not as cost effective.^[26]

DRAM based solid-state drives are especially useful on computers that already have the maximum amount of supported RAM. For example, some computer systems built on the x86-32 architecture can effectively be extended beyond the 4 gigabyte (GB) limit by placing the paging file or swap file on a DRAM based SSD. Since the paging file is used when there is no space left in main memory and at the expense of speed, placing the paging file on a DRAM based SSD will allow the computer to take advantage of the speed of the SSD when it requires more space in main memory and resorts to using the paging file.

Form factor

The size and shape of any device is largely driven by the size and shape of the components used to make that device. Traditional HDDs and optical drives are designed around the rotating platter or optical disc along with the spindle motor inside. If an SSD is made up of various interconnected integrated circuits (ICs) and an interface connector, then its shape could be virtually anything imaginable because it is no longer limited to the shape of rotating media drives. Some solid state storage solutions come in a larger chassis that may even be a rack-mount form factor with numerous SSDs inside. They would all connect to a common bus inside the chassis and connect outside the box with a single connector.^[1]

Standard HDD form factors

The benefit of using a current HDD form factor would be to take advantage of the extensive infrastructure already in place to mount and connect the drives to the host system.^[1][27] These traditional form factors are known by the size of the rotating media, e.g., 5.25", 3.5", 2.5", 1.8", not by the dimensions of the drive casing.^[28]

Non-HDD form factors

Form factors which were more common to memory modules are now being used by SSDs to take advantage of their flexibility in laying out the components. Some of these include PCIe, mini PCIe, mini-DIMM, MO-297, and many more.^[29] At least one manufacturer, InnoDisk, is producing a drive that sits directly on the SATA connector on the motherboard without any other support or mechanical mount.^[30] Some SSDs are based on the PCIe form factor and connect both the data interface and power through the PCIe connector to the host. These drives can use either direct PCIe Flash controllers^[31] or a PCIe-to-SATA bridge device which then connects to SATA Flash controller(s).^[32]

Comparison of SSD with hard disk drives

Making a comparison between SSDs and ordinary (spinning) Hard disk drives is difficult. Traditional HDD bench-marks are focused on finding the performance aspects where they are weak, such as rotational latency time and seek time. As SSDs do not spin, or seek, they may show huge superiority in such tests. However, SSDs have challenges with mixed reads and writes, and their performance may degrade over time. SSD testing must start from the (in use) full disk, as the new and empty disk may have much better write performance than it would show after years of use.^[33].

A comparison (with benchmarks) of SSDs, Secure Digital High Capacity (SDHC) drives, and hard disk drives (HDDs) is given in the reference.^[34]

The disassembled components of a hard disk drive (left) and of the PCB and components of a solid-state drive (right)

Comparisons reflect typical characteristics, and may not hold for a specific device.

Advantages

• Faster start-up because no spin-up is required.
• Fast random access because there is no "seeking" motion as is required with rotating disk platters and the read/write head and head-actuator mechanism ^[35]
  • Low read latency times for RAM drives.^[36] In applications where hard disk seeks are the limiting factor, this results in faster boot and application launch times (see Amdahl's law).^[37]
  • Consistent read performance because physical location of data is irrelevant for SSDs.^[38]
  • File fragmentation has negligible effect , again because data access degradation due to fragmentation is primarily due to much greater disk head seek activity, as data reads or writes are spread across many different locations on the disk; SSDs have no heads and thus no delays due to head motion (seeking).
• Silent operation due to the lack of moving parts.
• SSDs typically have lower power consumption than HDDs and, as a consequence, laptop manufacturers are starting to embrace them as optional replacements to standard HDDs.
• High mechanical reliability, as the lack of moving parts almost eliminates the risk of mechanical failure.
• Ability to endure extreme shock, high altitude, vibration and extremes of temperature.^[39][40] This makes SSDs useful for laptops, mobile computers, and devices that operate in extreme conditions (flash).^[37]
• Immune to magnets.
• For low-capacity SSDs, lower weight and size: although size and weight per unit storage are still better for traditional hard drives, and microdrives allow up to 20 GB storage in a CompactFlash form-factor. As of 2008, SSDs up to 256 GB are lighter than hard drives of the same capacity.^[39]
• Failures occur less frequently while writing/erasing data, which means there is a lower chance of irrecoverable data damage.^[41]
• Defragmenting the SSD is unnecessary. Since SSDs are random access by nature and can perform parallel reads on multiple sections of the drive (as opposed to a HDD, which requires seek time for each fragment, assuming a single head assembly), a certain degree of fragmentation is actually better for reads, and wear leveling intrinsically induces fragmentation.^[42] In fact, defragmenting a SSD is harmful since it adds wear to the SSD for no benefit.^[43]
• Can also be configured to smaller form factors and reduced weight.^[44]

Disadvantages

• Flash-memory drives have limited lifetimes and will often wear out after 1,000,000 to 2,000,000 P/E cycles (1,000 to 10,000 per cell) for MLC, and up to 5,000,000 P/E cycles (100,000 per cell) for SLC.^{[45][46][47][48]} Special file systems or firmware designs can mitigate this problem by spreading writes over the entire device, called wear leveling.^[49]
• Wear leveling used on flash-based SSDs has security implications. For example, encryption of existing unencrypted data on flash-based SSDs cannot be performed securely because wear leveling causes new encrypted drive sectors to be written to a physical location different from their original location—data remains unencrypted in the original physical location. It is also impossible to securely wipe files by overwriting their content on flash-based SSDs.^[50] However, drives that support the ATA command TRIM allow for secure file deletion, as deleted blocks are cleaned in the background before writes^[51].
• As of early 2010, SSDs are still more expensive per gigabyte than hard drives. Whereas hard drives are around US$0.10 per gigabyte for 3.5", or US$0.20 for 2.5", a typical flash drive is US$2 per gigabyte.
• The capacity of SSDs is currently lower than that of hard drives. However, flash SSD capacity is predicted to increase rapidly, with drives of 1 TB already released for enterprise and industrial applications.^{[52][53][54][55][56]}
• Older SSDs have asymmetric read vs. write performance that can cause problems with certain functions where the read and write operations are expected to be completed in a similar timeframe.
• SSD write performance is significantly impacted by the availability of free, programmable blocks. Previously written data blocks that are no longer in use can be reclaimed by TRIM; however, even with TRIM, fewer free, programmable blocks translates into reduced performance.^[57]
• As a result of wear leveling and write combining, the performance of SSDs degrades with use. However, most modern SSDs now support the TRIM command and thus return the SSD back to its factory performance when using OSes that support it like Windows 7.^[58][59]
• SATA-based SSDs generally exhibit much slower write speeds. As erase blocks on flash-based SSDs generally are quite large, e.g., 0.5 - 1 megabyte),^[23] they are far slower than conventional disks during small writes (write amplification effect) and can suffer from write fragmentation.^[60] Modern PCIe SSDs however have much faster write speeds than previously available.
• DRAM-based SSDs (but not flash-based SSDs) require more power than hard disks, when operating; they still use power when the computer is turned off, while HDDs do not.^[61][62]

Commercialization

Cost and capacity

Until recently, flash-based, solid-state drives were too costly for widespread use in mobile computing. As flash manufacturers transition from NOR flash to single-level cell (SLC) NAND flash and most recently to multi-level cell (MLC) NAND flash to maximize silicon die usage and reduce associated costs, "solid-state disks" are now being more accurately renamed "solid-state drives" – they have no disks but function as drives – for mobile computing in the enterprise and consumer electronics space. This technological trend is accompanied by an annual 50% decline in raw flash material costs, while capacities continue to double at the same rate. As a result, flash-based solid-state drives are becoming increasingly popular in markets such as notebook PCs and sub-notebooks for enterprises, Ultra-Mobile PCs (UMPC), and Tablet PCs for the healthcare and consumer electronics sectors. Major PC companies have now started to offer such technology.

Availability

Solid-state drive (SSD) technology has been marketed to the military and niche industrial markets since the mid-1990s .

CompactFlash card used as SSD

Along with the emerging enterprise market, SSDs have been appearing in ultra-mobile PCs and a few lightweight laptop systems, adding significantly to the price of the laptop, depending on the capacity, form factor and transfer speeds. As of 2008, some manufacturers have begun shipping affordable, fast, energy-efficient drives priced at $350 to computer manufacturers. For low-end applications, a USB flash drive may be obtained for $10 to $100 or so, depending on capacity, or a CompactFlash card may be paired with a CF-to-IDE or CF-to-SATA converter at a similar cost. Either of these requires that write-cycle endurance issues be managed, either by not storing frequently written files on the drive, or by using a Flash file system. Standard CompactFlash cards usually have write speeds of 7 to 15 MB/s while the more expensive upmarket cards claim speeds of up to 40 MB/s.

One of the first mainstream releases of SSD was the XO Laptop, built as part of the 'One Laptop Per Child' project. Mass production of these computers, built for children in developing countries, began in December 2007. These machines use 1024 MiB SLC NAND flash as primary storage which is considered more suitable for the harsher than normal conditions in which they are expected to be used. Dell began shipping ultra-portable laptops with SanDisk SSDs on April 26, 2007.^[63] Asus released the Eee PC subnotebook on October 16, 2007, and after a successful commercial start in 2007, it was expected to ship several million PCs in 2008, with 2, 4 or 8 gigabytes of flash memory.^[64] On January 31, 2008, Apple Inc. released the MacBook Air, a thin laptop with optional 64 GB SSD. The Apple store cost was $999 more for this option, as compared to that of an 80 GB 4200 rpm Hard Disk Drive.^[65] Another option—Lenovo ThinkPad X300 with a 64 gigabyte SSD—was announced by Lenovo in February 2008,^[66] and is, as of 2008, available to consumers in some countries. On August 26, 2008, Lenovo released ThinkPad X301 with 128GB SSD option which adds approximately $200 US.

The Mtron SSD

In 2008, low end netbooks appeared with SSDs. In 2009, SSDs began to appear in laptops.^[63][65]

On January 14, 2008, EMC became the first enterprise storage vendor to ship flash-based SSDs into its product portfolio.^[67]

In late 2008, Sun released the Sun Storage 7000 Unified Storage Systems (codenamed Amber Road), which use both solid state drives and conventional hard drives to take advantage of the speed offered by SSDs and the economy and capacity offered by conventional hard disks.^[68]

Dell began to offer optional 256 GB solid state drives on select notebook models in January 2009.

In May 2009, Toshiba launched a laptop with a 512 GB SSD^[69][70].

As of April 13, 2010, Apple's MacBook Pro line carry optional solid state drives of up to 512 GB at an additional cost.

Quality and performance

SSD is a rapidly developing technology. A January 2009 review of the market by technology reviewer Tom's Hardware concluded that comparatively few of the tested devices showed acceptable I/O performance, with several disappointments,^[71] and that Intel (who make their own SSD chipset) still produces the best performing SSD as of this time; a view also echoed by Anandtech.^[72] In particular, operations that require many small writes, such as log files, are particularly badly affected on some devices, potentially causing the entire host system to freeze for periods of up to one second at a time.^[73]

According to Anandtech, this is due to controller chip design issues with a widely used set of components, and at least partly arises because most manufacturers are memory manufacturers only, rather than full microchip design and fabrication businesses — they often rebrand others' products,^[74] inadvertently replicating their problems.^[75] Of the other manufacturers in the market, Memoright, Mtron, OCZ, Samsung and Soliware were also named positively for at least some areas of testing.

The overall conclusion by Tom's Hardware as of early 2009 was that "none of the [non-Intel] drives were really impressive. They all have significant weaknesses: usually either low I/O performance, poor write throughput or unacceptable power consumption".^[71]

Performance of flash SSDs are difficult to benchmark. In a test done by Xssist, using IOmeter, 4 KB RANDOM 70/30 RW, queue depth 4, the IOPS delivered by the Intel X25-E 64 GB G1 started around 10,000 IOPs, and dropped sharply after 8 minutes to 4,000 IOPS, and continued to decrease gradually for the next 42 minutes. IOPS vary between 3,000 to 4,000 from around the 50th minutes onwards for the rest of the 8+ hours test run.^[76]

OCZ has recently unveiled OCZ Vertex 2 Pro which is currently the fastest MLC SSD with a SandForce controller onboard performing more or less as the Intel X25-E series SSDs.^[77]

An April 2010 test of seven SSDs by Logan Harbaugh, which appeared in Network World, identified a performance problem with consumer grade SSDs. Dubbed the "write cliff" effect, consumer grade drives showed dramatic variations in response times under sustained write conditions. This dropoff occurred once the drive was filled for the first time and the drive's internal garbage collection and wear-leveling routines kicked in.^[78]

This only affected write performance with consumer grade drives. Enterprise grade drives avoid this problem by overprovisioning, and by employing wear-leveling algorithms that only move data around when the drives are not being heavily utilized.^[78]

Applications

Until 2009, SSDs were mainly used in mission critical applications where the speed of the storage system needed to be as fast as possible. Since Flash memory has become a common component of SSDs, the falling prices and increased densities have made it more financially attractive for many other applications. Consider the organizations that can benefit from faster access of system data like equity trading companies, telecommunication corporations, and video streaming and editing firms. The list of applications which could benefit from faster storage is vast. Any company can assess the ROI from adding SSDs to their own applications to best understand if that will be cost effective for them.^[1]

Flash-based Solid-state drives can be used to create network appliances from general-purpose PC hardware. A write protected flash drive containing the operating system and application software can substitute for larger, less reliable disk drives or CD-ROMs. Appliances built this way can provide an inexpensive alternative to expensive router and firewall hardware.

SSD optimized file systems

There are a number of file systems which are optimized for solid-state drives. Some of the more popular or notable are listed below.

Microsoft Windows

Versions of Windows prior to Windows 7 are optimized for hard disk drives rather than SSDs.^[79][80] Windows Vista includes ReadyBoost to exploit characteristics of USB-connected flash devices, but for SSDs it only improves the partition alignment to prevent read-modify-write operations because the SSD is typically aligned on 4 KB sectors and the OS is based on 512 byte sectors and they are not aligned.^[81] The proper alignment really does not help the SSD's endurance over the life of the drive. Some Vista operations, if not disabled, can shorten the life of the SSD. Disk defragmentation should be disabled because the location of the file components on an SSD don't significantly impact its performance, but moving the files to make them contiguous using the Windows Defrag routine will cause write wear on the limited number of P/E cycles on the SSD. The page file should also be disabled because the constant updates to the file will cause unnecessary additional wear on the SSD that is not offset by enough performance gain. The Superfetch feature will not materially change the performance of the system and causes additional overhead on the system and SSD, although it does not cause wear.^[82]

Windows 7 is optimized for SSDs^[83][84] as well as for hard disks. The OS looks for the presence of an SSD and operates differently with that drive. If an SSD is present, Windows 7 will disable disk defragmentation, Superfetch, ReadyBoost, and other boot-time and application prefetching operations. It also includes support for the TRIM command to reduce garbage collection of data which the OS has already determined is no longer valid, but the SSD is not aware of that.^[85]

ZFS

Solaris, as of 10u6 (released in October 2008), and recent versions of OpenSolaris and Solaris Express Community Edition on which OpenSolaris is based, can use SSDs as a performance booster for ZFS. There are two available modes—using an SSD for the ZFS Intent Log (ZIL), which is used every time a synchronous write to the disk occurs (a low write-latency SSD should be used for the ZIL), or for the L2ARC (Level 2 Adaptive Replacement Cache), which is used to cache data for reading (the L2ARC is infrequently written, and low write-latency SSD is thus not needed). When used either alone or in combination, large increases in performance are generally seen. ^[86]

Standardization organizations

The following are noted standardization organizations and bodies that work to create standards for solid-state drives (and other computer storage devices). It also includes organizations who promote the use of solid-state drives. This is not necessarily an exhaustive list.

See also

• Hybrid drive
• Computer storage
• Flash file system
• List of solid-state drives
• TRIM (SSD command)
• Write amplification

References

External links

• StorageReview.com SSD Guide
• JEDEC Continues SSD Standardization Efforts
• Understanding SSDs and New Drives from OCZ
• SSDs versus laptop HDDs and upgrade experiences