Testing ReadMemStats vs pprof.Lookup
This is a really simple webserver designed to test the performance of ReadMemStats vs pprof.Lookup. It exposes a couple of endpoints that both generate a long, sorted slice of float64s. It then reports on the allocated memory one of two ways.
This repo carries the Genuine "Written on an Airplane" seal of approval.
Why
I want to look into how much recording heap utilization will impact the performance of my webservers. ReadMemStats is fast, but stops the world. pprof.Lookup is about 100x slower, but shouldn't stop the world.
How
This exposes three endpoints:
localhost:8080/readMem0/-- Control. Doesn't read memory.localhost:8080/readMem1/-- ReadMemStatslocalhost:8080/readMem2/-- pprof.Lookup
I tested this using https://github.com/tsenart/vegeta .
echo "GET http://localhost:8080/readMem2/" | vegeta attack -duration 10s | vegeta report
First pass results
Control
Requests [total, rate] 500, 50.10
Duration [total, attack, wait] 9.981021439s, 9.980178389s, 843.05µs
Latencies [mean, 50, 95, 99, max] 1.076732ms, 991.45µs, 1.565569ms, 1.955077ms, 7.243589ms
Bytes In [total, mean] 9866887, 19733.77
Bytes Out [total, mean] 0, 0.00
Success [ratio] 100.00%
Status Codes [code:count] 200:500
Error Set:
ReadMemStats
Requests [total, rate] 500, 50.09
Duration [total, attack, wait] 9.982931537s, 9.982067082s, 864.455µs
Latencies [mean, 50, 95, 99, max] 1.077519ms, 966.665µs, 1.586833ms, 1.97482ms, 6.859381ms
Bytes In [total, mean] 9866533, 19733.07
Bytes Out [total, mean] 0, 0.00
Success [ratio] 100.00%
Status Codes [code:count] 200:500
Error Set:
pprof.Lookup
Requests [total, rate] 500, 50.09
Duration [total, attack, wait] 9.983083109s, 9.981228359s, 1.85475ms
Latencies [mean, 50, 95, 99, max] 1.637835ms, 1.557003ms, 2.705042ms, 3.478394ms, 7.55065ms
Bytes In [total, mean] 9866530, 19733.06
Bytes Out [total, mean] 0, 0.00
Success [ratio] 100.00%
Status Codes [code:count] 200:500
Error Set: