Services Flow Chart
The following diagram shows end-to-end performance testing. If test execution fails then system tuning (changes to application configuration or code) will be carried out, this will follow retest execution.
Planning
– A number of workshops or meetings will be carried out and the output will be a Test Plan or Strategy document. All documents are tailored for the given client. Test Plan can consist of the following information:
– Test Objectives & Non-Functional Requirements
– Risk Assessment
– Volumetric analysis
– Test Approach – Test Scripts and scenarios to be developed
– Physical & Logical system diagrams
– Test deliverables
– Application & Server monitoring approach
– Test tool approach or PoC
– Schedule with several cycles of test execution and at least one tuning cycle
– Entry & exit criteria
– RAID log – risks, assumptions, issues & dependencies
Preparation
– Setup test infrastructure consisting of several load injectors
– Creation of test scripts in LoadRunner, J-Meter or Gatling. LoadRunner is preferred
– Creation of test scenarios
– Setup of application & server monitoring
– Create test run log for recording; throughput, response times and server statistics
Execution
- – Execution of test scenarios; Peak Load, Stress & Soak. Volume, scalability and other scenarios will be considered if necessary.
- – Collect & analyse loadRunner results.
- – Collect & analyse application, server monitoring
- – Carry out tuning if performance is not acceptable
- – Repeat several cycles of retests until performance is acceptable.
- – Produce interim test results after each cycle
- – Update project team with regular or weekly reports to update status.
Reporting
– Analysis of results and monitoring data.
– Produce report consisting of the following:
– Recommendations
– Key Findings
– Non-Functional requirements with RAG status
– Throughput & response times analysis
– All system tuning & cycles executed
– Server & application statistics analysis
