Skip to main content

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.

The gap comes from scheduling overhead, uneven partitions, serialization, and shared resources like object storage.

Performance & optimizationDistributed compute

Related terms

  • 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.
  • Long tail: The situation where most tasks finish quickly but one or two run much longer, so the whole run waits on them while workers sit idle.
  • Compute-bound: A workload whose runtime is limited by how fast calculations can be performed. Adding workers to a compute-bound job generally makes it finish sooner.
  • Profiling: Measuring where a program actually spends its time, before deciding what to optimize.