Cluster management, a common software infrastructure among technology companies, aggregates compute resources from a collection of physical hosts into a shared resource pool, amplifying compute power and allowing for the flexible use of data center hardware. At Uber, cluster management …

The post Peloton: Uber’s Unified Resource Scheduler for Diverse Cluster Workloads appeared first on Uber Engineering Blog.

Disaster is really inevitable. Even with all the redundant power investments, ThePlanet (formerly EV1 and RackShack), had to shut down their backup generators at their H1 data center on the instructions of the fire crew. This happened after a wire-short in fault transformer led to an explosion that knocked off one of their walls, ultimately bringing 9,000 servers down. Luckily no one was injured.

This just goes on to show that just because a data center has redundant power and backup generators, it does not mean that a disaster cannot happen. IIRC, ThePlanet's last disaster was blamed on backup generators not kicking off properly.

While there was no damage to servers, I wonder how many MyISAM repairs need to be triggered once the servers do come back online?

- …