Worker Pool

Frame includes a worker pool backed by ants with retry scheduling and a typed job API.

Quick Start

job := workerpool.NewJob("job-1", func(ctx context.Context, pipe workerpool.JobResultPipe[string]) error {
    return pipe.WriteResult(ctx, "ok")
})

_ = frame.SubmitJob(ctx, svc, job)
result, _ := job.ReadResult(ctx)

Configure

Frame configures a worker pool using ConfigurationWorkerPool:

• CPU factor
• pool capacity
• pool count
• expiry duration

You can override with WithWorkerPoolOptions and workerpool.Option values:

• WithPoolCount
• WithSinglePoolCapacity
• WithConcurrency
• WithPoolExpiryDuration
• WithPoolNonblocking
• WithPoolPreAlloc
• WithPoolPanicHandler
• WithPoolLogger
• WithPoolDisablePurge

Retry Behavior

Jobs support retries and exponential backoff. The manager schedules retries via an internal queue. Maximum retry runs and backoff are enforced.

API Highlights

• workerpool.NewManager(ctx, cfg, stopErr, opts...)
• Manager.GetPool()
• workerpool.SubmitJob(ctx, manager, job)
• workerpool.NewJob(id, func)

Best Practices

• Keep job functions small and idempotent.
• Use typed results to avoid serialization overhead.
• Monitor retry rates to detect systemic failures.