Disk Drives - Test Methodology
Laboratory Home
Test Results
Papers
This Site
Archive
Email List
Information
& Assoc. Home
|
|
NEAL
NELSON
BENCHMARK
LABORATORY
Disk Drives - Test Methodology |
||||||
|
Benchmark Laboratory Home |
Summary Test Results |
White Papers |
About This Site |
Press Archive |
Our Email List |
Contact Information |
Neal Nelson & Assoc. Home |
|---|---|---|---|---|---|---|---|
The disk drive tests are performed on the client/server test bed located in our Chicago area offices. We use MySQL as the database. We install the disk drive to be tested in a server with two AMD Opteron 242 CPUs and 512 megabytes of RAM.
The server runs SUSE Linux 10.1 and has a 20 gigabyte IDE drive as a boot drive. The test MySQL database is loaded on the drive to be tested which is connected to a secondary controller/adapter.
The end user's experience of disk I/O speed is affected by the size of the database relative to the total capacity of the disk drive. For example, a four gigabyte database that is clustered in one small area of a 320 gigabyte drive will be accessed faster than an identical four gigabyte database which fills a four gigabyte drive. For this reason we run a series of tests with different sizes of databases. Our current test sizes are 3.5, 16, 62 and 210 gigabytes.
Click here to read white papers that describe this procedure in more detail.
|
Copyright © 2006-2008 Neal Nelson & Associates |
Trademarks which may be mentioned on this site are the property of their owners. |
|
|---|