[
https://issues.apache.org/jira/browse/WHIRR-290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13076295#comment-13076295
]
Karel Vervaeke edited comment on WHIRR-290 at 8/2/11 4:26 PM:
--------------------------------------------------------------
Here's a quick-n-dirty attempt at determining what would be needed to make this
work.
* A way to describe which volumes are needed for each template (size, device
node and mount point). (e.g. in the templates?)
* A provider mechanism to create (and manage) EBS volumes.
* A way to control which volumes are attached to which instances when
restarting (~/.whirr/{cluster-name}/volumes ?) We need control over which
volume is attached to which instance.
* A way to handle hostnames (think of a way to have a invariable public address
or update hadoop/hbase config on boot)? Work with DNS providers (route53,
dyndns, ...). Can Elastic IP help?
jclouds: http://code.google.com/p/jclouds/issues/detail?id=194 ?
was (Author: karel1980):
Here's a quick-n-dirty attempt at trying determine what would be needed to
make this work.
* A way to describe which volumes are needed for each the templates (size,
device node and mount point). (e.g. in the templates?)
* A provider mechanism to create (and manage) EBS volumes.
* A way to control which volumes are attached to which instances when
restarting (~/.whirr/{cluster-name}/volumes ?) We need control over which
volume is attached to which instance.
* A way to handle hostnames (think of a way to have a invariable public address
or update hadoop/hbase config on boot)? Work with DNS providers (route53,
dyndns, ...). Can Elastic IP help?
jclouds: http://code.google.com/p/jclouds/issues/detail?id=194 ?
> Add support for EBS-backed Hadoop clusters
> ------------------------------------------
>
> Key: WHIRR-290
> URL: https://issues.apache.org/jira/browse/WHIRR-290
> Project: Whirr
> Issue Type: New Feature
> Components: service/hadoop
> Reporter: Tom White
>
> An EBS-backed cluster can shutdown compute instances when the cluster is not
> being used, then restart later and continue where it left off.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
