JVM Tuning in SOA 10.1.3.4

Hi experts,
Thanks for the information.
We are currently facing issue. The description is as follows:
1. We are having a SOA cluster with version 10.1.3.4 and facing high memory utilization on both the application nodes.
2. We have observed that the Java process will be consuming huge amount of RAM.
3. The start parameters for the OC4J_SOA container is as mentioned below.
<ias-component id="SOA_GROUP" status="enabled">
<process-type id="OC4J_SOA" module-id="OC4J" status="enabled">
<module-data>
<category id="start-parameters">
<data id="java-options" value="-server -Xmx6144m -Xms6144m -Xmn3687m -Xrs -XX:+AggressiveHeap -XX:MaxPermSize=768M -XX:NewSize=3687M
-Djava.security.policy=$ORACLE_HOME/j2ee/OC4J_SOA/config/java2.policy -Djava.awt.headless=true -Dhttp.webdir.enable=false -Doc4j.userThreads=true
-Doracle.mdb.fastUndeploy=60 -Doc4j.formauth.redirect=true -Djava.net.preferIPv4Stack=true -Dorabpel.home=/app/oracle/product/10.1.3/bpel
-Xbootclasspath^/p:/app/oracle/product/10.1.3/bpel/lib/orabpel-boot.jar -Dhttp.proxySet=false
-Doraesb.home=/app/oracle/product/10.1.3/integration/esb -DHTTPClient.disableKeepAlives=true -Dstdstream.filenumber=30 -Dstdstream.rotatetime=00:00
-Dstdstream.filesize=20 "/>
<data id="oc4j-options" value="-out $ORACLE_HOME/opmn/logs/oc4j_soa.out -err $ORACLE_HOME/opmn/logs/oc4j_soa.err"/>
</category>
<category id="stop-parameters">
<data id="java-options" value="-Djava.security.policy=$ORACLE_HOME/j2ee/OC4J_SOA/config/java2.policy -Djava.awt.headless=true
-Dhttp.webdir.enable=false"/>
</category>
</module-data>
<start timeout="600" retry="2"/>
<stop timeout="120"/>
<restart timeout="720" retry="2"/>
<port id="default-web-site" range="12501-12600" protocol="ajp"/>
<port id="rmi" range="12401-12500"/>
<port id="rmis" range="12701-12800"/>
<port id="jms" range="12601-12700"/>
<process-set id="SOA_GROUP" numprocs="1"/>
</process-type>
</ias-component>
4. Total available RAM is 20GB.
5. I would like to know why the following parameter is set: "<process-set id="SOA_GROUP" numprocs="1"/>" ?
Could you please let me know whether there are any issues in the settings as we suspect the GC is not happening properly due to huge heap size.
It would be really helpful if you could answer this.
Thanks in Advance,
Kumar

Please raise a case with support if it is a production issue. SIMILAR threads I found are -
High Memory Utilization in Weblogic servers
XML Debatching in BPEL!!
Regards,
Anuj

Similar Messages

Linux command/shell script to find Memory consumed by JVM in EBS,SOA

Hi All,
Could anyone please let me know the following:
a)Linux command/shell script to find Memory consumed by JVM for every 1 minute frequency interval in EBS R12, SOA suite
b)Percentage of the Heap memory usage by the JVM for every 1 minute frequency interval in EBS R12, SOA suite
Thanks for your time!
Regards,

user10088255 wrote:
Hi All,
Could anyone please let me know the following:
a)Linux command/shell script to find Memory consumed by JVM for every 1 minute frequency interval in EBS R12, SOA suite
b)Percentage of the Heap memory usage by the JVM for every 1 minute frequency interval in EBS R12, SOA suite
Thanks for your time!
Regards,Please see these docs.
JVM: Guidelines to setup the Java Virtual Machine in Apps Ebusiness Suite 11i and R12 [ID 362851.1]
Generate JVM heap dump in E-Business suite 11i and R12 [ID 462550.1]
11i: How To Use Jconsole to Monitor JVM with Oracle E-Business Suite 11i [ID 415455.1]
Configuring various JVM tuning parameters for Oracle E-Business suite 11i and R12 [ID 567551.1]
Script to find Apache, Java, Jinitiator, Forms version and JVM details for Oracle E-Business Suite 11i [ID 466890.1]
JVM Memory Monitoring, Tuning, Garbage Collection, Out of Memory, and Heap Dump Analysis For SOA Suite Integration 11g [ID 1358719.1]
How to Perform Memory and Thread Analysis on SOA [ID 1450293.1]
Thanks,
Hussein

OBIEE JVM tuning

Does anyone have any links to this issue? I need to better understand how to tune JVM for our OBIEE (11g) installation on (RHEL 5) Linux 64 bit machine.

Hi,
Better to Use JRokit jdk1.6 version.
Thanks
Deva

Garbage collection pauses screw up my cluster. JVM tuning suggestions?

I am running Sun's 64 bit JVM 1.7.0. Everything hums along nicely until my storage enabled nodes all show GC log entries like these and then the cluster experiences timeouts. This happens like once a day at most. Suggestions? Thinking maybe add more storage enabled JVMs? Any JVM settings I should use to prevent this? What can determine the cause? I don't think the machine's CPU is maxed out at the time.
Thanks,
Andrew
2012-02-02T09:09:15.585-0600: 61965.351: [GC 61965.351: [ParNew: 915634K->5291K(1023552K), 0.0396299 secs] 1143514K->233236K(3299264K), 0.0406305 secs] [Times: user=1.26 sys=0.00, real=0.04 secs]
2012-02-02T09:09:22.816-0600: 61972.582: [GC 61972.583: [ParNew: 915115K->4454K(1023552K), 0.0497440 secs] 1143060K->233468K(3299264K), 0.0510981 secs] [Times: user=0.86 sys=0.02, real=0.05 secs]
2012-02-02T09:09:29.647-0600: 61979.413: [GC 61979.413: [ParNew: 914278K->2750K(1023552K), 0.0141793 secs] 1143292K->232548K(3299264K), 0.0152009 secs] [Times: user=0.44 sys=0.00, real=0.02 secs]
2012-02-02T09:09:37.432-0600: 61987.198: [GC 61987.199: [ParNew: 912574K->2926K(1023552K), 0.0145958 secs] 1142372K->232786K(3299264K), 0.0156003 secs] [Times: user=0.45 sys=0.00, real=0.02 secs]
2012-02-02T09:09:45.213-0600: 61994.979: [GC 61994.980: [ParNew: 912750K->3264K(1023552K), 0.0133675 secs] 1142610K->233225K(3299264K), 0.0144490 secs] [Times: user=0.02 sys=0.00, real=0.02 secs]
2012-02-02T09:09:50.709-0600: 62000.474: [GC 62000.475: [ParNew: 913088K->2920K(1023552K), 0.0169685 secs] 1143049K->232971K(3299264K), 0.0179272 secs] [Times: user=0.31 sys=0.00, real=0.02 secs]
2012-02-02T09:09:57.938-0600: 62007.704: [GC 62007.705: [ParNew: 912744K->4014K(1023552K), 0.0367154 secs] 1142795K->234128K(3299264K), 0.0377584 secs] [Times: user=0.36 sys=0.00, real=0.04 secs]
2012-02-02T09:10:04.134-0600: 62013.899: [GC 62013.899: [ParNew: 913838K->4306K(1023552K), 0.0179481 secs] 1143952K->234465K(3299264K), 0.0188570 secs] [Times: user=0.28 sys=0.00, real=0.02 secs]
2012-02-02T09:10:10.223-0600: 62019.988: [GC 62019.989: [ParNew: 914130K->4890K(1023552K), 0.0258524 secs] 1144289K->235119K(3299264K), 0.0267208 secs] [Times: user=0.37 sys=0.00, real=0.03 secs]
2012-02-02T09:10:19.538-0600: 62029.304: [GC 62029.305: [ParNew: 914714K->106275K(1023552K), 0.4447269 secs] 1144943K->336555K(3299264K), 0.4456875 secs] [Times: user=3.46 sys=0.14, real=0.45 secs]
2012-02-02T09:10:28.095-0600: 62037.860: [GC 62037.861: [ParNew: 1016099K->113728K(1023552K), 2.3496060 secs] 1246379K->627933K(3299264K), 2.3506549 secs] [Times: user=11.48 sys=1.48, real=2.35 secs]
2012-02-02T09:10:40.675-0600: 62050.440: [GC 62050.441: [ParNew: 1023552K->113728K(1023552K), 1.6754442 secs] 1537757K->946051K(3299264K), 1.6765925 secs] [Times: user=36.85 sys=0.37, real=1.68 secs]
2012-02-02T09:10:52.540-0600: 62062.306: [GC 62062.306: [ParNew: 1023552K->113728K(1023552K), 1.6465925 secs] 1855875K->1262979K(3299264K), 1.6475185 secs] [Times: user=38.24 sys=0.61, real=1.65 secs]
2012-02-02T09:10:54.190-0600: 62063.955: [GC [1 CMS-initial-mark: 1149251K(2275712K)] 1276188K(3299264K), 0.2430027 secs] [Times: user=0.25 sys=0.00, real=0.24 secs]
2012-02-02T09:10:54.434-0600: 62064.199: [CMS-concurrent-mark-start]
2012-02-02T09:10:55.375-0600: 62065.140: [CMS-concurrent-mark: 0.941/0.941 secs] [Times: user=10.44 sys=0.25, real=0.94 secs]
2012-02-02T09:10:55.375-0600: 62065.140: [CMS-concurrent-preclean-start]
2012-02-02T09:10:56.231-0600: 62065.996: [CMS-concurrent-preclean: 0.836/0.856 secs] [Times: user=1.78 sys=0.00, real=0.86 secs]
2012-02-02T09:10:56.231-0600: 62065.996: [CMS-concurrent-abortable-preclean-start]
2012-02-02T09:11:00.905-0600: 62070.669: [CMS-concurrent-abortable-preclean: 4.665/4.673 secs] [Times: user=9.75 sys=0.03, real=4.67 secs]
2012-02-02T09:11:00.906-0600: 62070.671: [GC[YG occupancy: 626192 K (1023552 K)]62070.672: [Rescan (parallel) , 0.4593515 secs]62071.132: [weak refs processing, 0.8817154 secs]62072.014: [scrub string table, 0.0006451 secs] [1 CMS-remark: 1149251K(2275712K)] 1775444K(3299264K), 1.7319597 secs] [Times: user=15.60 sys=0.00, real=1.73 secs]
2012-02-02T09:11:02.639-0600: 62072.404: [CMS-concurrent-sweep-start]
2012-02-02T09:11:04.088-0600: 62073.853: [CMS-concurrent-sweep: 1.443/1.449 secs] [Times: user=4.63 sys=0.05, real=1.45 secs]
2012-02-02T09:11:04.088-0600: 62073.853: [CMS-concurrent-reset-start]
2012-02-02T09:11:04.101-0600: 62073.866: [CMS-concurrent-reset: 0.013/0.013 secs] [Times: user=0.00 sys=0.00, real=0.01 secs]
2012-02-02T09:11:09.312-0600: 62079.077: [GC 62079.077: [ParNew: 1023552K->113728K(1023552K), 1.6178150 secs] 1587225K->979876K(3299264K), 1.6187922 secs] [Times: user=37.58 sys=0.11, real=1.62 secs]
2012-02-02T09:11:21.172-0600: 62090.937: [GC 62090.937: [ParNew: 1023552K->113728K(1023552K), 1.5327921 secs] 1889700K->1243425K(3299264K), 1.5335996 secs] [Times: user=34.29 sys=0.14, real=1.53 secs]
2012-02-02T09:11:33.305-0600: 62103.070: [GC 62103.070: [ParNew: 1023552K->113728K(1023552K), 1.6349678 secs] 2153249K->1475423K(3299264K), 1.6361176 secs] [Times: user=36.40 sys=0.17, real=1.64 secs]
2012-02-02T09:11:45.480-0600: 62115.245: [GC 62115.246: [ParNew: 1023552K->113728K(1023552K), 1.6566701 secs] 2385247K->1744596K(3299264K), 1.6579350 secs] [Times: user=37.52 sys=0.25, real=1.66 secs]
2012-02-02T09:11:47.140-0600: 62116.905: [GC [1 CMS-initial-mark: 1630868K(2275712K)] 1744677K(3299264K), 0.2085259 secs] [Times: user=0.22 sys=0.00, real=0.21 secs]
2012-02-02T09:11:47.350-0600: 62117.115: [CMS-concurrent-mark-start]
2012-02-02T09:11:48.481-0600: 62118.245: [CMS-concurrent-mark: 1.130/1.130 secs] [Times: user=12.03 sys=0.02, real=1.13 secs]
2012-02-02T09:11:48.481-0600: 62118.245: [CMS-concurrent-preclean-start]
2012-02-02T09:11:49.191-0600: 62118.955: [CMS-concurrent-preclean: 0.702/0.710 secs] [Times: user=1.51 sys=0.00, real=0.71 secs]
2012-02-02T09:11:49.191-0600: 62118.955: [CMS-concurrent-abortable-preclean-start]
2012-02-02T09:11:53.016-0600: 62122.780: [CMS-concurrent-abortable-preclean: 3.819/3.825 secs] [Times: user=8.44 sys=0.05, real=3.83 secs]
2012-02-02T09:11:53.017-0600: 62122.781: [GC[YG occupancy: 637511 K (1023552 K)]62122.782: [Rescan (parallel) , 0.6066033 secs]62123.389: [weak refs processing, 0.8883109 secs]62124.278: [scrub string table, 0.0005759 secs] [1 CMS-remark: 1630868K(2275712K)] 2268379K(3299264K), 1.8602409 secs] [Times: user=16.88 sys=0.02, real=1.86 secs]
2012-02-02T09:11:54.878-0600: 62124.643: [CMS-concurrent-sweep-start]
2012-02-02T09:11:57.491-0600: 62127.255: [CMS-concurrent-sweep: 2.610/2.613 secs] [Times: user=7.57 sys=0.16, real=2.61 secs]
2012-02-02T09:11:57.491-0600: 62127.256: [CMS-concurrent-reset-start]
2012-02-02T09:11:57.498-0600: 62127.262: [CMS-concurrent-reset: 0.007/0.007 secs] [Times: user=0.00 sys=0.00, real=0.01 secs]
2012-02-02T09:12:01.137-0600: 62130.902: [GC 62130.902: [ParNew: 1023552K->113728K(1023552K), 1.4807261 secs] 1678232K->1015472K(3299264K), 1.4816808 secs] [Times: user=26.47 sys=0.00, real=1.48 secs]
2012-02-02T09:12:13.113-0600: 62142.877: [GC 62142.878: [ParNew: 1023552K->113728K(1023552K), 1.5627711 secs] 1925296K->1258829K(3299264K), 1.5635313 secs] [Times: user=34.26 sys=0.00, real=1.56 secs]
2012-02-02T09:12:25.167-0600: 62154.931: [GC 62154.932: [ParNew: 1023552K->113728K(1023552K), 1.5751999 secs] 2168653K->1510600K(3299264K), 1.5762551 secs] [Times: user=31.75 sys=0.00, real=1.58 secs]
2012-02-02T09:12:36.964-0600: 62166.729: [GC 62166.730: [ParNew: 1023552K->113728K(1023552K), 1.6056739 secs] 2420424K->1768193K(3299264K), 1.6065643 secs] [Times: user=36.02 sys=0.08, real=1.61 secs]
2012-02-02T09:12:38.573-0600: 62168.337: [GC [1 CMS-initial-mark: 1654465K(2275712K)] 1768195K(3299264K), 0.1967081 secs] [Times: user=0.20 sys=0.00, real=0.20 secs]
2012-02-02T09:12:38.770-0600: 62168.535: [CMS-concurrent-mark-start]
2012-02-02T09:12:39.611-0600: 62169.376: [CMS-concurrent-mark: 0.841/0.841 secs] [Times: user=9.24 sys=0.00, real=0.84 secs]
2012-02-02T09:12:39.611-0600: 62169.376: [CMS-concurrent-preclean-start]
2012-02-02T09:12:40.036-0600: 62169.801: [CMS-concurrent-preclean: 0.419/0.425 secs] [Times: user=0.84 sys=0.02, real=0.42 secs]
2012-02-02T09:12:40.036-0600: 62169.801: [CMS-concurrent-abortable-preclean-start]
CMS: abort preclean due to time 2012-02-02T09:12:46.001-0600: 62175.765: [CMS-concurrent-abortable-preclean: 5.955/5.964 secs] [Times: user=12.00 sys=0.08, real=5.96 secs]
2012-02-02T09:12:46.003-0600: 62175.767: [GC[YG occupancy: 770945 K (1023552 K)]62175.767: [Rescan (parallel) , 0.7886971 secs]62176.557: [weak refs processing, 0.8179398 secs]62177.375: [scrub string table, 0.0008280 secs] [1 CMS-remark: 1654465K(2275712K)] 2425410K(3299264K), 1.8989263 secs] [Times: user=25.40 sys=0.02, real=1.90 secs]
2012-02-02T09:12:47.903-0600: 62177.667: [CMS-concurrent-sweep-start]
2012-02-02T09:12:50.676-0600: 62180.440: [CMS-concurrent-sweep: 2.772/2.774 secs] [Times: user=8.20 sys=0.06, real=2.77 secs]
2012-02-02T09:12:50.676-0600: 62180.441: [CMS-concurrent-reset-start]
2012-02-02T09:12:50.696-0600: 62180.460: [CMS-concurrent-reset: 0.019/0.019 secs] [Times: user=0.05 sys=0.00, real=0.02 secs]
2012-02-02T09:12:52.272-0600: 62182.037: [GC 62182.037: [ParNew: 1023552K->113728K(1023552K), 1.5954584 secs] 1584704K->951316K(3299264K), 1.5963712 secs] [Times: user=34.26 sys=0.00, real=1.60 secs]
2012-02-02T09:13:34.402-0600: 62224.165: [GC 62224.166: [ParNew: 1023552K->113728K(1023552K), 0.9755846 secs] 1861140K->1084527K(3299264K), 0.9767018 secs] [Times: user=19.19 sys=0.02, real=0.98 secs]
2012-02-02T09:16:04.628-0600: 62374.391: [GC 62374.391: [ParNew: 1023552K->113517K(1023552K), 0.1955335 secs] 1994351K->1173747K(3299264K), 0.1966082 secs] [Times: user=2.12 sys=0.05, real=0.20 secs]
2012-02-02T09:18:44.395-0600: 62534.156: [GC 62534.157: [ParNew: 1023341K->19351K(1023552K), 0.0506139 secs] 2083571K->1079581K(3299264K), 0.0517891 secs] [Times: user=0.37 sys=0.00, real=0.05 secs]
2012-02-02T09:21:20.270-0600: 62690.030: [GC 62690.030: [ParNew: 929175K->8239K(1023552K), 0.0405564 secs] 1989405K->1068470K(3299264K), 0.0417057 secs] [Times: user=0.76 sys=0.00, real=0.04 secs]
2012-02-02T09:23:54.367-0600: 62844.125: [GC 62844.126: [ParNew: 918063K->5778K(1023552K), 0.0497713 secs] 1978294K->1066008K(3299264K), 0.0508845 secs] [Times: user=0.58 sys=0.00, real=0.05 secs]
2012-02-02T09:26:33.522-0600: 63003.279: [GC 63003.279: [ParNew: 915602K->5421K(1023552K), 0.0524503 secs] 1975832K->1065651K(3299264K), 0.0536713 secs] [Times: user=0.59 sys=0.02, real=0.06 secs]
2012-02-02T09:29:13.270-0600: 63163.025: [GC 63163.026: [ParNew: 915245K->5390K(1023552K), 0.0317213 secs] 1975475K->1065620K(3299264K), 0.0330578 secs] [Times: user=0.25 sys=0.00, real=0.03 secs]
2012-02-02T09:31:58.001-0600: 63327.754: [GC 63327.755: [ParNew: 915214K->5524K(1023552K), 0.0617393 secs] 1975444K->1065894K(3299264K), 0.0629008 secs] [Times: user=1.05 sys=0.00, real=0.06 secs]
2012-02-02T09:34:43.454-0600: 63493.206: [GC 63493.207: [ParNew: 915348K->5600K(1023552K), 0.0376009 secs] 1975718K->1066058K(3299264K), 0.0388287 secs] [Times: user=0.36 sys=0.00, real=0.04 secs]
2012-02-02T09:37:15.601-0600: 63645.352: [GC 63645.353: [ParNew: 915424K->6656K(1023552K), 0.0273727 secs] 1975882K->1067278K(3299264K), 0.0285435 secs] [Times: user=0.56 sys=0.00, real=0.03 secs]
2012-02-02T09:39:41.873-0600: 63791.623: [GC 63791.623: [ParNew: 916480K->8588K(1023552K), 0.0239936 secs] 1977102K->1069403K(3299264K), 0.0252618 secs] [Times: user=0.75 sys=0.00, real=0.03 secs]
2012-02-02T09:42:02.276-0600: 63932.025: [GC 63932.025: [ParNew: 918412K->8408K(1023552K), 0.0259162 secs] 1979227K->1069313K(3299264K), 0.0269622 secs] [Times: user=0.34 sys=0.02, real=0.03 secs]
2012-02-02T09:44:22.990-0600: 64072.737: [GC 64072.738: [ParNew: 918232K->8777K(1023552K), 0.0208236 secs] 1979137K->1069755K(3299264K), 0.0220036 secs] [Times: user=0.41 sys=0.00, real=0.02 secs]

Today i used
-XX:+UseConcMarkSweepGC
-XX:+UseParNewGC
-XX:-CMSConcurrentMTEnabled
-XX:+CMSScavengeBeforeRemark
and had the same problem with all 3 storage enabled nodes simultaneously when I closed the Extend app. The GC logs from those storage nodes were nearly identical with each showing three "abort preclean" messages like below before eventually stabilizing again. CPU usage for the storage node JVMs was up a bit on the machine at the time then eventually came down. Extend proxy node CPU usage never goes up very high and GC times are always good there.
Could it be because the Extend app has many CQCs running and when the app disconnects without shutting them down properly their results queue up in the storage enabled nodes? Something the Extend proxy can't pass along is getting added to old space?
9:04:29.313-0600: 34201.081: [GC 34201.081: [ParNew: 914605K->4950K(1023552K), 0.0469807 secs] 1057777K->148216K(3299264K), 0.0480525 secs] [Times: user=0.78 sys=0.03, real=0.05 secs]
9:04:35.742-0600: 34207.509: [GC 34207.509: [ParNew: 914774K->4559K(1023552K), 0.0317232 secs] 1058040K->147868K(3299264K), 0.0324390 secs] [Times: user=0.48 sys=0.00, real=0.03 secs]
9:04:44.806-0600: 34216.573: [GC 34216.574: [ParNew: 914383K->113728K(1023552K), 1.4924395 secs] 1057692K->348748K(3299264K), 1.4934226 secs] [Times: user=7.52 sys=0.81, real=1.49 secs]
9:04:56.498-0600: 34228.265: [GC 34228.265: [ParNew: 1023552K->113728K(1023552K), 1.3332836 secs] 1258572K->634502K(3299264K), 1.3341394 secs] [Times: user=28.58 sys=0.44, real=1.34 secs]
9:05:09.845-0600: 34241.612: [GC 34241.612: [ParNew: 1023552K->113728K(1023552K), 1.5617413 secs] 1544326K->920121K(3299264K), 1.5628521 secs] [Times: user=33.43 sys=0.34, real=1.56 secs]
9:05:24.569-0600: 34256.335: [GC 34256.336: [ParNew: 1023552K->113728K(1023552K), 2.3253881 secs] 1829945K->1186221K(3299264K), 2.3263253 secs] [Times: user=33.95 sys=0.48, real=2.33 secs]
9:05:39.812-0600: 34271.579: [GC 34271.580: [ParNew: 1023552K->113728K(1023552K), 1.6817535 secs] 2096045K->1453026K(3299264K), 1.6827555 secs] [Times: user=33.60 sys=0.20, real=1.68 secs]
9:05:41.498-0600: 34273.264: [GC [1 CMS-initial-mark: 1339298K(2275712K)] 1453026K(3299264K), 0.2348391 secs] [Times: user=0.22 sys=0.02, real=0.24 secs]
9:05:41.733-0600: 34273.500: [CMS-concurrent-mark-start]
9:05:44.732-0600: 34276.498: [CMS-concurrent-mark: 2.999/2.999 secs] [Times: user=6.49 sys=0.05, real=3.00 secs]
9:05:44.732-0600: 34276.498: [CMS-concurrent-preclean-start]
9:05:45.839-0600: 34277.606: [CMS-concurrent-preclean: 1.095/1.107 secs] [Times: user=2.25 sys=0.00, real=1.11 secs]
9:05:45.839-0600: 34277.606: [CMS-concurrent-abortable-preclean-start]
CMS: abort preclean due to time 9:05:50.866-0600: 34282.633: [CMS-concurrent-abortable-preclean: 5.021/5.027 secs] [Times: user=10.14 sys=0.05, real=5.03 secs]
9:05:50.868-0600: 34282.634: [GC[YG occupancy: 809255 K (1023552 K)]9:05:50.868-0600: 34282.635: [GC 34282.635: [ParNew: 809255K->113728K(1023552K), 1.0828740 secs] 2148553K->1611793K(3299264K), 1.0835767 secs] [Times: user=19.41 sys=0.06, real=1.08 secs]
34283.719: [Rescan (parallel) , 0.1498045 secs]34283.869: [weak refs processing, 1.0258216 secs]34284.896: [scrub string table, 0.0006202 secs] [1 CMS-remark: 1498065K(2275712K)] 1611793K(3299264K), 2.7794183 secs] [Times: user=24.56 sys=0.06, real=2.78 secs]
9:05:53.648-0600: 34285.415: [CMS-concurrent-sweep-start]
9:05:56.112-0600: 34287.879: [CMS-concurrent-sweep: 2.464/2.464 secs] [Times: user=7.38 sys=0.03, real=2.46 secs]
9:05:56.112-0600: 34287.879: [CMS-concurrent-reset-start]
9:05:56.120-0600: 34287.887: [CMS-concurrent-reset: 0.008/0.008 secs] [Times: user=0.05 sys=0.00, real=0.01 secs]
9:06:06.204-0600: 34297.970: [GC 34297.971: [ParNew: 1023552K->113728K(1023552K), 1.5001175 secs] 1845197K->1139597K(3299264K), 1.5011371 secs] [Times: user=31.31 sys=0.00, real=1.50 secs]
9:06:20.044-0600: 34311.810: [GC 34311.811: [ParNew: 1023552K->113728K(1023552K), 1.6438234 secs] 2049421K->1389674K(3299264K), 1.6446616 secs] [Times: user=36.97 sys=0.03, real=1.64 secs]
9:06:34.063-0600: 34325.829: [GC 34325.830: [ParNew: 1023552K->113728K(1023552K), 2.0298132 secs] 2299498K->1648212K(3299264K), 2.0308665 secs] [Times: user=34.10 sys=0.03, real=2.03 secs]
9:06:36.096-0600: 34327.862: [GC [1 CMS-initial-mark: 1534484K(2275712K)] 1648282K(3299264K), 0.2109338 secs] [Times: user=0.22 sys=0.00, real=0.21 secs]
9:06:36.308-0600: 34328.074: [CMS-concurrent-mark-start]
9:06:39.105-0600: 34330.871: [CMS-concurrent-mark: 2.797/2.797 secs] [Times: user=6.02 sys=0.00, real=2.80 secs]
9:06:39.105-0600: 34330.871: [CMS-concurrent-preclean-start]
9:06:39.795-0600: 34331.561: [CMS-concurrent-preclean: 0.686/0.690 secs] [Times: user=1.40 sys=0.00, real=0.69 secs]
9:06:39.795-0600: 34331.561: [CMS-concurrent-abortable-preclean-start]
CMS: abort preclean due to time 9:06:46.623-0600: 34338.389: [CMS-concurrent-abortable-preclean: 6.820/6.828 secs] [Times: user=13.85 sys=0.03, real=6.83 secs]
9:06:46.624-0600: 34338.390: [GC[YG occupancy: 891686 K (1023552 K)]9:06:46.624-0600: 34338.391: [GC 34338.391: [ParNew: 891686K->113728K(1023552K), 1.2772255 secs] 2426171K->1846761K(3299264K), 1.2780695 secs] [Times: user=25.66 sys=0.11, real=1.28 secs]
34339.669: [Rescan (parallel) , 0.1456700 secs]34339.815: [weak refs processing, 0.6135547 secs]34340.431: [scrub string table, 0.0005247 secs] [1 CMS-remark: 1733033K(2275712K)] 1846761K(3299264K), 2.3580749 secs] [Times: user=30.44 sys=0.11, real=2.36 secs]
9:06:48.983-0600: 34340.749: [CMS-concurrent-sweep-start]
9:06:51.895-0600: 34343.660: [CMS-concurrent-sweep: 2.911/2.911 secs] [Times: user=7.71 sys=0.06, real=2.91 secs]
9:06:51.895-0600: 34343.660: [CMS-concurrent-reset-start]
9:06:51.941-0600: 34343.707: [CMS-concurrent-reset: 0.047/0.047 secs] [Times: user=0.09 sys=0.00, real=0.05 secs]
9:07:02.441-0600: 34354.207: [GC 34354.207: [ParNew: 1023552K->113728K(1023552K), 1.5969685 secs] 1805173K->1136718K(3299264K), 1.5980247 secs] [Times: user=33.85 sys=0.00, real=1.60 secs]
9:07:18.155-0600: 34369.921: [GC 34369.922: [ParNew: 1023552K->113728K(1023552K), 2.2864633 secs] 2046542K->1381538K(3299264K), 2.2874742 secs] [Times: user=33.74 sys=0.02, real=2.29 secs]
9:07:33.522-0600: 34385.288: [GC 34385.288: [ParNew: 1023552K->113728K(1023552K), 1.6890223 secs] 2291362K->1632772K(3299264K), 1.6898157 secs] [Times: user=35.18 sys=0.01, real=1.69 secs]
9:07:35.213-0600: 34386.979: [GC [1 CMS-initial-mark: 1519044K(2275712K)] 1632799K(3299264K), 0.2011283 secs] [Times: user=0.20 sys=0.00, real=0.20 secs]
9:07:35.415-0600: 34387.181: [CMS-concurrent-mark-start]
9:07:37.810-0600: 34389.576: [CMS-concurrent-mark: 2.395/2.395 secs] [Times: user=5.63 sys=0.00, real=2.39 secs]
9:07:37.810-0600: 34389.576: [CMS-concurrent-preclean-start]
9:07:38.623-0600: 34390.389: [CMS-concurrent-preclean: 0.808/0.813 secs] [Times: user=1.70 sys=0.05, real=0.81 secs]
9:07:38.623-0600: 34390.389: [CMS-concurrent-abortable-preclean-start]
CMS: abort preclean due to time 9:07:45.370-0600: 34397.135: [CMS-concurrent-abortable-preclean: 6.742/6.747 secs] [Times: user=13.76 sys=0.11, real=6.75 secs]
9:07:45.371-0600: 34397.136: [GC[YG occupancy: 950924 K (1023552 K)]9:07:45.372-0600: 34397.137: [GC 34397.137: [ParNew: 950924K->113728K(1023552K), 1.3829715 secs] 2469968K->1894060K(3299264K), 1.3840623 secs] [Times: user=31.28 sys=0.00, real=1.38 secs]
34398.522: [Rescan (parallel) , 0.1888848 secs]34398.711: [weak refs processing, 0.3561537 secs]34399.068: [scrub string table, 0.0004774 secs] [1 CMS-remark: 1780332K(2275712K)] 1894060K(3299264K), 2.1091730 secs] [Times: user=36.55 sys=0.00, real=2.11 secs]
9:07:47.482-0600: 34399.247: [CMS-concurrent-sweep-start]
9:07:50.500-0600: 34402.265: [CMS-concurrent-sweep: 3.017/3.018 secs] [Times: user=8.63 sys=0.33, real=3.02 secs]
9:07:50.500-0600: 34402.265: [CMS-concurrent-reset-start]
9:07:50.507-0600: 34402.273: [CMS-concurrent-reset: 0.008/0.008 secs] [Times: user=0.00 sys=0.00, real=0.01 secs]
9:09:58.395-0600: 34530.160: [GC 34530.160: [ParNew: 1023552K->113438K(1023552K), 0.4065570 secs] 1764220K->964374K(3299264K), 0.4077345 secs] [Times: user=4.63 sys=0.03, real=0.41 secs]
9:12:06.040-0600: 34657.802: [GC 34657.803: [ParNew: 1023262K->27778K(1023552K), 0.0334396 secs] 1874198K->878714K(3299264K), 0.0345056 secs]-Andrew

JVM Tuning

Is there any way to attach to a running JVM and read/change values for max heap and stack size? It seems like this would be useful for debugging and trying to determine the root cause of OutOfMemory errors.
Thanks,
Bill

If you are getting OutOfMemoryErrors (not StackOverflowErrors), then your stack size is ok. By playing with max heap size you are unlikely to get any clues about who is responsible for overflowing the heap. You need to see what objects are allocated and where, and who is on top of the list. Then you will have to check those top-number objects - whether you really need to allocate so many of them, etc. To get this info, your best bet is to use some profiling tool. E.g. Sun has a free profiler called JFluid, check for it on google. Other good (but expensive) tools are OptimizeIt, JProbe, JProfiler, etc.

Better JVM Tuning options

Hi
I'm working on a swing based application- JRE1.4.2+ Windows XP.
In our case around 10% of the compoenents are having the same lifesapn as the application. When the application is invoked it would take around 80MB(This inludes basic jre runtime+custom frame work classes) without opening any screens.
In most of the cases user may open the screens and keep it opened for long time ,but around 10% of scrreens will be closed in 2 or 3 minutes.This application is supposed to work on 256 MB RAM with JRE 1.4.2.
Following are the two sets of GC tuning parameters we have set.
Pls advise me which one is the better.
1. -client -Xms256m -Xmx256m
2. -XX:MaxTenuringThreshold=0 -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -client -Xms256m -Xmx256m
Would like to know which would bethe better one, or whether I need to add any extra parameter.?
Thanks in advance
Jobinesh

Thanks alot jon999.
I'll try with default GC and see the performance. I believe that deafult GCsettings with heap size of 256 MB would be fine for me.
Thanks
Jobinesh

Performance tuning of BPEL processes in SOA Suite 11g

Hi,
We are working with a customer for performance tuning of SOA Suite 11g, one of the areas is to tune the BPEL processes. I am new to this and started out with stress testing Hello World process using SOAPUI tool. I would like help with the below topics -
1. How do I interpret the statistics collected during stress testing? Do we have any benchmark set that can indicate that the performance is ok.
2. Do we need to run stress tests for every BPEL process deployed?
2. Is there any performance tuning strategy documentation available? Or can anybody share his/her experiences to guide me?
Thanks in advance!
Sritama

1. How do I interpret the statistics collected during stress testing? Do we have any benchmark set that can indicate that the performance is ok.
You need
pay attention to:
java heap usage vs java heap capacity
java eden usage vs java eden capacity
JDBC pool initial connections vs JDBC pool capacity connections
if you are using linux: top
if you are using aix: topas
2. Do we need to run stress tests for every BPEL process deployed?
yes, you need test each BPEL. You can use "Jmeter" tool.
Download Jmeter from here: Apache JMeter - Apache JMeter&trade;
Other tools:
jstat
jstack
jps -v
Enterprise Manager
WebLogic Console
VisualVM
JRockit Mission Control
3. Is there any performance tuning strategy documentation available? Or can anybody share his/her experiences to guide me?
I recommend "Oracle SOA Suite 11g Performance Tuning Cookbook" http://www.amazon.com/Oracle-Suite-Performance-Tuning-Cookbook/dp/1849688842/ref=sr_1_1?ie=UTF8&qid=1378482031&sr=8-1&keywords=oracle+soa+suite+11g+performance+tuning+cookbook

Tuning Oracle APPS 11i Start with

Hi Hussain/Helios,
I am new to Oracle APPS, and am working oracle apps 11i 11.5.10.2 with 10.2.5.0, the database and application was upgraded by previous dba. which left almost 3-4 months before.
now my question is i want to have a performance tuning for Oracle apps environment, can you please suggest some key points for starting with and some good links for future usage.
Other thing is i have used find command on my linux box and found many log files which are greater than 1000K on apps tier (forms/reports/web), and similiarly db tier (db/admin/concurrent).
can i just remove all these logfiles using os (rm) command and is it feasible with production environment on running mode or is it not feasible and i should use some other ways/etc.
APPS Tier folder locations.
****podappl/admin/PROD/log
****prodora/iAS/Apache/Apache/logs/sec_audit.log
****prodora/iAS/Apache/Jserv/logs/mod_jserv.log
****prodora/iAS/Apache/Jserv/logs/jserv.log
****prodcomn/admin/scripts/PROD_****/sqlnet.log
****prodcomn/admin/log/PROD_****/
****prodcomn/_pages/
****prodcomn/rgf/PROD_****/
****prodcomn/rgf/PROD_****/oam/
DB Tier files folder locations.
****oracle/proddb/10.2.0/oc4j/j2ee/OC4J_DBConsole_*****_PROD/log/em-application.log
****oracle/proddb/10.2.0/oc4j/j2ee/OC4J_DBConsole_*****_PROD/log/http-web-access.log
****oracle/proddb/10.2.0/oc4j/j2ee/OC4J_DBConsole_****_PROD.upgrade/log/http-web-access.log
****oracle/proddb/10.2.0/appsutil/log/PROD_****/StageDBTier_10161830.log
****oracle/proddb/10.2.0/*****_PROD/sysman/log/emoms.log
****oracle/proddb/10.2.0/*****_PROD.upgrade/sysman/emd/sqlnet.log
****oracle/proddb/10.2.0/******_PROD.upgrade/sysman/log/
****oracle/prodappl/admin/PROD/log/For Performance Tuning, right now i have following points with me.
1). Compile APPS Schema in maintenace which i will take.
2). I have checked some Gather Stats/Purge jobs are already scheduled for maintenance.
Please tell me what other Jobs i can schedule using OAM for weekend for All tiers in APPS.
I have been also asked to apply latest patches on PROD APPS/DB/* system, I can check using sql, which patches are applied, but how can i know which patches are required, (don't know about the licensed products with never ask policy here).
Thanks,
RAJ
Edited by: 870344 on Jul 5, 2011 6:36 AM

can you please suggest some key points for starting with and some good links for future usage.Please check below MOS doc:-
A Holistic Approach to Performance Tuning Oracle Applications Systems [ID 69565.1]
Configuring various JVM tuning parameters for Oracle E-Business suite 11i and R12 [ID 567551.1]
Tuning is something that require more analysis less action, first you need to answer:-
1) Why do you need tuning? is it proactive or some incident happend?
2) Where are you facing issues? i.e. which module, form, report or whole application.
3) Number of users facing issues? i.e. is it single user, user at one location or all users.
4) Has something changed in past that might have triggered this need?
Theri are various components involved in tuning that start from :-
1) user machine
2) network
3) Application tier.
4) Database tier.
Application tier can be further divided into:-
1) Apache server(JVM, modplsql, jserv etc etc
2) Form Server
3) Concurrent processing server
4) Discovered if using,
5) XML publisher..
and like wise other component using.
If you are looking to tune then i believe start from application tier first, pick one by one componets, like if your users are facing issues with login then use note that i have provided, if users are not satisifed with time taken to complete concurrent request then you might have to tune it and likewise.
can i just remove all these logfiles using os (rm) command and is it feasible with production environment on running mode or is it not feasible and i should use some other ways/etc.Do not rm files because it might be in use by some process and ti might then not be able to release space, its better to trim the file. do cat file >
Log file you have mentioned can be removed depedning upon your company policy if any.
Thanks,
JD

JVM Setting of Max Heap Equal To Max Perm Size

Hello,
I was wondering based on anyones experience or knowledge if there would be any bad effects in setting the Maximum heap size (-Xmxs) equal to the Maximum Permanent Size (-XX:MaxPermSize)?
I have a person that has the following settings of:
-native -Xnoclassgc -ms256m -mx256m -XX:MaxPermSize=256M
running under WLS 6.1 sp1 on a Sun Solaris box.
Just curious how this would affect applications running under this environment and if it could attribute to a java.lang.outofmemory error be generated.
Thanks,
Bob Krause
PHS

"Bob Krause" <[email protected]> wrote in message news:19491214.1100812698044.JavaMail.root@jserv5...
I was wondering based on anyones experience or knowledge if there would be any bad effects in setting the Maximum heap size(-Xmxs) equal to the Maximum Permanent Size (-XX:MaxPermSize)?
>
I have a person that has the following settings of:
-native -Xnoclassgc -ms256m -mx256m -XX:MaxPermSize=256M
running under WLS 6.1 sp1 on a Sun Solaris box.
Just curious how this would affect applications running under this environment and if it could attribute to ajava.lang.outofmemory error be generated.
The settings above don't make much sense. Check java.sun.com for JVM tuning for details.
Regards,
Slava Imeshev

R12 Performance Tuning...

Are there any docs or notes that one can to follow for EBS R12 performance tuning ?
Regards,
Rupdipt

Hi,
Please refer to the following links:
Performance Tuning the Apps Database Layer
http://blogs.oracle.com/schan/newsItems/departments/release12/2007/05/17#a1562
Understanding JDBC Connections From the eBusiness Middle Tier
http://blogs.oracle.com/schan/newsItems/departments/release12/2007/09/20#a2051
Performance Tuning for the E-Business Suite
http://blogs.oracle.com/schan/newsItems/departments/release12/2007/12/11#a2297
The following notes/links provide guidelines for setting up JVMs (Jserv groups/OC4J instances) for online applications in Apps 11i and R12:
Note: 362851.1 - Guidelines to setup the JVM in Apps Ebusiness Suite 11i and R12
https://metalink.oracle.com/metalink/plsql/ml2_documents.showDocument?p_database_id=NOT&p_id=362851.1
Note: :462550.1 - Generate JVM heap dump in E-Business suite 11i and R12
https://metalink.oracle.com/metalink/plsql/ml2_documents.showDocument?p_database_id=NOT&p_id=462550.1
Latest JVM Tuning Recommendations for Apps 11i
http://blogs.oracle.com/schan/2007/03/01
Tuning JVMs with Apps 11i
http://blogs.oracle.com/schan/2007/02/16#a1230
Configuring Middle-Tier JVMs for Applications 11i
http://blogs.oracle.com/schan/2006/08/01#a494
For tuning other application components (i.e. CM, initialization parameters, top SQL statements, application tier, ..etc), the following note should be helpful:
Note: 169935.1 - Troubleshooting Oracle Applications Performance Issues
https://metalink.oracle.com/metalink/plsql/ml2_documents.showDocument?p_database_id=NOT&p_id=169935.1
Note: 244040.1 - Oracle E-Business Suite Recommended Performance Patches
https://metalink.oracle.com/metalink/plsql/ml2_documents.showDocument?p_database_id=NOT&p_id=244040.1
Note: 744143.1 - Tuning performance on eBusiness suite
https://metalink.oracle.com/metalink/plsql/ml2_documents.showDocument?p_database_id=NOT&p_id=744143.1
Regards,
Hussein

SUN IDM 8.1.0.3: performance tuning on SPARC platform

Hi all!
In our environment we have SUN IDM 8.1.0.3 on Sun SPARC Enterprise T5120 Server, 4 Core 1.2GHz UltraSPARC T2 processor, 4GB FBDIMM memory (4 1GB). As application server we use Sun GlassFish 2.1. In this configuration we have noticed that there is some performance issue: all IDM web pages are opened very slowly. For example, opening the page “Roles - > List Roles” takes about 20 seconds whereas in the virtual machine with 2.4 GHz x86 CPU and 1 Gb RAM opening this page takes about 3 seconds.
I have performed Java JVM tuning on Sparc machine and has obtained throughput about 98% but the page is still open slow.
I assume that problem is in the GlassFish tuning on Sparc platform.
Does anybody have an experience in the GlassFish tuning on Sparc platform and can give some advice or links how I can do this?
Or maybe anybody has other idea than tuning on Sparc platform which cause this performance issue?
Any ideas are welcome.
Thanks’ in advance!

Where's your database for this located? On the same server or another server? It is more likely the database than Glassfish being the weak and slow link here unless badly tuned GC is taking up all the CPU cycles.

SOA Suite: SoaDistributedCache

Hi all,
we have developed a custom identity store for a soa suite installation because the user information is stored in an external system. This queries are not so performant so we decided to cache the user information using coherence. I have noticed that the SOA Suite has already a cache configured called "SoaDistributedCache". So my question is if it is recommended to use this cache (since our custom identity store is tightly coupled to the soa suite) or if we should define another one.
best regards Nicolas

Hi NJ,
thanks for your contribution. Unfortunately we can't use that approach because we are using Coherence inside a custom OPSS Identity Provider, which has to run on the same JVM of the SOA Suite.
Nevertheless we found another solution. We created a new instance of the DefaultConfirableCacheFactory, loaded our own coherence-cache-factory.xml, put it in a static variable (a kind of Singleton) and referenced it on our IdentityProvider. It seems to work very fine (we are still testing) and we can leave the SOA Coherence configuration unaltered. The only inconvenience from this approach is that we can't use the CacheFactory object in our Identity Provider, because it has the SOA configuration and not ours. Instead we have to reference our static DefaultConfirableCacheFactory object and call the ensureCache method.
cheers Nicolas

IS INCREASING number of JVM = good performance on forms

Hi,
Will increasing number of JVMs make the oracle forms work faster. :). [Release is R12].
Regards
Taher

Taher,
Not necessarily -- See these documents for details.
Note: 362851.1 - Guidelines to setup the JVM in Apps Ebusiness Suite 11i and R12 (Customer Recommended)
Note: 567551.1 - Configuring various JVM tuning parameters for Oracle E-Business suite 11i and R12
Note: 415455.1 - How To Use Jconsole to Monitor JVM with Oracle E-Business Suite 11i10 and R12
Note: 763658.1 - R12: What Is the OAFM JVM Process Used For?
Regards,
Hussein

DB Adapter Performance Problem

Hello,
We are trying to update oracle database using DB Adapter. The insertion into database via DBAdapter (& only with DBAdapter alone) is slow. Even for transferring 50 records of ~1K data, 5-6 seconds are spent.
Environment:
Oracle SOA suite 10.1.3 with 10.1.3.3 Patch Applied
AIX 5
8 CPU & 20 GB RAM
Our test setup.
Tool:ESB
Inbound Adapter to read data from Oracle Table
TransformActivity to convert source schema to destination schema
Outbound Adapter to write data into same oracle table in the same machine. (This has performance problem).
If we are to read data using DB adapter from oracle table & write it to a file using File adapter, transfer of 10,000 records (~2K each) happens in 2 secs.Only writing into database is taking long time. We are unsure why writing into database takes so much. Any help would be appreciated to solve this problem.
We have modified the DB values stated by Oracle documentation for performance improvement. We have done the JVM tuning. We tried using "UsesBatchWriting" & UseDirectSql=true. However there is no improvement.
We also tried creating an outbound adapter which executes custom sql. Custom sql inserts 10000 records into destination table. (insert into dest_table select * from source_table). There is no performance issue with this approach. Customsql is executed in less than 2 seconds. Also we dont see any performance problem if we are to use any of the sql clients to update data in the same destination table. Only via DB Adapter we face this issue.
Please let me know if you would like know setting of any parameter in the system.We would appreciate if any help can be provided to find where the bottleneck is.
Thanks

I'm presuming this is just merge and not insert.
do alter system set sql_Trace=true and capture the trace files on the database . It's probably only waiting on SQLNET message from client but we need to rule that out.
dmstool should show you some of the activity stuff inside the client, it may also be worth doing a truss on the java process to see what syscalls it is waiting on.
Also are you up to MLR7 , the latest ESB release ?

10.1.3.3 ESB DB Adapter Performance Problem

Hello,
We are trying to update oracle database using DB Adapter. The insertion into database via DBAdapter (& only with DBAdapter alone) is slow. Even for transferring 50 records of ~1K data, 5-6 seconds are spent.
Environment:
Oracle SOA suite 10.1.3 with 10.1.3.3 Patch Applied
AIX 5
8 CPU & 20 GB RAM
Our test setup.
Tool:ESB & BPEL
Inbound Adapter to read data from Oracle Table
TransformActivity to convert source schema to destination schema
Outbound Adapter to write data into same oracle table in the same machine. (This has performance problem).
ESB Console shows much of total time spent in the Outbound Adapter activity.
We also created a BPEL process to do the data transfer between Oracle Databases. Adapter statistics for outbound insert activity in BPEL console shows higher value under "Commit" listed under "Adapter Post Processing".
If we are to read data using DB adapter from oracle table & write it to a file using File adapter, transfer of 10,000 records (~2K each) happens in 2 secs.Only writing into database is taking long time. We are unsure why writing into database takes so much. Any help would be appreciated to solve this problem.
We have modified the DB values stated by Oracle documentation for performance improvement. We have done the JVM tuning. We tried using "UsesBatchWriting" & UseDirectSql=true. However there is no improvement.
We also tried creating an outbound adapter which executes custom sql. Custom sql inserts 10000 records into destination table. (insert into dest_table select * from source_table). There is no performance issue with this approach. Customsql is executed in less than 2 seconds. Also we dont see any performance problem if we are to use any of the sql clients to update data in the same destination table. Only via DB Adapter we face this issue.
Interestingly, in a different setup,a Windows machine with just 1CPU, 1GB RAM running 10.1.3 is able to transfer 10,000 records (~2K per record) to a different Oracle database over the network(with in LAN).
Please let me know if you would like know setting of any parameter in the system.We would appreciate if any help can be provided to find where the bottleneck is.
Thanks

I'm presuming this is just merge and not insert.
do alter system set sql_Trace=true and capture the trace files on the database . It's probably only waiting on SQLNET message from client but we need to rule that out.
dmstool should show you some of the activity stuff inside the client, it may also be worth doing a truss on the java process to see what syscalls it is waiting on.
Also are you up to MLR7 , the latest ESB release ?

JVM Tuning in SOA 10.1.3.4

Similar Messages

Maybe you are looking for