Hazelcast example in Eta

Note that this example is not using the Direct Java Interop feature and I've manually written the FFI imports for now. A better example is yet to come ;)

Hazelcast is a distributed data system that supports high scalability and data distribution in a clustered environment.

In this example we will create a cluster member, distributed Map to share Map data among the cluster nodes and create a client to connect and query data in the cluster.

Maven Dependencies

The maven dependencies for Hazelcast Cluster Member and Hazelcast Java Client are added in the eta-hazelcast.cabal file.

maven-depends:       com.hazelcast:hazelcast:3.10
...
maven-depends:       com.hazelcast:hazelcast:3.10, com.hazelcast:hazelcast-client:3.10

Running the server

1. Fire up the terminal and enter the commands:

$ git clone https://github.com/Jyothsnasrinivas/eta-hazelcast.git
$ cd eta-hazelcast
$ etlas run hazelcast-server

2. When we start the hazelcast-server application, we can see the flowing text in the console.

Members {size:1, ver:1} [
	Member [192.168.0.104]:5701 - 564aa798-70a8-48d8-8ef9-c0590adb853b this
]

3. To create multiple nodes we can start the multiple instances of hazelcast-server. That means open up another terminal instance and run the command etlas run hazelcast-server simultaneously with the first instance. You will notice the following log in the console.

Members {size:2, ver:2} [
	Member [192.168.0.104]:5701 - 564aa798-70a8-48d8-8ef9-c0590adb853b this
	Member [192.168.0.104]:5702 - e2d32f12-b7af-4991-8c9d-3b8f85f197a9
]

Running the client

1. Fire up another terminal instance and enter the command.

$ etlas run hazelcast-client

2. You can try few more instances to see Hazelcast work.

Credits

This example is inspired from the Hazelcast tutorial by Eugen.