* Price - As of early 2007, flash memory prices are still considerably higher per gigabyte than those of comparable conventional hard drives - around US$8 per GiB compared to about US$0.25 for mechanical drives.
* Price - As of early 2007, flash memory prices are still considerably higher per gigabyte than those of comparable conventional hard drives - around US$8 per GiB compared to about US$0.25 for mechanical drives.
* Capacity - The capacity of SSDs tends to be significantly smaller than the capacity of HDDs.
* Lower recoverability - After mechanical failure the data is completely lost as the cell is destroyed, while if normal HDD suffers mechanical failure the data is often [[Data Recovery|recoverable]] using expert help. Subsequent investigations in to this field, however, have found that data can be recovered from SSD memory.
* Lower recoverability - After mechanical failure the data is completely lost as the cell is destroyed, while if normal HDD suffers mechanical failure the data is often [[Data Recovery|recoverable]] using expert help. Subsequent investigations in to this field, however, have found that data can be recovered from SSD memory.
* Vulnerability to certain types of effects, including abrupt power loss (especially DRAM based SSDs), magnetic fields and electric/static charges compared to normal HDDs (which store the data inside a [[Faraday cage]]).
* Vulnerability to certain types of effects, including abrupt power loss (especially DRAM based SSDs), magnetic fields and electric/static charges compared to normal HDDs (which store the data inside a [[Faraday cage]]).
Revision as of 18:32, 11 September 2007
This article refers to both flash and DRAM-based solid state drives. For other flash-based solid state storage, see flash drive and USB flash drive.
With no moving parts, a solid state drive largely eliminates seek time, latency and other electro-mechanical delays and failures associated with a conventional hard disk drive.
While SSD is not technically a disk, the term solid state disk emphasises the typical usage as an alternative for a disk drive.
SSDs based on volatile memory such as SDRAM are categorized by fast data access, less than 0.01 milliseconds (over 250 times faster than the fastest hard drives in 2004) and are used primarily to accelerate applications that would otherwise be held back by the latency of disk drives.
DRAM-based SSDs typically incorporate internal battery and backup disk systems to ensure data persistence. If power is lost for whatever reason, the battery would keep the unit powered long enough to copy all data from random access memory (RAM) to backup disk. Upon the restoration of power, data is copied back from backup disk to RAM and the SSD resumes normal operation.
However, most SSD manufacturers use nonvolatile flash memory to create more rugged and compact alternatives to DRAM-based SSDs. These flash memory-based SSDs, also known as flash drives, do not require batteries, allowing makers to replicate standard disk drive form factors (1.8-inch, 2.5-inch, and 3.5-inch). In addition, nonvolatility allows flash SSDs to retain memory even during sudden power outages, ensuring data retrievability. Just like DRAM SSDs, flash SSDs are extremely fast since these devices have no moving parts, eliminating seek time, latency and other electro-mechanical delays inherent in conventional disk drives. (Though flash SSDs are significantly slower than DRAM SSDs).
Solid state drives are especially useful on a computer which already has the maximum amount of RAM. For example, some x86 architectures have a 4 GB limit, but this can effectively be extended by putting the swap file on a SSD. These SSD do not provide as fast storage as main RAM because of the bandwidth bottleneck of the bus they connect to, but would still provide a performance increase over placing the swap file on a traditional hard disk drive.
File:Open HDD and SSD.JPGOpen casing of 2.5” traditional hard disk drive (left) and solid state drive (center), a drop-in replacement for hard disk drives for mobile computing.
DRAM based SSDs may also work like a buffer cache mechanism. Whenever data is written to memory, the corresponding block in memory is marked as dirty and all dirty blocks can be flushed to the actual hard drive based on the following two strategies: 1. Time (e.g. every 10 seconds, flush all dirty data), 2. Threshold (when the ratio of dirty data to SSD size exceeds some predetermined value, flush the dirty data).
Advantages
Faster startup - Since no spin-up required.
Far faster than conventional disks on random I/O
Extremely low read and write latency (seek) times, roughly 15 times faster than the best current mechanical disks.
Faster boot and application launch time when hard disk seeks are the limiting factor. See Amdahl's law.
In some cases, somewhat longer lifetime - Flash storage typically has a data lifetime on the order of 10 years before degradation. If data is periodically refreshed, it can store data indefinitely. Flash drives have limited endurance (typically, 100,000-300,000 write cycles), which, if a single block is written once per second, leads to failure in a few days at most. However, all flash drives employ a technique known as wear levelling, where writes are smoothly distibuted over all blocks. This means that if one write occurs per second, and n is the number of writes before failure and m is the number of blocks on the disk, failure no longer occurs in n seconds, but in (n*m) seconds. Given that blocks are typically on the order of 1kb and an 8 GB disk will have 8,192 blocks, this gives about 9,500 days before failure; remember also this is with one write per second for that entire time [1]. In consumer level devices you can expect the drive's data storage component to last roughly 10 years in normal use [4]
The disassembled components of a hard disk drive (left) and of the PCB and components of a solid state drive (right).
However, it should be noted that certain SCSI hard drives have MTBF of 1.5 million hours (~175 years) and normal SATA harddrives have MTBF of 500,000 hours (~57 years). [5] The actual expected time-to-failure is typically several years, with manufacturers giving warranties of up to and around 5 years in consumer level products [6]. The high MTBFs are not calculated based on expected hard drive time to failure, but rather based on failure rates of hard drives without much wear.
Few to no mechanical parts
For small drives, lower power consumption and heat production.
For small hard drives, no noise - Lack of mechanical parts makes the SSD completely silent (many high-end SSDs include cooling fans).
Better mechanical reliability - Lack of mechanical parts results in less wear and tear. High level of ability to endure extreme shock, high altitude, vibration and temperatures[citation needed], which apply to laptops and other mobile devices, or when transported.
Security - allowing a very quick "wipe" of all data stored. [citation needed]
Relatively deterministic performance [7] - unlike mechanical hard drives, performance of SSDs is almost constant and deterministic across the entire storage. "Seek" time can be constant, and performance does not deteriorate as the media fills up (See: Fragmentation).
However, this is not always the case, as explained below.
Flash memory is organised in blocks which can be erased, written or read, but only as whole blocks.
The access time is the same for each block.
If one or more blocks are used as Access Unit (AU), fragmentation has no harmful effect on access speed.
However, for high capacity flash memories the AU would be too big, causing a lot of wasted bytes due to unused space within allocated AU's.
Hence, in these cases each block is split up in a number of AU's.
Initially AU's will be used sequentially within blocks.
So a file with a size of N blocks will use no more than N+2 blocks, the first and last block only partially.
However, after some time a situation will occur where no block is available of which all AU's are free, so that one or more extra blocks are needed
The result is that said file will need more than N+2 blocks and accessing more blocks takes more time.
In the worst case only one AE per block is free and said file will need S*N blocks, where S is the number of AE's per block.
The conclusion is that, with bigger flash memory, fragmentation has a detoriating effect on access time.
For very low-capacity drives, lower weight and size. Size and weight per unit storage are still better for traditional hard drives, and microdrives allow up to 20GB storage in a CompactFlash 42.8×36.4×5 mm (1.7×1.4×.2 in) form factor.
Disadvantages
Flash based SSDs also have several disadvantages:
Price - As of early 2007, flash memory prices are still considerably higher per gigabyte than those of comparable conventional hard drives - around US$8 per GiB compared to about US$0.25 for mechanical drives.
Lower recoverability - After mechanical failure the data is completely lost as the cell is destroyed, while if normal HDD suffers mechanical failure the data is often recoverable using expert help. Subsequent investigations in to this field, however, have found that data can be recovered from SSD memory.
Vulnerability to certain types of effects, including abrupt power loss (especially DRAM based SSDs), magnetic fields and electric/static charges compared to normal HDDs (which store the data inside a Faraday cage).
Somewhat slower than conventional disks on sequential I/O, the latest perpendicular hard disks doing about 150 Mebibytes/sec read, with the latest SSDs doing about 120 Mebibytes/sec read.[2]
Limited write cycles. Typical Flash storage will typically wear out after 100,000-300,000 write cycles, while high endurance Flash storage is often marketed with endurance of 1-5 million write cycles (many log files, file allocation tables, and other commonly used parts of the file system exceed this over the lifetime of a computer). Special file systems or firmware designs can mitigate this problem by spreading writes over the entire device, rather than rewriting files in place. [3]
Slow random write speeds - as erase blocks on SSDs generally are quite large, they're far slower than conventional disks for random writes [4]
History of the solid state drive
This article needs attention from an expert on the subject. Please add a reason or a talk parameter to this template to explain the issue with the article. When placing this tag, consider associating this request with a WikiProject.
In the mid 1980s a company named Santa Clara Systems introduced a product named BatRam which consisted of an array of 1 megabit DIP RAM Chips and a custom controller card that emulated a hard drive. The package included a rechargeable battery to preserve the memory chip contents when the power was off.
msystems introduced flash-based solid state drives in 1995 (SanDisk completed acquisition of msystem in November 2006). Since then, they have been used successfully as hard disk drive replacements by the military and aerospace industries, as well as other mission-critical applications that require the exceptional mean time between failure (MTBF) rates that solid state drives achieve based on their ability to withstand extreme shock, vibration and temperature ranges.
Commercialization of solid state drives
Cost and capacity considerations
Until recently, solid state disks were too costly for 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 technology 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.
Availability
SSD pioneers such as BiTMICRO Networks and MemTech have marketed solid state disks to the military and industrial markets since the mid-1990s. Afterwards, the enterprise sector also realized the benefits of using SSDs as cache for storage networks and began deploying SSDs in their systems.
SSDs have been appearing in ultra mobile PCs and a few light weight laptop systems, adding a US$ 200 to $800 premium to the systems, depending on the capacity, form factor and transfer speeds. Only a handful of companies offer large (64 GB or larger) SSD drives with write speeds adequate for replacing traditional drives, but these drives are available in limited quantities and are very expensive. Already Sandisk has begun shipping an affordable, fast, energy efficient drive priced at $350 to computer manufacturers. For low-end applications, a USB memory stick may be used as a Flash hard drive for around $10-$100, 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.
Mtron announces the fastest flash memory solid state disk, performing 100MB/s Read, 80MB/s Write, 72,000 Max IOPS.[December 2005] [6]File:IMG 0085.jpg-m.jpgThe Mtron SSD Read: 100MB/s Write: 80MB/s, 76,000 sequential / 16,000 random Max IOPS with less than 0.1ms access time
Hyperdrive (storage) release the rev.4 designed to use 8 standard DDR ECC Registered memory modules on a native SATA and IDE interface. February 2007[7]
Sandisk released a 32GB 2.5-inch solid state drive on March 132007. The SSD SATA 5000 is being sold to computer manufacturers for $350. Sandisk has also released a 32GB 1.8-inch solid state drive on January 42007.
TaiwaneseA-DATA introduced at the Las VegasCES 2007 SSD drives at capacities of 32GB, 64GB (1.8" model) and 128GB (2.5" model).[14] It is expected to be commercially available by mid-2007.[15]
SimpleTech has announced a 64GB SSD that is only 9.5mm thick, half the size of competing SSDs. On April 18, 2007 SimpleTech announced 256GB capacity enterprise level drives available immediately and 512GB capacity drives available late 2007.[16][17]
+Flash+Card+Model+MCAQE32G5APP-0XA00+-+OEM){kind=link}