This patch set adds how-to-use documentation for Watcher Overload standard deviation algorithm. Change-Id: I75d7cd0ff8507ca70efb6d9668ae9fbf651a7f97
2.7 KiB
Strategy name
Synopsis
display name:
goal:
Add here a complete description of your strategy
Requirements
Metrics
- Write here the list of metrics required by your strategy algorithm (in the form
-
of a table). If these metrics requires specific Telemetry plugin or other additional software, please explain here how to deploy them (and add link to dedicated installation guide).
Example:
metric | service name | plugins | comment |
---|---|---|---|
compute.node.* vm.cpu.utilization_perc |
none none |
one point every 60s |
|
power | ceilometer | kwapi | one point every 60s |
Cluster data model
Default Watcher's cluster data model.
or
- If your strategy implementation requires a new cluster data model, please
-
describe it in this section, with a link to model plugin's installation guide.
Actions
Default Watcher's actions.
or
- If your strategy implementation requires new actions, add the list of Action
-
plugins here (in the form of a table) with a link to the plugin's installation procedure.
action | description |
---|---|
action1 | This action1 ... |
action2 | This action2 ... |
Planner
Default Watcher's planner.
or
- If your strategy requires also a new planner to schedule built actions in time,
-
please describe it in this section, with a link to planner plugin's installation guide.
Configuration
If your strategy use configurable parameters, explain here how to tune them.
Efficacy Indicator
Add here the Efficacy indicator computed by your strategy.
Algorithm
Add here either the description of your algorithm or link to the existing description.
How to use it ?
$ Write the command line to create an audit with your strategy.
External Links
- If you have written papers, blog articles .... about your strategy into Watcher,
-
or if your strategy is based from external publication(s), please add HTTP links and references in this section.