|
|
The following sections describe how to tune Java Virtual Machine (JVM) garbage collection performance for engine tier servers:
Production installations of WebLogic SIP Server generally require extremely small response times (under 50 milliseconds) for clients at all times, even under peak server loads. A key factor in maintaining brief response times is the proper selection and tuning of the JVM’s Garbage Collection (GC) algorithm for WebLogic SIP Server instances in the engine tier.
Whereas certain tuning strategies are designed to yield the lowest average garbage collection times or to minimize the frequency of full GCs, those strategies can sometimes result in one or more very long periods of garbage collection (often several seconds long) that are offset by shorter GC intervals. With a production SIP Server installation, all long GC intervals must be avoided in order to maintain response time goals.
The sections that follow describe GC tuning strategies for JRockit and Sun’s JVM that generally result in best response time performance.
If you use custom startup scripts to start WebLogic SIP Server engines and replicas, simply edit those scripts to include the recommended JVM options described in the sections that follow.
The BEA Configuration Wizard also installs default startup scripts when you configure a new domain. These scripts are installed in the BEA_HOME/user_projects/domains/domain_name/bin directory by default, and include:
If you use the BEA-installed scripts to start engines and replicas, you can override JVM memory arguments by first setting the USER_MEM_ARGS environment variable in your command shell.
| Note: | Setting the USER_MEM_ARGS environment variable overrides all default JVM memory arguments specified in the BEA-installed scripts. Always set USER_MEM_ARGS to the full list of JVM memory arguments you intend to use. For example, when using the Sun JVM, always add -XX:MaxPermSize=128m to the USER_MEM_ARGS value, even if you only intend to change the default heap space (-Xms, -Xmx) parameters. |
JRockit provides several monitoring tools that you can use to analyze the JVM heap at any given moment, including:
Use these and other tools in a controlled environment to determine the effects of JVM settings before you use the settings in a production deployment. See the BEA WebLogic JRockit 1.4.2 SDK Documentation for more information about JRockit and JRockit profiling tools.
The following sections describe suggested starting JVM options for use with the JRockit JVM. The JRockit JVM is available in a standard edition (included with WebLogic SIP Server) or as part of BEA WebLogic Real Time, available separately. The standard JRockit JVM can be tuned to provide high throughput and low response times using the information in Using JRockit without Deterministic Garbage Collection.
The version of JRockit included with BEA WebLogic Real Time uses a deterministic garbage collector, and is recommended when extremely low latency is the overriding requirement of your application. The deterministic garbage collector attempts to maximize throughput, but places the highest priority on guaranteeing short, predictable pause times. This focus on guaranteeing short pause times may result in lower overall throughput when compared to the standard JRockit JVM. See Using JRockit with Deterministic Garbage Collection (WebLogic Real Time) for more information.
When using BEA’s JRockit JVM without deterministic garbage collection (the version included with WebLogic SIP Server), the best response time performance is obtained by using the generational concurrent garbage collector.
The full list of example startup options for an engine tier server are:
-Xms1024m -Xmx1024m -Xgc:gencon -XXnosystemgc -XXtlasize:min=3k -XXkeeparearatio=0 -Xns:48m
| Note: | Fine tune the heap size according to the amount of live data used by deployed applications. |
The full list of example startup options for a replica server are:
-Xms3072m -Xmx3072m -Xgc:gencon -XXnosystemgc -XXtlasize:min=3k -XXkeeparearatio=0 -Xns:48m
Very short response times are most easily achieved by using JRockit’s deterministic garbage collector, which is available with the WebLogic Real Time product. See JVM Tuning for Real-Time Applications in the WebLogic Real Time documentation for basic information about tuning with deterministic garbage collection.
BEA recommends using the following JVM arguments for engine tier servers in replicated cluster configurations:
-Xms1024m -Xmx1024m -XgcPrio:deterministic -XpauseTarget=30ms -XXnosystemgc
| Note: | You may need to increase the -XpauseTarget value for allocation-intensive applications. The value can be decreased for smaller applications under light loads. |
| Note: | Adjust the heap size according to the amount of live data used by deployed applications. As a starting point, set the heap size from 2 to 3 times the amount required by your applications. A value closer to 3 times the required amount generally yields the best performance. |
For replica servers, use the arguments:
-Xms3072m -Xmx3072m -XgcPrio:deterministic -XpauseTarget=30ms -XXnosystemgc
These settings fix the heap size and enable the dynamic garbage collector with deterministic garbage collection. -XpauseTarget sets the maximum pause time and -XXtlasize=3k sets the thread-local area size. -XXnosystemgc prevents System.gc() application calls from forcing garbage collection.
When using Sun’s JDK, the goal in tuning garbage collection performance is to reduce the time required to perform a full garbage collection cycle. You should not attempt to tune the JVM to minimize the frequency of full garbage collections, because this generally results in an eventual forced garbage collection cycle that may take up to several full seconds to complete.
The simplest and most reliable way to achieve short garbage collection times over the lifetime of a production server is to use a fixed heap size with the default collector and the parallel young generation collector, restricting the new generation size to at most one third of the overall heap.
The following example JVM settings are recommended for most engine tier servers:
-server -Xmx1024m -XX:MaxPermSize=128m -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:+UseTLAB -XX:+CMSIncrementalMode -XX:+CMSIncrementalPacing -XX:CMSIncrementalDutyCycleMin=0 -XX:CMSIncrementalDutyCycle=10 -XX:MaxTenuringThreshold=0 -XX:SurvivorRatio=256 -XX:CMSInitiatingOccupancyFraction=60 -XX:+DisableExplicitGC
For replica servers, use the example settings:
-server -Xmx3072m -XX:MaxPermSize=128m -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:+UseTLAB -XX:+CMSIncrementalMode -XX:+CMSIncrementalPacing -XX:CMSIncrementalDutyCycleMin=0 -XX:CMSIncrementalDutyCycle=10 -XX:MaxTenuringThreshold=0 -XX:SurvivorRatio=256 -XX:CMSInitiatingOccupancyFraction=60 -XX:+DisableExplicitGC
The above options have the following effect:
-XX:+UseTLAB—Uses thread-local object allocation blocks. This improves concurrency by reducing contention on the shared heap lock.-XX:+UseParNewGC—Uses a parallel version of the young generation copying collector alongside the concurrent mark-and-sweep collector. This minimizes pauses by using all available CPUs in parallel. The collector is compatible with both the default collector and the Concurrent Mark and Sweep (CMS) collector.-Xms, -Xmx—Places boundaries on the heap size to increase the predictability of garbage collection. The heap size is limited in replica servers so that even Full GCs do not trigger SIP retransmissions. -Xms sets the starting size to prevent pauses caused by heap expansion.-XX:MaxTenuringThreshold=0—Makes the full NewSize available to every NewGC cycle, and reduces the pause time by not evaluating tenured objects. Technically, this setting promotes all live objects to the older generation, rather than copying them.-XX:SurvivorRatio=128—Specifies a high survivor ratio, which goes along with the zero tenuring threshold to ensure that little space is reserved for absent survivors.
|