Distributed Cache Failover

If a single cache server goes down or gets disconnected from the network, you will get an error until the cache server comes back up, and the application will not function as expected.

If a cache server in a fault tolerant cache setup (presuming at least two backup servers) goes down, the failure will not impact the system except for a temporary slowdown. In case of a clustered distributed cache architecture, the following mechanism is provided for:

Wait and Retry

When a “fetch” operation on the cache fails with an exception, a Wait and Retry approach is used to access the cache. If the cache is not available after a predefined number of times, it is deemed a sustained cache failure and the application shuts down.

Backup Policy

The Distributed cache service allows you to configure the number of backups; If any of the cluster nodes fail, as long as you have one or more backups, there will be no data loss.

Death Detection

When the death or departure of a cluster member is automatically and quickly detected, failover occurs very quickly and transparently and the responsibilities of the failed cluster member can be assumed by the cluster.