Performance testing is essential to ensure that when the system goes into production it performs adequately under anticipated loads. Broadly speaking, our performance testing goals are focused on ensuring:
- The system is stable under load
- The system can scale to handle larger volumes in the future
- The system responds quickly enough
- The infrastructure (hardware, software, network) is capable of supporting the system
Performance Testing Metrics:
The Performance Testing process involves deciding on the Measurement types for the live application and then categorizing the metrics related to the application components. here are some of the key metrics that we capture as part of our performance testing effort:
- Concurrent Usage – measure the number of virtual users active at a given point of time (Concurrent users).
- ART (Average Response Time) – calculate the time the request started until the time to complete the same. Response times can be measured either as Time To First Byte or Time To Last Byte.
- Peak Response Time – measure the longest round trip duration of a request-and-response cycle.
- Throughput – measure the bandwidth consumed during the test.
- Error Rate – calculated as a percentage of the requests that errored out against the total number of requests.