Collecting data: Questions for performance issues
Merative ™ Social Program Management is now Cúram ™ by Merative™
Collecting data: Specific questions for Merative ™ Social Program Management performance issues
Question & Answer
Question
What questions should be answered when reporting a performance issue with Merative Social Program Management?
Answer
When reporting a performance issue in Merative Social Program Management you must provide details about the topology of your system and any recent changes to it, observations about the environment in which the problem occurs and operational data from the running application.
Before you send Merative Support any performance related information, your local database administrator and application server administrator must have reviewed the information and made observations which point to an Merative Social Program Management issue.
Providing this performance related information will help clarify the problem and save time analyzing the issue.
In addition to this performance related information you must also provide the minimum information required for every support case which is described in "Collecting Data for Merative Social Program Management issues"
Full details of the performance related information required to be sent to Merative Support are listed below:
Topology
Answer the questions below to provide an overview of the topology of your system:
- How many clusters are there?
- What are the hardware specifications on these clusters (CPUs, memory etc)?
- What is the topology of these clusters?
- What is the topology of associated hosts (Deployment manager, IHS, Database etc)?
- Provide Operating System Information on each relevant machine or node in the configuration, where machine_name uniquely identifies each machine or node.
Document Information
More support for:
Merative Social Program Management
Software version:
All Version(s)
Operating system(s):
AIX, HP-UX, Linux, Solaris,Windows, z/OS
Modified date:
28 January 2020
| Platform | Commands to gather Operating System Information | Command to gather current process and CPU information |
|---|---|---|
Microsoft Windows |
systeminfo > systeminfo_machine_name.txt \ msinfo32 /report c:\Temp\SystemInfo_machine_name.txt |
Take a snapshot of Process Explorer or Task Manager (order by CPU usage highest first) |
AIX |
oslevel > oslevel_machine_name.txt \ uname -a > uname_machine_name.txt \ prtconf > prtconf_machine_name.txt |
topas >topas.out vmstat 10 >vmstat.out (or any output file name) |
UNIX (Linux) |
uname -a > uname_machine_name.txt | UNIX (Linux) - top >top.out and vmstat 10 |
Solaris |
uname -a > uname_machine_name.txt |
prstat >prstat.out (or any output file name) |
Recent Changes to your System
- What recent changes have been made to the environment e.g. software, hardware, firewall, configuration, etc?
- Was there a change in the number of users around the time the issue was first identified?
- Have there been changes to the number of users in the past 6 months?
- How many users is the system scaled for?
- If a change in volume please review your ongoing performance measurement and monitoring data; if performance was acceptable previously then is your hardware or tuning now inadequate for the new volume?
- Has there been a change to the usage of the system since the performance issue was identified, e.g. new business processes or initiative to clear application backlog? If yes, then please describe the change.
- Describe any engagement to date with local subject matter experts and/or support services for the database, application server, etc. in use. If not applicable, why not?
Environmental Factors
If the issue manifests itself during specific times of the day, describe those times in general, but most specifically in terms of user load at those times or particular business practices that might occur. Examples could include, "Monday afternoons are used for client interviews which results in a lot of evidence being updated at this time".
Are the affected users running other software/applications at the time the issue is experienced?
Is the issue isolated to specific offices/floors or consistent across all offices? If isolated to specific locations, describe:
- the commonalities between all locations experiencing the issue (network bandwidth, machines in use, etc.)
- the differences between locations experiencing and not experiencing the issue
Performance Measurement Information
- Has a performance baseline been established in this area? If so, when was the baseline created?
- Quantify the change in performance as measured against your baseline.
- Describe the variance in user performance measured against maximum acceptable response times since the performance issue was identified.
- If there is no related baseline then please advise how the degradation in performance is being measured.
JMX Statistics
Follow the instructions in the documentation at "Access to the JMX statistics" to collect JMX statistics for your system.
Profiler Snapshots, Thread Dump, Heap Dump, GC Logs
If Profiler Snapshot, Thread Dump, Heap Dump, and GC Logs were collected at the time the issue was observed, please provide them. Screenshots from your chosen heap analyzer tool could also greatly help with problem diagnosis. e.g. screenshot of tree view of heap
Webserver Logs
If webserver logs demonstrating the issue are available, please provide these.
Related Information