go-pool - A generic pool implementation in Go

A lightweight, generic object pool implementation in Go

Overview

go-pool is a generic, efficient object pool for Go designed to manage reusable, expensive-to-create objects such as virtual machines, interpreters, or database connections.

Instead of repeatedly allocating and destroying large objects, you can use this pool to acquire, reuse, and release instances safely and efficiently.

Features

• Generic API – works with any object type
• Optimized for performance and low contention
• Simple, idiomatic interface
• Supports timeouts for acquiring objects
• Optional helper for automatic acquire/release management

Use Cases

• Embedding scripting engines (Lua, JS, etc.)
• Managing reusable network connections or sessions
• Pooling pre-initialized workers or buffers
• Any case where object creation is expensive

Usage Example

package main

import (
	"log"
	"time"

	"github.com/epikur-io/go-pool"
)

type PoolEntry struct{}

func (pe *PoolEntry) DoSomeWork() {
	log.Println("Do some work...")
	time.Sleep(time.Second * 1)
}

func main() {
	factory := func() *PoolEntry {
		return &PoolEntry{}
	}
	pool := pool.NewPool(10, factory)

	// get an entry:
	entry := pool.Acquire()
	entry.DoSomeWork()

	// release entry
	pool.Release(entry)

	// get an entry or timeout after 1 second:
	entry, err := pool.AcquireWithTimeout(time.Second * 1)
	if err != nil {
		log.Fatalln("error:", err)
	}

	entry.DoSomeWork()

	// release entry, since entry is nil, a new entry will be created and put into the pool
	pool.Release(nil)
}