Hazelcast comes with connectors for the following data structures:
It’s possible to use the data structures that are part of the same Hazelcast cluster, and share the same memory and computation resources with running jobs. For a more in-depth discussion on this topic, see the In-Memory Storage.
Accessing Data Structures on Remote Clusters
With the Jet API, you can also read from or write to data structures from remote Hazelcast clusters.
| With SQL, you can only access local maps. |
When reading or writing to remote data structures, Hazelcast internally creates a client using the supplied configuration and will create connections to the other cluster.
ClientConfig cfg = new ClientConfig();
cfg.setClusterName("cluster-name");
cfg.getNetworkConfig().addAddress("node1.mydomain.com", "node2.mydomain.com");
Pipeline p = Pipeline.create();
p.readFrom(Sources.remoteMap("inputMap", cfg));
...This client uses the embedded Hazelcast version to connect to the remote cluster. Hazelcast server and client versions are backward and forward compatible within the same major version.