incubator-clerezza-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Reto Bachmann-Gmür (JIRA) <>
Subject [jira] Created: (CLEREZZA-347) Performance degradation after the firts million requests
Date Tue, 02 Nov 2010 08:58:25 GMT
Performance degradation after the firts million requests

                 Key: CLEREZZA-347
             Project: Clerezza
          Issue Type: Bug
         Environment: java version "1.6.0_20"
OpenJDK Runtime Environment (IcedTea6 1.9.1) (6b20-1.9.1-1ubuntu3)
OpenJDK 64-Bit Server VM (build 17.0-b16, mixed mode)
            Reporter: Reto Bachmann-Gmür

After installing the java-ssp-ttl prototype sample I do the following to measure performance

for i in `seq 1000`; do echo round $i; ab -n 10000 -H "Accept: application/xhtml+xml" -c 100
-A admin:admin http://localhost:8080/greeting/list; done

Depending on the heap settings of clerezza the performance starts degradating massively after
a period of relative constant sped.

Degradation starts at (> 10000000 = no degradation observed)

-Xmx200m: > 10'000'000
-Xmx600m: > 10'000'000
-Xmx1g: 610'000
-Xms1200: 750'000
-Xmx2g: 1'300'000
-Xmx3g: > 10'000'000

no mx argument set (on a machine with 5964MB of ram):  970'000

Repeated measurement with the same mx-value yield to almost identical results.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message