Tag

crush Archives - AWS Managed Services by Anchor

Pulling apart Ceph’s CRUSH algorithm

By | Technical | One Comment

As you’ve probably noticed, we’ve been evaluating Ceph in recent months for our petabyte-scale distributed storage needs. It’s a pretty great solution and works well, but it’s not the easiest thing to setup and administer properly. One of the bits we’ve been grappling with recently is Ceph’s CRUSH map. In certain circumstances, which aren’t entirely clearly documented, it can fail to do the job and lead to a lack of guaranteed redundancy. How CRUSH maps work The CRUSH map algorithm is one of the jewels in Ceph’s crown, and provides a mostly deterministic way for clients to locate and distribute data on disks across the cluster. This avoids the need for an index server to coordinate reads and writes. Clusters with index servers, such as the MDS in Lustre, funnel…

Read More