gammay
gammay

Reputation: 6215

Non-distributed data structure in Hazelcast

The Hazelcast datastructures are by default distributed (replicated/accessible by any node in the cluster). Is it possible to make a datastructure as local or non-distributed, if desired so?

Context: We have the need to cache some data in-memory. It doesn't really need to be distributed but local to the node where it is inserted. We could choose not to use Hazelcast but the language datastructures directly (such as java.util.List). However, we would like to make use of the capabilities that Hazelcast offers, such as time-to-live.

Is it possible to specify if a datastructure should not be distributed?

Upvotes: 1

Views: 502

Answers (1)

noctarius
noctarius

Reputation: 6094

This is normally not a Hazelcast usecase, so even if you use a one-node-cluster you probably won't get the expected performance, since Hazelcast always serializes data. You might want to have a look at Google Guava as it might be better suited.

Upvotes: 1

Related Questions