File: README-dgm.md

package info (click to toggle)
golang-github-couchbase-moss 0.0~git20170914.0.07c86e8-4
  • links: PTS, VCS
  • area: main
  • in suites: bullseye, buster, sid
  • size: 664 kB
  • sloc: python: 230; makefile: 37
file content (54 lines) | stat: -rw-r--r-- 1,502 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
# Instructions for DGM testing for moss

dgm_moss_test.go is a moss component level test that exercises moss.

Included in the test is a simulator of the moss herder memory stalling
functionality.

The test can be configured to generate a Results*.json file that can be
used with python script graph/dgm-moss-ploy.py to plot the resutls.

## Steps

Here is an example shell script to exercise the test.

```
#!/bin/bash
findDisk()
{
    dev=`df ${MossStore} | grep -v Filesystem | awk {'print $1'}`
    dirName=`dirname ${dev}`
    devName=`basename ${dev} ${dirName}`
    echo ${devName}
}

clearCaches()
{
    sudo sync
    sudo sh -c "echo 1 > /proc/sys/vm/drop_caches"
    sudo sh -c "echo 2 > /proc/sys/vm/drop_caches"
    sudo sh -c "echo 3 > /proc/sys/vm/drop_caches"

}

MossStore="/mossstore/MossStoreTest"
mkdir -p -m=0777 ${MossStore}
diskMonitor=`findDisk`
runDesc="${gitInfo}"

go test -v -test.run TestMossDGM -test.timeout 99999s -runTime 5m -memQuota 4gb -keyLength 48 -valueLength 48 -keyOrder random -numWriters 1 -writeBatchSize 100000 -writeBatchThinkTime 1ms -numReaders 16 -readBatchSize 100 -readBatchThinkTime 1ms -outputToFile -dbPath ${MossStore} -diskMonitor ${diskMonitor} -runDescription ${runDesc} -dbCreate
```

## Ploting a graph of the results

The python script requires numpy, pandas, matploglib

```
python dgm-moss-plot.py Results_...json
```

This will process a Results...png file.

## Notes

This test only runs properly on Linux as it gathers stats from /proc.