OPPerTune

Post-Deployment Configuration Tuning of Services Made Easy

Real-world application deployments have hundreds of inter-dependent configuration parameters, many of which significantly influence performance and efficiency. With today’s complex and dynamic services, operators need to continuously monitor and set the right configuration values (configuration tuning) well after a service is widely deployed. This is challenging since experimenting with different configurations post-deployment may reduce application performance or cause disruptions.

While state-of-the-art ML approaches do help to automate configuration tuning, they do not fully address the multiple challenges in end-to-end configuration tuning of deployed applications.

We present OPPerTune, a service that enables configuration tuning of applications in deployment at Microsoft. OPPerTune reduces application interruptions while maximizing the performance of deployed applications as and when the workload or the underlying infrastructure changes. It automates three essential processes that facilitate post-deployment configuration tuning:

  1. Determining which configurations to tune
  2. Automatically managing the scope at which to tune the configurations, and
  3. Using a novel reinforcement learning algorithm to simultaneously and quickly tune numerical and categorical configurations, thereby keeping the overhead of configuration tuning low.
diagram, schematic
OPPerTune service architecture – Applications create tuning instances on the server to tune various configuration parameters.
Autoscoper helps automatically create, manage, and scope tuning instances based on the application’s dynamic context.
Selector helps pick the most promising configuration parameters to tune.