Evaluating std, crossbeam, disruptor for inter-thread communication based on Blog Post
We use 2 producer threads and 1 consumer thread in MPSC & 1 producer thread and 1 consumer thread in SPSC setup.
My Blog post: https://venkat.eu/the-power-of-mechanical-sympathy-in-software-engineering#heading-lmax-disruptor
On my 13-inch, M1, 2020, 16 GB MacBook Pro:
$ make run-all-optimized
cargo run --release --bin main -- --lib std
Finished `release` profile [optimized] target(s) in 1.75s
Running `target/release/main --lib std`
Running std demo:
SPSC Sum: 10000000, processed time: 140
MPSC Sum: 20000000, processed time: 555
cargo run --release --bin main -- --lib crossbeam
Finished `release` profile [optimized] target(s) in 0.01s
Running `target/release/main --lib crossbeam`
Running crossbeam demo:
SPSC Sum: 10000000, processed time: 347
MPSC Sum: 20000000, processed time: 249
cargo run --release --bin main -- --lib disruptor
Finished `release` profile [optimized] target(s) in 0.01s
Running `target/release/main --lib disruptor`
Running disruptor demo:
SPSC Sum: 10000000, processed time: 183
MPSC Sum: 20000000, processed time: 686
disruptor lib official benchmark: See disruptor results
A simple throughput benchmark using criterion create as recommended here
As of May 6, 2024:
SPSC: Throughput: std > disruptor > crossbeam
MPSC: Throughput: std > mpsc > disruptor
$ make bench
Benchmarking std_spsc: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 12.6s, or reduce sample count to 30.
std_spsc time: [126.08 ms 127.48 ms 129.76 ms]
Found 10 outliers among 100 measurements (10.00%)
6 (6.00%) high mild
4 (4.00%) high severe
Benchmarking crossbeam_spsc: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 35.9s, or reduce sample count to 10.
crossbeam_spsc time: [319.63 ms 326.99 ms 334.20 ms]
Found 6 outliers among 100 measurements (6.00%)
1 (1.00%) low severe
4 (4.00%) low mild
1 (1.00%) high severe
Benchmarking disruptor_spsc: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 25.2s, or reduce sample count to 10.
disruptor_spsc time: [291.63 ms 314.71 ms 337.12 ms]
Benchmarking std_mpsc: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 57.4s, or reduce sample count to 10.
std_mpsc time: [581.07 ms 589.71 ms 598.16 ms]
Found 7 outliers among 100 measurements (7.00%)
3 (3.00%) low severe
1 (1.00%) low mild
1 (1.00%) high mild
2 (2.00%) high severe
Benchmarking crossbeam_mpsc: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 29.4s, or reduce sample count to 10.
crossbeam_mpsc time: [264.37 ms 274.54 ms 286.64 ms]
Found 3 outliers among 100 measurements (3.00%)
1 (1.00%) high mild
2 (2.00%) high severe
Benchmarking disruptor_mpsc: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 80.8s, or reduce sample count to 10.
disruptor_mpsc time: [782.81 ms 792.41 ms 801.15 ms]
Found 20 outliers among 100 measurements (20.00%)
12 (12.00%) low severe
3 (3.00%) low mild
2 (2.00%) high mild
3 (3.00%) high severe