Closed Loop Server Compliance test results

The following steps were followed to execute each of these scenarios for the Closed Loop Server Compliance workflow:

1. Installed RSCD agents on 100 virtual machines (targets).
   • Fifty were installed on Red Hat Enterprise Linux 4.0 32-bit VM targets.
   • Fifty were installed on Red Hat Enterprise Linux 5.0 32-bit VM targets.
2. Created the rest of the targets to scale by creating aliases.
3. Added all the target servers to BMC BladeLogic Application Server.
4. Created a component template with four rule groups each having five rules.
5. Executed a Compliance job on 500, 800, 1500, 2000, 3000, 4000, 5000, 6000, 7000, and 8000 targets having 1 percent noncompliant targets each, using a component template with 20 rules.

The following KPIs were observed:

• Time taken by BMC BladeLogic Application Server in running the Compliance job.
• Time taken to receive an SNMP trap at BMC Atrium Orchestrator.
• Time taken to parse the BladeLogic export log. This value was calculated by determining the difference between:
  • Time snapshot of log on request to BMC Remedy ITSM
  • CSV file creation time
• Time taken to create all the change requests in BMC Remedy ITSM.
• Total time taken in execution of the workflow. This value was calculated by determining the difference between:
  • Time when BladeLogic started executing the Compliance job
  • Allocation time of last noncompliant change requests

The following results show the time taken to execute each activity by BMC BladeLogic and BMC Atrium Orchestrator server.

Total time taken to execute the complete tests

Figure 1 shows the total time taken to execute the complete test (that is, from the start of the BladeLogic Compliance job to the creation of the last change request for the 500, 800, 1500, 2000, 3000, 4000, 5000, 6000, 7000, and 8000 target tests).

Figure 1: Total test time

Time taken to execute the BladeLogic Compliance job for each test

Figure 2 shows the time that the BladeLogic Compliance job took to complete for the 500, 800, 1500, 2000, 3000, 4000, 5000, 6000, 7000, and 8000 target tests, with 1 percent noncompliant targets.

Figure 2: BladeLogic Compliance job execution times

Time taken to parse the BladeLogic export log

On completion of the BladeLogic Compliance job, the export log is created by the BladeLogic server and contains information about the compliant and noncompliant targets. Figure 3 shows the time taken by the BMC Atrium Orchestrator server to parse the export log for the 500, 800, 1500, 2000, 3000, 4000, 5000, 6000, 7000, and 8000 target tests.

Figure 3: BladeLogic export log parsing times

Figure 3 shows that the export log parsing time increases exponentially even though the targets increase linearly.

Time taken to create change requests in BMC Remedy ITSM

Figure 4 shows the time taken by BMC Atrium Orchestrator to create the change requests in BMC Remedy ITSM for the 500, 800, 1500, 2000, 3000, 4000, 5000, 6000, 7000, and 8000 target tests with 1 percent noncompliant targets (5, 8, 15, 20, 30, 40, 50, 60, 70, and 80 respectively).

Figure 4: Change request processing times

Comparison of all times/charts

Figure 5 shows the comparative time taken by each test (500, 800, 1500, 2000, 3000, 4000, 5000, 6000, 7000, and 8000 target tests having 1 percent noncompliant targets — 5, 8, 15, 20, 30, 40, 50, 60, 70, and 80 respectively).

Figure 5 shows that the BladeLogic Compliance job execution time and the change request creation time are minimal compared to the parsing time. The majority of the time is spent parsing the file created upon execution of the job.

Figure 5: Consolidated chart