Netflix runs thousands of microservices and thousands of backend data store nodes on Amazon's Web Services (AWS) instances to serve and store data for more than 100M users. Hence, we are not always aware of the performance implications that bundled microservices may generate on our backend systems. Hence, we architected and developed NDBench. Netflix Data Benchmark (NDBench) is a pluggable cloud-enabled benchmarking tool that can be used across any NoSQL or SQL data store system. NDBench aids in simulating the performance benchmark by mimicking several production use cases. NDBench allows for dynamically changing the benchmark configurations while a test is running, hence perform tests along with our production microservices. NDBench is well integrated with common platform cloud services such as dynamic configuration management, load balancing services, and metrics. NDBench allows us to run infinite horizon tests to identify potential memory leaks from long running processes or garbage collection issues, or to test long running maintenance jobs. In this talk, we will cover our NDBench deployment, and use cases. Note: This will talk will be given along with Vinay Chella.
