Mirroristas/etcd
2
0
Fork 0
mirror of https://github.com/etcd-io/etcd.git synced 2024-09-27 06:25:44 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #14111 from patrocinio/check_perf_doc

Browse Source 
Doc: provide more clarify to the usage of check perf command
This commit is contained in:
Benjamin Wang 2022-06-18 05:07:26 +08:00 committed by GitHub
commit 43434afd48
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 22 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
etcdctl/README.md
View File

@ -376,7 +376,7 @@ Prints the compacted revision.
### WATCH [options] [key or prefix] [range_end] [--] [exec-command arg1 arg2 ...]
Watch watches events stream on keys or prefixes, [key or prefix, range_end) if range_end is given. The watch command runs until it encounters an error or is terminated by the user. If range_end is given, it must be lexicographically greater than key or "\x00".
Watch watches events stream on keys or prefixes, [key or prefix, range_end) if range_end is given. The watch command runs until it encounters an error or is terminated by the user. If range_end is given, it must be lexicographically greater than key or "\x00".
RPC: Watch
@ -1505,6 +1505,26 @@ CHECK provides commands for checking properties of the etcd cluster.
CHECK PERF checks the performance of the etcd cluster for 60 seconds. Running the `check perf` often can create a large keyspace history which can be auto compacted and defragmented using the `--auto-compact` and `--auto-defrag` options as described below.
Notice that different workload models use different configurations in terms of number of clients and throughtput. Here is the configuration for each load:
| Load | Number of clients | Number of put requests (requests/sec) |
|---------|------|---------|
| Small | 50 | 10000 |
| Medium | 200 | 100000 |
| Large | 500 | 1000000 |
| xLarge | 1000 | 3000000 |
The test checks for the following conditions:
- The throughput should be at least 90% of the issued requets
- All the requests should be done in less than 500 ms
- The standard deviation of the requests should be less than 100 ms
Hence, a workload model may work while another one might fail.
RPC: CheckPerf
#### Options

2
etcdctl/ctlv3/command/check.go
View File

@ -126,7 +126,7 @@ func NewCheckPerfCommand() *cobra.Command {
}
// TODO: support customized configuration
cmd.Flags().StringVar(&checkPerfLoad, "load", "s", "The performance check's workload model. Accepted workloads: s(small), m(medium), l(large), xl(xLarge)")
cmd.Flags().StringVar(&checkPerfLoad, "load", "s", "The performance check's workload model. Accepted workloads: s(small), m(medium), l(large), xl(xLarge). Different workload models use different configurations in terms of number of clients and expected throughtput.")
cmd.Flags().StringVar(&checkPerfPrefix, "prefix", "/etcdctl-check-perf/", "The prefix for writing the performance check's keys.")
cmd.Flags().BoolVar(&autoCompact, "auto-compact", false, "Compact storage with last revision after test is finished.")
cmd.Flags().BoolVar(&autoDefrag, "auto-defrag", false, "Defragment storage after test is finished.")