Solid State Drive (SSD)
A Solid State Drive (SSD), in its simplest form, is a drive (disk) emulator, that uses solid state devices (memory) to store the data instead of using rotating magnetic platters. The memory used today could be volatile but it is backed up by a battery-based power system, or it could be flash RAM, that does not loose the data when power is removed.
Believe it or not, this is not a new technology. Way back in the mid 1980’s, when I was working for an automatic test equipment (ATE) company, they designed and made a RAM Disk for a few select customers that needed to have faster load times of the data patterns that were used to test integrated circuits. This was because the hard disks of that day were very slow when it comes to access time, usually in the hundreds of milliseconds.
Pros and Cons of Solid State Drives
The major benefit of solid state drives is speed. This is because these drives eliminate the physical pieces of a hard disk that cause access delays. When you read or write a block of data, the platter has to rotate into a position where that data block can be accessed, which takes time. The actuator arm may also have to move the head across the platter, and that takes time too. So any of these physical movements that can be removed will speed up the system.
There are a few other things that make solid state drives a good thing. Since there is no physical pieces to move around, solid state drives are less susceptible to vibration in a computer system or storage array. They are also less susceptible to changes in temperature. A standard disk drive has to recalibrate internally whenever there is a temperature change of just a few percentage points. That is no longer needed with SSD.
Speaking of power, solid state drives require less power than conventional disks, therefore making them part of the “Green Computing” or “Green Storage” movement happening today.
Now for the cons of SSD. Cost is the biggest obstacle of using solid state disks. But as the market evolves and they become more available to the general public, the price should come down. Capacity is the other weak point of solid state drives. To make a really big, really fast memory system is not that hard, but to make it fit into a form factor similar to what it is replacing, such as a 2.5 or 3.5 inch hard disk, then it becomes a challenge.
Incoming search terms for the article:
- solid state drive training
- solid state san drive
- top 10 storage area network solid state disk
- storage area network solid state drive
- storage area nertwork solid state memory drives
- ssd disk san storage
- Solid-state Storage area network
- solid-state drive for SAN
- solid state storage training
- arm ssd disc driver emulator
SCSI
SCSI stands for Small Computer System Interface, and was based on SASI (Shugart Associates System Interface), which was developed and released back in the early 1980s. The use of the term “Small” is interesting, as that was the initial target audience for this system, but the protocol is now used in very large (enterprise class) storage systems around the world.
The original SCSI systems were based on a parallel set of wires (cable) that had separate control lines and data lines, and a few other conductors for termination power, Grounds (shields or “return lines” really) and of course data lines. The older SCSI systems would allow data transfers at a speed of up to 5 MB/s (Mega-Bytes per second).
Over time, speeds were increased, along with the number of bits that could be carried across the interface. What is now known as a “wide” SCSI bus carries two bytes of data at a top speed of up to 320 MB/s. There was talk in the technical committee that defined parameters to allow the interface to double its speed (up to 640 MB/s), but about that same time the serial interfaces for carrying data traffic (Fibre Channel mainly) were becoming more available for the same or lower cost and higher reliability.
The basic use of SCSI is to move “blocks” of data (vs. files) using a simple protocol. The protocol consists of an “Initiator” sending a command (such as Read or Write) to a “Target” device (disk, tape, etc.) and then transferring the data between these two entities, depending on the command itself. At the end of the command, a status is transmitted back to the Initiator to let it know if the command worked or not, from the Target’s point of view.
All of this makes up an “intelligent” system where these two entities have different roles to play, and the Target is expected to know where the data can be found, based on a Logical Block Address (LBA).
Most commands contain another parameter that tells the device “how much” data is going to be moving. For blocks, this is called “transfer length”.
SCSI is still around today, but mostly in the form of the Command-Data-Status protocol, and not so much as a parallel interface. Almost all data storage interfaces, including Fibre Channel, iSCSI, SAS, and even USB use this same protocol to move blocks of data between a computer and a storage device. That’s good news because virtually all operating systems know how to communicate using SCSI protocol, and that does not change when you move the command language to a different transport.
(vs. files)