Home > Storage > PowerFlex > White Papers > Elastic Cloud on Kubernetes on Dell PowerFlex 4.0 > Benchmark methodology
The Elasticsearch development team uses Rally to run regular nightly and release benchmarks early during development. Rally is also used to tune Elasticsearch and benchmark against an existing Elasticsearch cluster, manage benchmark configurations, run and compare results, and capture potential performance results. For more information, see Rally 2.8.0 documentation.
A race in Rally is the execution of a benchmarking experiment. You can choose different benchmarking scenarios (called tracks) for your benchmarks.
The following figure shows the fundamental blocks of the Rally:
Figure 13. Rally functional overview
For this whitepaper, some of the prevalent use cases of Elastic are run that demonstrate PowerFlex performance.
Use case 1: This use case is rally-eventdata-track, a rally track for simulating event-based data use cases. The track supports bulk indexing of autogenerated events and simulated Kibana queries and a range of management operations to make the track self-contained. For more information, see rally-eventdata-track.
Use case 2: This use case is http_logs, a rally track based on web server logs from the 1998 Football World Cup. For more information, see http_logs-tracks.
Use case 3: This use case is Elastic security track. This track simulates security workloads and supports bulk-indexing and simulated queries. For more information, see elastic-security-tracks.