Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 7 Next »

1.Overview

To better support more HPA capabilities without modifying the implementation code for Multi-cloud. so we need to extract the HPA discovery part of the code make it into many drivers(HPA Discovery Driver). if we want to add new HPA feature or modify the implementation code for the currently feature, we just need to modify HPA Discovery Driver code.


Extract the original HPA discovery part code and make it into a driver.
we need to add hook in Openstack, Kubernetes and others plugin. then we develop driver for various languages.

We plan to develop python language and GO language(in R6 support python language).


2.Driver code Management

For python language driver, we put it into multi-cloud openstack plugin.

For golang language driver, we put it into multi-cloud kubernetes plugin.

3.Add New hardware feature driver

  1. The hardware vendor determines whether to share the driver to community.
  2. If vendor share the driver to community, then submitted the driver code to multi-cloud repo, then we  will put the driver to the docker image and release it.
  3. if vendor didn't share the driver to community, he/she can make new image base on the releasing docker image and upload to local docker image repo, then change the oom helm charts file to point this image repo.

           # Dockerfile

           from multi-windriver:1.4.0

           add hpa-discovery-driver1(device vendor1)

           add hpa-discovery-driver2(device vendor2)

4.Driver Version Control

We will put HPA discovery drivers as pip package to release pip REPO. name like onap-device-driver-<drivername>

4.1 Construct package:

python setup.py compile


4.2 upload package

python


5. Usage

For python, run below command in openstack plugin

$ python setup.py

6. Test

Unit Tests with tox

  • No labels