Skip to main content

Design

This diagram illustrates the workflow of a ProflingExperiment.

morphlingui

  • A user submits a ProflingExperiment via a RPC or front-end UI interface, specifying the ML model, tunable configuration parameters, optimization objectives, and sampling budgets.

  • Within the sampling budget, Morphling iteratively communicates with the algorithm server to get the next configuration for sampling.

  • Then Morphling starts a Trial to evaluate that sampling.

  • For each Trial, a model serving inference instance Deployment is launched, and its “readiness” is reported to trigger a client-side RPS stress-test Job.

  • After the client Job completes, the measured peak RPS is stored in the DB.

  • A Trial finishes, and the result is sent to the ProflingExperiment.

  • The ProflingExperiment completes when the sampling budget is reached.

The sequence diagram of the ProflingExperiment workfolow is shown as follows:

morphlingui