Studying for the A+, Network+ or Security+ exams? Get over 2,600 pages of FREE study guides at CertiGuide.com!|
Join the PC homebuilding revolution! Read the all-new, FREE 200-page online guide: How to Build Your Own PC!
NOTE: Using robot software to mass-download the site degrades the server and is prohibited. See here for more.
Find The PC Guide helpful? Please consider a donation to The PC Guide Tip Jar. Visa/MC/Paypal accepted.
|Take a virtual vacation any time at DesktopScenes.com - view my art photos online for FREE in either Flash or HTML!|
The current IDE/ATA standard is a parallel interface; this means that multiple bits of data are transmitted at one time. In the case of ATA, 16 bits are moved across the interface simultaneously during each transfer. The advantage of a parallel interface is that it allows for high throughput; the problem with it is that as the frequency of the interface is increased, signaling problems and interference between signals become common. To combat this, techniques such as CRC and special 80-conductor cables are used in higher-speed transfer modes such as Ultra DMA. These are really "kludges" that are used to work around problems with the interface as it moves to higher speeds. A different approach, however, is to abandon the parallel concept in favor of a serial interface, where only one bit is transferred at a time. This is what the Serial ATA proposal is all about: creating a serial version of ATA for attaching IDE/ATA hard disks.
Obviously, in going from 16 bits to 1 bit, the speed of the interface must be increased by a factor of 16 just to "break even". The idea is that the simplicity of the serial interface will enable much higher speeds than would be possible from a parallel implementation, because the signaling problems are largely eliminated. In fact, this is the same reasoning that led to the creation of other high-speed serial interfaces, such as IEEE-1394. Serial ATA is still in development at the time that I write this page, but indications are that it will support maximum throughput of somewhere between 150 and 300 MB/s.
As enticing as the higher speed of the interface is its promise of improvements to some of the well-known (and well-hated) weaknesses of IDE/ATA. Since Serial ATA is a point-to-point serial protocol, each device communicates directly with the host system over a flexible, thin cable that can be made a reasonable length. This means no more master/slave jumpering hassles, elimination of the difficult-to-deal-with ribbon cables, and more flexibility in the placement of devices within the PC. It's also possible that hot-swapping will be supported when Serial ATA is implemented, which would be a welcome feature as far as I am concerned!
We'll have to see what happens regarding Serial ATA. I am a little skeptical about both its likelihood of being successfully implemented, and in fact with whether it is even needed. Proponents of serial hard disk interfaces have been saying "the end is nigh" regarding parallel interfaces for many years; I remember hearing about the "end of the road" when parallel IDE/ATA was at 16.6 MB/s, and parallel SCSI was at 40 MB/s, but these interfaces continue to get faster and faster. However, if they can really make Serial ATA work, especially at such high speeds, it may be a welcome improvement on what has really become a rather outdated design.
I guess we'll just have to see what happens. At any rate, I wouldn't look for Serial ATA to even be introduced to the market until late 2001 or even 2002. Even if it is successful, conventional parallel IDE/ATA will be popular for many, many years due to the hundreds of millions of systems that use it, so don't worry about obsolescence just yet. :^)