Scheduling overhead
The fixed cost of dispatching a task and returning its result. When it approaches the cost of the task body, adding parallelism stops helping.
A fixed cost per task, which is why task size matters more than task count. A task taking a millisecond spends most of its life being scheduled; batching a hundred of them into one changes the ratio decisively.
Performance & optimizationDistributed compute
Related terms
- Batching: Grouping several small units of work into one task so that scheduling overhead does not dominate the time spent doing useful work.
- Partition: One independently processable chunk of a dataset. Partition size is the main lever on how well a workload parallelizes.
- Scheduler: The component on the head node that decides which compute node runs each task, honoring the resources that task requested.
- Speedup: How many times faster a distributed run is than the same work on one worker. Ideal speedup equals worker count; real speedup falls short of it.