Fragmentation and RAID
Diskeeper is compatible with all levels of RAID software and hardwarebased with and without parity. The defragmentation method is the same whether it's Software or Hardware based.
Physical members in the RAID environment are not read or written todirectly by an application. Even the Windows NT file system sees it as one single "virtual" drive. Thisvirtual drive has logical cluster numbering just like any other partition supported under Windows NT.
As an application reads and writes to this virtual environment (creatingnew files, extending existing ones, as well as deleting others) the free space becomes fragmented, thus causingnew and existing files to be fragmented as they are written. Because of this fact, fragmentation on this virtualdrive WILL HAVE a substantial negative performance effect. When an I/O request is processed by the file system,there are a number of attributes that must be checked which cost valuable system time.
If an application has to issue multiple "unnecessary" I/Orequests, as in the case of fragmentation, not only is the processor kept busier than needed, but once the I/Orequest has been issued, the RAID software must process it and determine which physical member to direct the I/Orequest. In fact, this can negate any and all benefits of having RAID in the first place as these unnecessary fragmentedI/O requests take up extra bandwidth causing the RAID to be less and less effective.
The question now becomes how does Diskeeper effect this? Diskeepersees the RAID environment just as the file system does. That is, Diskeeper defragments the virtual drive. Diskeeperimproves the speed and performance of a RAID environment by eliminating these wasteful and unnecessary I/Os frombeing issued by the file system. This occurs because the file system sees the files and free space as being morecontiguous.
The file system will spend less time checking file attributes whichmeans more processor time can be dedicated to doing real useful work for the user/application. In addition, theseI/O requests are now much more likely to fill the entire 64K chunk size with the I/O now taking full advantageof the RAID.
If you have any comments about this article orany requests for new technical articles e-mail
Executive Software Europe
