Serialization
Converting Python objects into bytes so they can be sent to another machine. Large task arguments are serialized and copied to every worker that needs them.
A frequent hidden cost: passing a large object as an argument to many tasks can move more data than the computation saves.
Distributed computePerformance & optimization
Related terms
- Task: One unit of distributed work: a single call to a function decorated with the distribute decorator. Tasks are stateless, so each call starts fresh.
- Transfer cost: The time spent moving data between machines, or between host and GPU memory. Often the reason an operation is not worth accelerating.