Many higher-end systems, especially servers, now employ a technology called redundant
arrays of inexpensive disks, or RAID. This concept allows for great
improvements in both reliability and performance. The idea is to store data on multiple
disk drives running in parallel. The primary motivation in many cases is reliability. From a performance standpoint, most RAID
levels improve performance by allowing multiple accesses to happen simultaneously, and
also by using algorithms that reduce seek time and latency by taking advantage of having
multiple drives at their disposal. The exact performance impact depends entirely on the
level of RAID used; some improve read performance at the expense of write performance, for
example.
I have written an entire section on RAID that covers
its issues, levels and implementation in some detail, including a discussion of its impact
on storage subsystem performance. Once used almost exclusively in a corporate setting for
large, expensive machines, a new crop of inexpensive RAID controllers and hard disks is
bringing RAID into the "mainstream", and small RAID arrays are now commonly seen
in the machines of "power users".
Next: File System Factors
