This is an implementation of the redis-benchmark in Go. This was done as a proof-of-concept to see if Go could achieve
similar performance to the C implementation. As a proof-of-concept, only a subset of the reference implementation's
functionality was programmed (i.e. just the PING command) though it should be relatively easy to extend.
As a caveat to inferring too much from these results, please read Pitfalls and misconceptions about redis benchmarking.
| Implementation | Requests/sec | Elapsed (s) | Mem (Kb) | CPU (%) | Notes |
|---|---|---|---|---|---|
| Go | 247905.40 | 0:04.03 | 10520 | 194 | GOMAXPROCS = 2, Go 1.9 |
| Go | 219339.87 | 0:04.56 | 18052 | 316 | GOMAXPROCS = 8, Go 1.9 |
| Reference (C) | 214454.20 | 0:04.73 | 22428 | 99 | |
| Go | 162336.18 | 0:06.16 | 8380 | 99 | GOMAXPROCS = 1, Go 1.9 |
| Node.js | 88983.80 | 0:11.33 | 72580 | 101 | Node.js 8.5.0 |
- Median value of 10 runs, lower elapsed time is better
- The benchmark implementation was run on the same machine as the server.
- Command:
$ time -f "Command: %C\nElapsed: %Es\nMEM (RSS): %Mkb\nCPU: %P\n" redis-benchmark -n 1000000
System
$ lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 8
On-line CPU(s) list: 0-7
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 1
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 94
Model name: Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz
Stepping: 3
CPU MHz: 1082.453
CPU max MHz: 3500.0000
CPU min MHz: 800.0000
BogoMIPS: 5183.91
Virtualisation: VT-x
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 6144K
Memory
$ free -m
total used free shared buff/cache available
Mem: 15901 6657 3525 217 5719 8494
Swap: 16239 0 16239
Redis Server: 4.0.2 64 bit