Analyze Application Server Performance

App Server Performance Analytics provides you with key information and findings that let you quickly and effectively analyze the performance of your application servers.

When you go to the application from the Oracle Management Cloud home page or the corresponding dashboard, you can view information for the following:

• Garbage Collection Issues: You can identify your server applications that are over the garbage collection limit. A long garbage collection period could affect the amount of time server requests have to wait in the queue before they’re fulfilled by the application server. The threshold limit for garbage collection is set by the application, and is not user-defined. The application presents you with a finding based on the collected data. The following is an example of an application finding:

  12 have JDK version 1.8.0_60.
  23 have increasing Request Rate

  According to this finding, based on the collected data, 12 of your application servers that are over the garbage collection limit have the JDK version 1.8.0_60, which might be causing problems. Also, 23 of your application servers have an increasing request rate.

• Heap Issues: You can view the application servers with a heap usage above a specified percentage. The application provides you with insights or findings based on the available data. The following is an example of an application insight:

  21 have Heap Size above 40GB.
  23 have an increasing Request Rate

  According to this finding, based on the collected data, 21 of your application servers show a high heap size of above 40 GB. Also, 23 of your application servers have an increasing request rate.

• Availability: You can view the critical events across your application servers. If no server is down, then the application lets you know that there are no critical events across your application servers.

• Memory Issues: In some cases, issues with host performance can affect the performance of your application server.

• Workload Performance: You can drill down to view information specific to the performance of selected servers. The following is an example of an application finding related to workload performance:

  12pm-1pm has the highest number of request 7.7M
  Thursday has the highest number of request 30.3M

  According to this finding, based on the collected data, the peak hour with the highest number of requests (7.7 M requests) was at 12 p.m. to 1 p.m.. Also, the day with the highest number of requests (30.3 M requests) was Thursday.

• Distribution Analysis: You can view information about the percentage of time spent in garbage collection and the total JDBC connection waits.

• Inventory Analysis: From here, you can view information about the JDK version, operating system vendor or version, and the XMX size across your application servers.

You can decide to view information for all your application servers or only for specific servers. You can also select the time period for which you want the application to load the information.