Child pages
  • Allocating Disk Space for Fault Tolerance and Availability

This is documentation for a previous version of ClustrixDB. Documentation for the latest version can be found here

Skip to end of metadata
Go to start of metadata

To enable the cluster to recover from a node failure without intervention, you must maintain sufficient free disk space. To calculate the amount of space required, use the following formula:

Recovery Threshold % = (N - 1) * 95 / N

For example, for a four-node cluster to maintain full cluster operation and fault tolerance in the event of a node failure, N=71.25% is the threshold. This number improves as your cluster gets larger:

Recovery Threshold (N=6) = (N-1) * 95 / N = 79.17
Recovery Threshold (N=10) = (N-1) * 95 / N = 85.50
Recovery Threshold (N=20) = (N-1) * 95 / N = 90.25

If your applications write data at a rate that fills the cluster aggressively, use a value that is more conservative than 95% in your calculations, to ensure that data can continue to be written while you are waiting for a replacement node to join the cluster and for data redistribution to complete.

To examine the disk space usage of your cluster, use the Clustrix  CLX utility.