Appendix A. Running HATS applications on a System z platform

This appendix describes unique topics when running HATS applications on a System z platform. It is expected that you also use the HATS User's and Administrator's Guide, the HATS Programmer's Guide, and the 2HATS Rich Client Platform Programmer's Guide as sources of information on HATS, and that you review the HATS Readme.

2 2 2

Installation considerations

2

There can be implications installing the .ear file on a z/OS platform. 2If you use WebSphere administrative console and browse to a local copy of the file, it is likely 2that you will receive the following error message:

2
BBOO0271E HTTP REQUEST EXCEEDED 10485760 BYTE INPUT BUFFER 
2

The explanation for this message is:

2
2
BBOO0271E HTTP REQUEST EXCEEDED dstring BYTE INPUT BUFFER
2
2The default buffer is not large enough for HATS applications. It is possible 2to increase this variable, and then use WebSphere administrative console to install the application 2from a local copy. 2
2
2 2 2

Optimizing settings for performance

2
2Note to reviewers 2

BruceG, I added all of this based on the info you supplied from the capacity 2guide. I have a few questions:

2

1. The capacity guide lists Servers?Application Servers?servername?Server 2Instance?Server Instance?Minimum Number of Instances and Servers?Application 2Servers?servername?Server Instance?Server Instance? Maximum Number of Instances. 2I assume Server Instance is only in the path once and not twice as shown above.

2

2. In java heap sizes it says garbage collection should be no more than 215% but in garbage collection it says should be 2%-5%. Which is correct?

2

3. In garbage collection, where do you set -Xgcpolicy=optavgpause?

2

4. In WebSphere variables what does "under Manage WebSphere Variables " 2refer to?

2

You can use the WebSphere administrative console to adjust the following settings and values 2which can enhance the performance of HATS applications running on a z/OS platform:

2 2 2

Workload Profile

2

On a z/OS platform, setting the Workload Profile parameter to LONGWAIT is important for 2application processing that involves sending or receiving information across 2a network. The LONGWAIT setting allows each 2servant region to have a pool of 40 threads to handle concurrent HATS requests. 2The default pool size is 4 threads. This parameter setting is important when 2sizing the number of servant regions.

2

Find this parameter by navigating to Servers>Application 2Servers>servername>ORB Service>Advanced Settings>Workload Profile, where servername represents the name of your server.

2 2

Number of servant regions

2

On a z/OS platform, HATS applications run in the WebSphere Application 2Server servant regions. If you specify LONGWAIT for the Workload Profile parameter, each 2servant region has a pool of 40 threads, . By default, there is only one servant 2region to handle all concurrent HATS requests. You can increase the number 2of servant regions to handle your application workload by setting the following 2parameters. In the parameter locations listed below, servername represents 2the name of your server.

2
2
Multiple Instances Enabled
2
Select this check box to allow multiple servant regions to be initialized. 2

Find this parameter by navigating to Servers>Application 2Servers>servername>Server Instance>Multiple Instances Enabled.

2
2
Minimum Number of Instances
2
This parameter specifies the number of servant regions to be kept running 2once servant regions are initialized during WebSphere Application Server start 2up. Use this parameter when the response time for the workload requires that 2several servant regions are needed to be ready to process work. The default 2for J2EE servers is 1. The maximum value is 20. 2

Find this parameter by 2navigating to Servers>Application Servers>servername>Server Instance>Minimum Number of Instances.

2
2
Maximum Number of Instances
2
This parameter specifies the total number of servant regions allowed 2by workload management to run concurrently in the server's application 2environment. Set this value to limit the number of servant regions created 2by workload management for a server. The default is zero, which means there 2is no limit. 2

Find this parameter by navigating to Servers>Application Servers>servername>Server Instance>Maximum 2Number of Instances.

2
2
2 2

Java heap sizes

2

The Java heap parameters influence performance and the behavior of garbage 2collection. Larger heap sizes take longer to fill and longer to garbage collect. 2With large heap sizes, applications run longer before a garbage collection 2occurs, and it takes a longer time to perform the garbage collection. You should 2monitor garbage collection and adjust your heap sizes so that no more than 215% of the total execution time is spent in garbage collection. A good way 2to see what is going on with garbage collection is to use the JVM's verbose 2garbage collection option which is enabled by adding -verbosegc to 2the JVM arguments. Verbosegc provides an output file that gives the date and 2time of each garbage collection cycle. From the output file, you can calculate 2how much of the total execution time is spent in garbage collection.

2

On a z/OS platform, there is a JVM in both the controller and the servant 2region. Usually, the JVM in the controller does not need to be tuned. This 2information applies to the JVM in the servant region.

2

There are two JVM parameters that specify how much memory is allocated 2for the JVM heap. The settings of these two parameters will affect the performance 2of your HATS applications. In the parameter locations listed below, servername represents the name of your server.

2
2
Initial Heap Size
2
This parameter specifies how much memory is allocated for the heap when 2the JVM starts. Properly tuning this parameter reduces the overhead of garbage 2collection, improving server response time and throughput. For most HATS applications, 2the default setting for this option is too low, resulting in a high number 2of minor garbage collections. This setting will be workload specific, but 2it should be at least 512 MB. If the heap size changes frequently, you might 2improve performance if you specify the same value for the initial and maximum 2heap parameters. 2

Find this parameter by navigating to Servers>Application Servers>servername>Process Definitions>JVM>Initial 2Heap Size.

2
2
Maximum Heap Size
2
This parameter specifies the maximum heap size the Java interpreter 2will use for dynamically allocated objects and arrays. Properly tuning this 2parameter also reduces the overhead of garbage collection, improving server 2response time and throughput. For most HATS applications, the default setting 2for this option is too low, resulting in a high number of minor garbage collections. 2This setting will be workload specific, but it should be at least 768 MB. 2

Find this parameter by navigating to Servers>Application 2Servers>servername>Process Definitions>JVM>Max Heap Size.

2
2
2 2

Garbage collection

2

The JVM memory management function, or garbage collection, provides one 2of the biggest opportunities for improving JVM performance.

2

Garbage collection should normally consume from 2% to 5% of the total execution 2of a properly functioning application. It takes monitoring to understand the 2effects of garbage collection.

2

If you have pause time problems which are denoted by erratic application 2response times which are caused by normal garbage collection, you should set -Xgcpolicy=optavgpause.

2 2

WebSphere variables

2

In the Manage WebSphere Variables section 2of the WebSphere administrative console, navigate to Servers>Application 2Servers>servername>Custom Properties and add the variables 2listed below:

2

2
    2
  1. protocol_http_tcp_no_delay =1 (for non-secure 2connections)
  2. 2
  3. protocol_https_tcp_no_delay =1 (for secure 2connections)
  4. 2
  5. protocol_http_max_persist request=1
  6. 2
  7. protocol_http_timeout_input=0
  8. 2
  9. protocol_http_timeout_output=0
  10. 2
  11. protocol_http_timeout_persistentSession=0
  12. 2
  13. wlm_stateful_session_placement_on=1