What is Fuzzball Federate and why use it?

Gregory Kurtzer gives a one-minute explanation of Fuzzball Federate. Federate bridges multiple Fuzzball clusters together, so an organization can have a cluster in each of several geographically separated data centers as well as clusters in different cloud availability zones, all presented as one.

Federate decides where a job should run based on resource and architecture availability, cost, and where execution makes the most sense, while data locality, mobility, gravity, and security policies also influence placement. The result is that users no longer have to think about which cluster their work is running on. It is a quick primer for anyone hearing about Federate for the first time.

Key takeaways

• Fuzzball Federate bridges multiple Fuzzball clusters, whether in separate data centers or in different cloud availability zones.
• Job placement decisions weigh resource and architecture availability, cost, and where a job makes the most sense to run.
• Data locality, mobility, gravity, and security policies also influence where Federate sends a job.
• Users no longer need to think about which cluster their jobs are actually running on.

Questions this video answers

How does Fuzzball Federate decide where a job runs?

Federate looks at resource and architecture availability across the bridged clusters, considers cost and where it is cheapest and most sensible to run, and factors in data locality, mobility, gravity, and security policies. Users submit work without needing to know which data center or cloud cluster ultimately executes it.

This video is part of the Fuzzball playlist. Browse every CIQ video by product and topic.