Can you give an example of a bottleneck that was identified through profiling with pprof?

One example of a bottleneck identified using pprof was a situation with excessive memory allocation in a loop processing large volumes of data. Profiling showed that the function responsible for parsing incoming messages created many temporary objects, which were quickly collected by the garbage collector, causing delays.

The solution involved optimizing the parsing algorithm: instead of creating new structures on each call, a sync.Pool object pool was used, significantly reducing GC load and improving overall performance.

Example of using sync.Pool:

var bufPool = sync.Pool{
    New: func() interface{} {
        return make([]byte, 1024)
    },
}

func parseMessage(data []byte) {
    buf := bufPool.Get().([]byte)
    defer bufPool.Put(buf)
    // Use buf for processing, avoiding new memory allocation
}

This approach helped reduce response time and memory consumption.