This presentation provides an overview of the IBM ECM Performance Methodology for building and maintaining high performance and scalability in complex real-world ECM deployments, including performance planning, pre-production performance testing, performance tuning, monitoring and maintaining performance, and performance troubleshooting. IBM Content Navigator and IBM Case Manager are used as models to illustrate the concepts, but the Methodology can be applied to any Enterprise Content Management system.
This presentation provides an overview of the IBM ECM Performance Methodology for building and maintaining high performance and scalability in complex real-world ECM deployments. The IBM ECM Performance Methodology includes
- Planning for performance
- Pre-production performance testing
- Performance tuning
- Monitoring and maintaining performance
- Performance troubleshooting
In this presentation IBM Content Navigator and IBM Case Manager are used as models to illustrate the concepts, but the Methodology can be applied to any Enterprise Content Management system.
Overview of topics covered
Planning for performance
- Start with an initial sizing and capacity planning estimate
- Design your solution with performance in mind
- Model and test your own expected workload and system configuration
- Create a "performance profile" as a baseline for production monitoring and capacity planning
- Thorough pre-production testing increases confidence and reduces risk for the production system
- Evaluate end-user performance under realistic conditions
- Test types: low-load baseline, longevity at typical production load, expected peak load, stress test, scalability test
- "Cold / warm / hot" end-user testing
- Strategies for Rational Performance Tester scripting and test repository population
- The tuning process is similar to the performance troubleshooting process
- Performance tuning can touch all components in the system
- Common aspects to tune for IBM Case Manager and the FileNet Content Platform Engine repository
- Maintain an ongoing "performance profile"
- The Green / Yellow / Red Zone approach to monitoring
- Use a disciplined and organized approach to solve the problem
- Troubleshooting checklists
- Key system metrics for performance troubleshooting
Was this topic helpful?
17 June 2018