Email this Page   Printer Friendly Version

Solid State disks (SSD) are devices that use exclusively semiconductor memory components to store digital data. The memory components are the same ones used for the different types of computer memory: work memory, cache memory and embedded memory.

The two primary advantages resulting from using solid state memory components instead of mechanical devices to store data are higher ruggedness and significantly improved performance. Performance improvements result from the very fast and predictable access times associated with the semiconductor memory components and from the drastic reduction of latency and seek times. Additional benefits are the extended operating temperature range and the lower power consumption.

Most solid state disks have a HDD-compatible form factor, system interconnect and electric interconnect. This enables solid state disks to be added to the system very much like the magnetic or optical disk drives. From a system solution perspective, solid state disks can be regarded nowadays as HDD replacements. From a system architecture and functional perspective, they complement mechanical disk drives and have the capability to enhance the performance of individual mechanical drives and drive clusters.

In general, there are three classifications of solid state disks based on the type of memory components utilized:

DRAM-based Solid State Disk (D-SSD)
SRAM-based Solid State Disk (S-SSD)
Flash memory-based Solid State Disk (F-SSD)

The first two (D-SSD and S-SSD) are based on volatile memory components and need a data retention mechanism when the power supply is removed. This can be either built-in batteries, or a non-volatile mechanic or semiconductor backup. F-SSDs use Flash memory chips that are non-volatile memory components. Some solid state disks may use a combination of memory components in order to improve certain specifications.

SRAMs have the fastest read and write cycle times. They do not need a periodic refresh cycle to preserve the stored contents. S-SSD drawbacks are the volatile nature of the memory cell and its larger size, which leads to a lower storage density and higher cost per stored bit.

DRAMs have fast read and write cycle times as well. However, they need to periodically refresh their contents, which downgrades the average transfer performance. The DRAM cell size is considerably smaller than that of SRAMs and the storage density accordingly higher. The DRAM cost per stored bit is currently the lowest.

Flash memory has a read access time that is comparable to that of DRAMs, however, the write cycle time is significantly longer. In addition, the write operation needs to be performed in conjunction with an erase operation, simultaneously for a group of locations, called page or sector. This complicates the write process and the associated circuitry. Flash memory cell size is smaller than that of DRAM and can be reduced at a faster pace. This creates the premise for lower storage cost. From a system perspective, the main advantage of F-SSD is the non-volatile nature of its storage cell, which puts it at par with magnetic and optical disk drives.

[ Top ]