A benchmark is a performance test of hardware or software. A benchmark simulates real-world application workloads. These models of real applications can then be run against the system being evaluated. Models work better than the actual application and offer reproducible results. The real application, on the other hand, has too many variables that can change the results over several test sessions.
Industry standard benchmarks represent real application workloads, their run results depend on workload definition (user load, latency, etc.), as well as server configuration and tuning parameters. During the running of a benchmark, deadlock errors can occur. This is not an issue with Benchmark Factory, but more of an issue with the tuning of the workload as well as the system-under-test. For example, it would be unrealistic to expect no errors from a workload that runs 1000 users updating a single row on a small table. To further troubleshoot any tuning issues, we recommend the Quest Spotlight® products. Learn more at: http://www.quest.com/
Benchmarks measure performance that analyze:
Benchmark Factory provides the following benchmark tests:
|Benchmark||Benchmark Version (TPC)|