Difference between revisions of "RemnantServers"

Latest revision as of 23:30, 17 February 2013

• Launchpad Entry: OpenstackSpec:foo
• Created:
• Contributors:
• Packages affected:

Summary

Nova compute will support running transient servers that are launched when capacity is available, and terminated without notice when capacity is required for standard servers.

Release Note

Existing servers and storage should continue to function as normal - in a cloud environment where requested remnant servers fill up all remaining capacity, it is important to configure nova such that nodes do not become oversubscribed beyond the SLA of the standard servers.

Rationale

In a cloud environment using converged hardware (such as NASA's Nebula), it's not feasible to turn off underutilized compute nodes, since they are also functioning as storage nodes. The most energy-efficient approach is to keep all the nodes busy, all the time. We can achieve this by providing a second tier of virtual server services, with a 0% uptime SLA.

User stories

In a science computing environment, a default remnant job might function as a typical grid computing node, and consume batch jobs from a globus or condor system.

Requests for remnant capacity are made using the standard API calls.

Remnant servers will be terminated without notice. (A future enhancement would provide a webhooks callback on termination.)

Remnant servers will be launched when available.

Standard API metadata, available from within the remnant server, will provide access to a count of the number of remnant servers currently running for a specific job.

It is the job of the Scheduler to select which remnant request(s) should be running at any given time. (This is pluggable, and can support arbitrary rules regarding quota, priority, etc.)

This feature is similar to 'Spot Instances'.

Assumptions

Design

You can have subsections that better describe specific parts of the issue.

Implementation

This section should describe a plan of action (the "how") to implement the changes discussed. Could include subsections like:

UI Changes

Should cover changes required to the UI, or specific UI that is required to implement this

Code Changes

Code changes should include an overview of what needs to change, and in some cases even the specific details.

Migration

Include:

• data migration, if any
• redirects from old URLs to new ones, if any
• how users will be pointed to the new way of doing things, if necessary.

Test/Demo Plan

It's important that we are able to test new features, and demonstrate them to users. Use this section to describe a short plan that anybody can follow that demonstrates the feature is working. This can then be used during testing, and to show off after release. Please add an entry to http://testcases.qa.ubuntu.com/Coverage/NewFeatures for tracking test coverage.

This need not be added or completed until the specification is nearing beta.

Unresolved issues

This should highlight any issues that should be addressed in further specifications, and not problems with the specification itself; since any specification with problems cannot be approved.

BoF agenda and discussion

Use this section to take notes during the BoF; if you keep it in the approved spec, use it for summarising what was discussed and note any options that were rejected.