Performance
GraphQLade comes with a few built-in performance optimizations. Other than that, performance should be optimized at resolver level (e.g. data loader pattern) or transport level (e.g. compression, streamed body parsing).
Parser cache
GraphQLade caches parse and validation results in an LRU (least recently used) cache
with a cache size of 50 by default.
The cache size can be configured via parserOptions.cacheSize.
Benchmarks
Take these with a grain of salt, as the methodology is probably flawed. Updated July 1st 2024.
graphqlade@1.9.0 + Node.js http
Running 20s test @ http://localhost:3000/graphql?query={hello}
10 connections
┌─────────┬──────┬──────┬───────┬──────┬─────────┬─────────┬───────┐
│ Stat │ 2.5% │ 50% │ 97.5% │ 99% │ Avg │ Stdev │ Max │
├─────────┼──────┼──────┼───────┼──────┼─────────┼─────────┼───────┤
│ Latency │ 0 ms │ 0 ms │ 1 ms │ 1 ms │ 0.07 ms │ 0.47 ms │ 55 ms │
└─────────┴──────┴──────┴───────┴──────┴─────────┴─────────┴───────┘
┌───────────┬─────────┬─────────┬─────────┬─────────┬─────────┬──────────┬─────────┐
│ Stat │ 1% │ 2.5% │ 50% │ 97.5% │ Avg │ Stdev │ Min │
├───────────┼─────────┼─────────┼─────────┼─────────┼─────────┼──────────┼─────────┤
│ Req/Sec │ 12,951 │ 12,951 │ 17,199 │ 17,583 │ 16,882 │ 1,025.81 │ 12,948 │
├───────────┼─────────┼─────────┼─────────┼─────────┼─────────┼──────────┼─────────┤
│ Bytes/Sec │ 1.94 MB │ 1.94 MB │ 2.58 MB │ 2.64 MB │ 2.53 MB │ 154 kB │ 1.94 MB │
└───────────┴─────────┴─────────┴─────────┴─────────┴─────────┴──────────┴─────────┘
Req/Bytes counts sampled once per second.
# of samples: 20
338k requests in 20.02s, 50.6 MB read
graphqlade@1.9.0 + express@4.19.2
Running 20s test @ http://localhost:3000/graphql?query={hello}
10 connections
┌─────────┬──────┬──────┬───────┬──────┬────────┬─────────┬───────┐
│ Stat │ 2.5% │ 50% │ 97.5% │ 99% │ Avg │ Stdev │ Max │
├─────────┼──────┼──────┼───────┼──────┼────────┼─────────┼───────┤
│ Latency │ 1 ms │ 1 ms │ 3 ms │ 4 ms │ 1.3 ms │ 0.93 ms │ 59 ms │
└─────────┴──────┴──────┴───────┴──────┴────────┴─────────┴───────┘
┌───────────┬────────┬────────┬─────────┬─────────┬──────────┬────────┬────────┐
│ Stat │ 1% │ 2.5% │ 50% │ 97.5% │ Avg │ Stdev │ Min │
├───────────┼────────┼────────┼─────────┼─────────┼──────────┼────────┼────────┤
│ Req/Sec │ 3,705 │ 3,705 │ 6,207 │ 6,299 │ 5,845.55 │ 725.17 │ 3,704 │
├───────────┼────────┼────────┼─────────┼─────────┼──────────┼────────┼────────┤
│ Bytes/Sec │ 971 kB │ 971 kB │ 1.63 MB │ 1.65 MB │ 1.53 MB │ 190 kB │ 970 kB │
└───────────┴────────┴────────┴─────────┴─────────┴──────────┴────────┴────────┘
Req/Bytes counts sampled once per second.
# of samples: 20
117k requests in 20.02s, 30.6 MB read
graphql-yoga@5.6.0
Running 20s test @ http://localhost:3000/graphql?query={hello}
10 connections
┌─────────┬──────┬──────┬───────┬──────┬─────────┬─────────┬────────┐
│ Stat │ 2.5% │ 50% │ 97.5% │ 99% │ Avg │ Stdev │ Max │
├─────────┼──────┼──────┼───────┼──────┼─────────┼─────────┼────────┤
│ Latency │ 0 ms │ 0 ms │ 3 ms │ 5 ms │ 0.69 ms │ 2.78 ms │ 160 ms │
└─────────┴──────┴──────┴───────┴──────┴─────────┴─────────┴────────┘
┌───────────┬────────┬────────┬─────────┬─────────┬─────────┬──────────┬────────┐
│ Stat │ 1% │ 2.5% │ 50% │ 97.5% │ Avg │ Stdev │ Min │
├───────────┼────────┼────────┼─────────┼─────────┼─────────┼──────────┼────────┤
│ Req/Sec │ 3,103 │ 3,103 │ 7,647 │ 10,495 │ 7,619.8 │ 2,060.51 │ 3,103 │
├───────────┼────────┼────────┼─────────┼─────────┼─────────┼──────────┼────────┤
│ Bytes/Sec │ 611 kB │ 611 kB │ 1.51 MB │ 2.07 MB │ 1.5 MB │ 406 kB │ 611 kB │
└───────────┴────────┴────────┴─────────┴─────────┴─────────┴──────────┴────────┘
Req/Bytes counts sampled once per second.
# of samples: 20
152k requests in 20.04s, 30 MB read
