CycleCloud is the cornerstone of your high performance computational grid, so including plans for keeping CycleCloud running in the event of a disaster is important. This guide describes various disaster recovery options for CycleCloud, and provides you with a template disaster recovery script to place into your organization’s disaster recovery guide.
You currently have two options when it comes to disaster recovery approaches for your CycleCloud installation:
This option requires separate hardware to host a cold instance of CycleCloud, and second license for CycleCloud on the secondary hardware. You mirror the hot CycleCloud instance to the cold hardware and keep the automatic backups from CycleCloud in sync on a fairly regular basis. In the event of a disaster you restore the cold instance from a recent backup and bring it to life.
Running CycleCloud under the control of virtualization software allows you to take snapshots of CycleCloud at regular intervals. You copy the snapshots to remote hardware and, in the event of a disaster, resume the CycleCloud virtual machine on secondary hard using the most recent snapshot.