You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@libcloud.apache.org by "Tomaz Muraus (JIRA)" <ji...@apache.org> on 2013/11/20 23:43:35 UTC

[jira] [Commented] (LIBCLOUD-444) Support scaling APIs

    [ https://issues.apache.org/jira/browse/LIBCLOUD-444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13828227#comment-13828227 ] 

Tomaz Muraus commented on LIBCLOUD-444:
---------------------------------------

Usually the work-flow when creating / proposing a new API is something like this:

1. Spent a good deal of time researching and understanding the existing APIs.
2. Determine a common set of functionalists.
2. Prepare a proposal for the base API, send it to the mailing list for comments.
3. Write a prototype driver for at least two different providers (the more the better). Two are the minimum because if you only write one it's really easy to make it biased against a single provider.

> Support scaling APIs
> --------------------
>
>                 Key: LIBCLOUD-444
>                 URL: https://issues.apache.org/jira/browse/LIBCLOUD-444
>             Project: Libcloud
>          Issue Type: New Feature
>          Components: Core
>            Reporter: Brian Curtin
>              Labels: features
>
> I propose to kick off a new package in the libcloud namespace, e.g., libcloud.scaling, to support the various automatic scaling APIs that some providers are offering.
> * Amazon has had Auto Scaling available for some time now.
> * Rackspace released its Autoscale product to general availability today.
> * Azure supports Autoscale via its Management API.
> * I've found some Google documentation around autoscaling, but it seems more tied to AppEngine than GCE, so maybe not applicable here.
> At least Amazon and Rackspace's offerings are built on or make use of their CloudWatch and Cloud Monitoring products, respectively, so we'd probably want to create a package, e.g., libcloud.monitoring, as well. Azure
> I have to imagine the first step is getting to understand the various APIs and boiling down a common set of goals and functionality. Before going down that path, is there anything I should know?



--
This message was sent by Atlassian JIRA
(v6.1#6144)