Question 1

What is faster in Go: strings.Builder or bytes.Buffer for concatenation?

Accepted Answer

Based on the benchmarks, strings.Builder is slightly faster, at 2.415 ns/op compared to bytes.Buffer at 3.775 ns/op on an M2 Max. Both are vastly superior to simple string concatenation, which runs at 34768 ns/op due to allocations.

Question 2

How to benchmark concurrency patterns in Go?

Accepted Answer

Use the concurrency_counter benchmarks as a reference, which compare mutexes, atomic operations, channels, and worker pools. Clone the repository and run `go test -bench .` in that directory to see metrics tailored to your hardware.

Question 3

Is regexp faster than strings.Contains for substring checks?

Accepted Answer

No, strings.Contains is much faster, at around 4.874 ns/op, while regexp.MatchString can be over 700 ns/op due to compilation and matching overhead. Pre-compiled regexp helps but still lags behind direct string methods.

Question 4

What hash function is fastest for non-cryptographic use in Go?

Accepted Answer

CRC32 is among the fastest at 245.2 ns/op for 2048-byte data, making it suitable for checksums. For cryptographic needs, SHA256 is faster at 858.6 ns/op but slower than non-cryptographic options.

Question 5

How does embedded file performance compare to disk reads?

Accepted Answer

Embedded files are drastically faster, with benchmarks showing near-zero time (0.3151 ns/op) versus disk reads at about 11737 ns/op, ideal for static assets to reduce I/O latency.

Question 6

Are int keys faster than string keys in Go maps?

Accepted Answer

Yes, int keys are significantly faster, with map int index at 19.06 ns/op versus map string index at 46.07 ns/op, due to the overhead of string hashing and comparison in lookups.

golang-benchmarks

What is golang-benchmarks?

Overview

Use Cases

Best For

Related Projects

Found a gem we're missing?

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions