Mozilla is testing out using a small cluster of Ganeti VMs to manage failover for Hadoop/HBase admin services such as NameNode, SecondaryNameNode, JobTracker, and HBase Master.
This isn't an extremely high traffic cluster, and so far, it has been working very well.
Also, the Mozilla IT team has been using it in a few other systems, but I don't know the details of those.
This isn't an extremely high traffic cluster, and so far, it has been working very well.
Also, the Mozilla IT team has been using it in a few other systems, but I don't know the details of those.