 |
|
Oracle Tips by Burleson |
Choosing a Benchmark
In deciding to perform a series of tests to
evaluate the effectiveness of using solid-state disk drives with the
Oracle database system, a test bed that showcased state-of-the-art
technology had to be obtained. Texas Memory Systems provided access
and sole use to a 64 gigabyte TMS solid state drive, a set of Linux
servers and a 64 gigabyte SCSI (and later SATA) disk array. Over the
period of May through August of 2004 a series of tests was conducted
utilizing version 9.2.0.4 of the Oracle database system and various
configurations of the SSD, server and SCSI/SATA drives to evaluate the
effectiveness of various combinations.
In reviewing the existing research and applying
logic to the results a large performance gain should be seen from
placing entire specific types of databases on SSD. Specifically
databases that are predominantly read-based such as decision support (DSS)
or data warehouses should benefit the most. In order to determine if
this is indeed a fact, the TPCH benchmarkwill be
used. The TPCH benchmark was designed for DSS testing
The next section will provide an overview of the
details behind the TPCH benchmark in order to demonstrate the
reasoning for this choice.
The above book excerpt is from:
Oracle
Solid State Disk Tuning
High Performance Oracle
tuning with RAM disk
ISBN
0-9744486-5-6
Donald K. Burleson & Mike Ault
http://www.rampant-books.com/book_2005_1_ssd.htm
|
