atom feed12 messages in org.openldap.openldap-technicalIssues in Performance Tesing of OpenL...
FromSent OnAttachments
Shilpa SethiNov 10, 2008 3:58 am 
Quanah Gibson-MountNov 10, 2008 8:20 am 
Shilpa SethiNov 14, 2008 4:18 am 
Quanah Gibson-MountNov 14, 2008 11:33 am 
Shilpa SethiNov 18, 2008 4:28 am.xls
Quanah Gibson-MountNov 19, 2008 2:02 pm 
Howard ChuNov 19, 2008 2:23 pm 
Quanah Gibson-MountNov 19, 2008 2:28 pm 
Shilpa SethiNov 20, 2008 6:29 am 
Quanah Gibson-MountNov 20, 2008 10:32 am 
Shilpa SethiNov 24, 2008 10:35 pm 
Howard ChuNov 24, 2008 10:51 pm 
Subject:Issues in Performance Tesing of OpenLDAP using SLAMD
From:Shilpa Sethi (shil@aricent.com)
Date:Nov 10, 2008 3:58:24 am
List:org.openldap.openldap-technical

Hi all,

Currently I am involved in an activity that deals with measuring performance of
OpenLDAP on Solaris 10 - x_86 machine.

Hardware Specs -

1. Two identical servers (Servers 'NodeA' and 'NodeB') with the following
configuration: * Sun Fire X4200 M2 x64 Server * 2x AMD Opteron Model 2220 (2.8GHz/1MB) processor * 1x 4GB PC2-5300 DDR2-667 memory * 2x 146GB HDD 15K RPM (Hardware RAID-1, so available disk space is 146
GB). 2. One server 'NodeC' is used as a traffic injector and has the following
hardware configuration: * Sun Fire X4200 M2 x64 Server * 2x AMD Opteron Model 2218 * 4x 2GB PC2-5300 DDR2-667 memory * 4x 73GB HDD 10K RPM. Software installation common to all servers:

1. JDK 1.6 2. Symas CDS Version 3.9.3

For traffic injection, SLAMD v2.0.0-alpha1 is installed on NodeC.

Both Node A and B are Mirror-Mode replicas.

The database is having 1 Million subscribers and both the cache sizes are set as
follows in slapd.conf : cachesize 250000 idlcachesize 250000 i.e 25% of the subscriber base.

Two identical SLAMD SearchRate Jobs, one pointing to Node A and other to Node B
are run from Node C. However, there is a wide difference in the number of lookups/sec performed in
both the servers.

The read rates obtained vary in the ratio of 1:7 or so on both the servers.

I am using 1 SLAMD client and 2 threads for targeting on 1 server.

If I increase the number of thread, SLAMD starts reporting exceptions.

Please suggest a way by which Predictable/ similar read rates could be achieved
on both the servers as both being exactly same.

Thanks a ton in advance for suggesting a solution.

A R I C E N T Plot 6, Electronic City, Sector - 18, Gurgaon 122015 Haryana, India