GoVPP/Performance

From fd.io
< GoVPP
Revision as of 15:04, 11 July 2017 by Raszabo (Talk | contribs)

Jump to: navigation, search

GoVPP Performance

GoVPP performance has been measured using the perf-bench example binary on an Ubuntu 16.04 LTS virtual machine running on a laptop with Intel® Core™ i7-6820HQ CPU @ 2.70GHz and 16 GB of RAM. The virtual machine has been assigned all CPU cores and 8 GB of RAM.

The benchmark application is able to measure the performance of both synchronous and asynchronous binary API. As the results of the benchmark show; in case that speed is an issue, asynchronous API is recommended:

Asynchronous API
1 C application (vppapiclient) - control ping 762 000 req/second
2 GoVPP - control ping 251 878 req/second
3 GoVPP - l2fib add 245 560 req/second
4 GoVPP - interface dump 210 305 req/second
Synchronous API
5 C application - control ping 2340 req/second
6 GoVPP - control ping 107 req/second

As the results show, there is quite big difference between the performance of a C test application and the GoVPP. In order to identify the difference and bottlenecks, we:

  • performed profiling of the execution of the test 2 (asynchronous GoVPP) and 6 (synchronous GoVPP).
  • performed the measurements from the bare Go application that called vppapiclient library directly, without any further processing in Go, with following results:
Bare Go
7 Go async - no encode & decode, callback once per 100 replies 761 836 req/second
8 Go async - no encode & decode 554 215 req/second
9 Go async - with encode & decode, no message passing 284 283 req/second
10 Go async - with encode & decode, with message passing 250 861 req/second