Swarm Dispatch (spawn-collect-die)
A coordinator spawns N ephemeral workers, collects outputs, terminates them. Workers are short-lived and non-communicating.
Verified against arxiv.org
A swarm-dispatch coordinator spawns N ephemeral workers for a job, collects their outputs, and terminates them, a fan-out and fan-in that exists only for the task. Unlike a standing colony the workers are disposable, which keeps state simple and cost bounded to the job. It suits embarrassingly-parallel work: search a space or process partitions, then reduce. In anatomy terms it is a coordinator core spawning short-lived scoped cores.
A coordinator core spawns N short-lived worker cores (map), collects and reduces their outputs (reduce), then discards them; workers hold no durable state; the perimeter gates the coordinator's writes.
- ·Embarrassingly-parallel processing: partition, process, reduce.
- ·Fan-out search across many candidates.
- ·Batch jobs where workers are disposable.
- ·Bounded cost: workers live only for the job.
- ·Simple state: nothing durable in the workers.
- ·Scales to the width of the fan-out.
- ·Clean lifecycle: spawn, collect, die.
- ·The coordinator is a bottleneck and single point of failure.
- ·Spawn and teardown overhead per job.
- ·No cross-worker learning or memory.
- ·Only fits decomposable, parallel work.
Description, use cases and trade-offs: Field Book analysis, synthesised from established agent-architecture literature.
The corpus does not rewrite vendor documentation. Where copy is quoted verbatim it is marked as such and attributed to its source.