Course Description

Before you can troubleshoot and view reports about performance, you must collect data about monitored resources. Performance monitors are data collection jobs that gather performance information about resources. This information includes metrics that measure the performance of the components within a resource. Metrics measure the performance characteristics of volumes, ports, and disks on storage systems

Performance monitoring and analysis are critical to deciphering the often-complex behavior of applications. They help identify regions of code that are most frequently executed, thus allowing fine-tuning of application behavior for optimal runtime speed and resource usage. Elimination of various inefficiencies either introduced by the programmer or arising due to less than perfect mapping to a specific execution environment is vital to achieving acceptable application execution rates on a supercomputer and increasing the users' productivity

Troubleshooting can be a tedious process after we collect the information on performance of application, or an issue occurred in application. There as several problem-solving steps to follow that will help ease the stress of solving nearly any IT issue:


1)  Identification of an Issue
Getting to the bottom of a computer issue can sometimes feel like playing 20 questions, so it’s crucial to ask the right questions first if you want to discover the root problem, The key is to find out the “who, what, and when” of the problem, like does the issue affect just one person or many people? Did it occur right after a significant event , etc.

2) Gather more details, eliminate variables
In calypso application logs we have log categories to provide more information around the issue that has occurred. So we need to gather Error messages, Events logs, run utilities with help of Calypso TA to collect evidence to raise a request with helpdesk.

3) Reproduce the problem, develop hypothesis of root cause
Reproducing an issue simply means verifying that you can recreate the same error the user reported. This requires a complete set of instructions and repeat them as performed by users , some times we may require the technical and functional expertise to address the issue in such case reach out to calypso TA and BA with relevant details to get guidance to resolve.

4) Fix based on findings
With all the details in hand we can now work towards resolving it or providing a workaround to address the issue or error encountered by users

This eLearning course outline Calypso performance and troubleshooting techniques, performance is critical to application and it starts from the time we are selecting hardware/software for implementation and will travel along with life cycle of application. Especially in fast growing world every application needs to perform better and same is the case with Calypso, it provides some of the best in class techniques, methods and metrics to achieve better performance. Troubleshooting arise in case of any issue that is not resolved by a simple config change, in case of issue with code change there are numerous ways to address it by following the steps mentioned above, however here in this course we will discuss about some the Calypso out of box utilities to help in troubleshooting application.

Calypso Learning Services

Course curriculum

  • Troubleshooting

    • Introduction

    • How to write a simple performance related HD Call

    • JSTACK Mechanics

    • Environment Management

    • Preparing Calypso Systems For Performance P2s

    • Minimal Installation Scripts

    • Installation Checksum

  • Resource

    • Implementation Guidelines

  • Assessment

    • Test Your Understanding

  • Appendix

    • Troubleshooting-Presentation