Red Hat Directory Server 7.1 for HP-UX11i v2 Integrity System Performance Tuning and Sizing Guideline

Red Hat Directory Server 7.1 Performance Tuning and Sizing Guidelines
Page 11
lower. Setting the value to 0 turns off the cache autosizing and uses
the normal nsslapd-cachememsize and nsslapd-dbcachesize attributes.
Entry DN: cn=config,cn=ldbm database,cn=plugins,cn=config
Valid Range: 0 (turns cache autosizing off) to 100
Default Value: 0
Syntax: Integer
Example: nsslapd-cache-autosize: 80
If you set nsslapd-cache-autosize to a non-zero value between 0 and 100, RHDS will override the
nsslapd-cachememsize and nsslapd-dbcachesize with a value calculated based on the amount of free
memory at the time RHDS starts up. This is often not preferable as free memory might fluctuate from time to time
depending on usage of the system.
Logging
Directory Server provides three types of logs to help you better manage your directory and tune performance. These
logs include:
• Access log: The access log contains detailed information about client connections to the directory. It
provides beneficial troubleshooting information. Access logging is enabled by default.
• Error log: The error log contains detailed messages of errors and events the directory experiences during
normal operations. Error logging is enabled by default.
• Audit log: The audit log contains detailed information about changes made to each database as well as to
server configuration. By default, audit logging is disabled.
In a typical production environment, access log should be turned off, otherwise it will cause excessive disk I/O
despite buffering which will affect performance. For customers who want to have access log turned on, they can put
the access log file on a separate physical disk to improve the performance. In the performance test environment,
turning off the access log can increase the search throughput by 4%. Please see table 6.
Performance Measurements
This section describes performance testing for Red Hat Directory Server version 7.1 under controlled environment
with databases containing 100, 10K, 100K, 250K, 500K, 1M, 5M
∗
and 10M
∗
entries. The directory entries were
inetorgPerson entries generated by DirMark. Performance data was measured for exact search on CN only.
Purpose
• Find out how number of CPUs affects the performance.
• Find out how threadnumbers affects the performance.
• Find out how dbcachesize affects the performance.
• Find out how cachememsize and cachesize affect the performance.
• Find out how logging affects the performance.
• Find out the performance differences between a SSL connection and a non-SSL connection.
Test Result
Data collection 1: (Different Number of CPUs)
This set of data is collected to help us understand the performance differences between different numbers of CPUs.
• #of entries: 100k entries
∗
See Appendix C and Appendix D for detail