Instantiating a node based client is the simplest way to get a Client that can execute operations against elasticsearch.

import static org.elasticsearch.node.NodeBuilder.*;

// on startup

Node node = nodeBuilder().node();
Client client = node.client();

// on shutdown

node.close();

When you start a Node, it joins an elasticsearch cluster. You can have different clusters by simply setting the cluster.name setting, or explicitly using the clusterName method on the builder.

You can define cluster.name in the /src/main/resources/elasticsearch.yml file in your project. As long as elasticsearch.yml is present in the classpath, it will be used when you start your node.

cluster.name: yourclustername

Or in Java:

Node node = nodeBuilder().clusterName("yourclustername").node();
Client client = node.client();

The benefit of using the Client is the fact that operations are automatically routed to the node(s) the operations need to be executed on, without performing a "double hop". For example, the index operation will automatically be executed on the shard that it will end up existing at.

When you start a Node, the most important decision is whether it should hold data or not. In other words, should indices and shards be allocated to it. Many times we would like to have the clients just be clients, without shards being allocated to them. This is simple to configure by setting either node.data setting to false or node.client to true (the NodeBuilder respective helper methods on it):

import static org.elasticsearch.node.NodeBuilder.*;

// on startup

Node node = nodeBuilder().client(true).node();
Client client = node.client();

// on shutdown

node.close();

Another common usage is to start the Node and use the Client in unit/integration tests. In such a case, we would like to start a "local" Node (with a "local" discovery and transport). Again, this is just a matter of a simple setting when starting the Node. Note, "local" here means local on the JVM (well, actually class loader) level, meaning that two local servers started within the same JVM will discover themselves and form a cluster.

import static org.elasticsearch.node.NodeBuilder.*;

// on startup

Node node = nodeBuilder().local(true).node();
Client client = node.client();

// on shutdown

node.close();

The TransportClient connects remotely to an elasticsearch cluster using the transport module. It does not join the cluster, but simply gets one or more initial transport addresses and communicates with them in round robin fashion on each action (though most actions will probably be "two hop" operations).

// on startup

Client client = new TransportClient()
        .addTransportAddress(new InetSocketTransportAddress("host1", 9300))
        .addTransportAddress(new InetSocketTransportAddress("host2", 9300));

// on shutdown

client.close();

Note that you have to set the cluster name if you use one different than "elasticsearch":

Settings settings = ImmutableSettings.settingsBuilder()
        .put("cluster.name", "myClusterName").build();
Client client =    new TransportClient(settings);
//Add transport addresses and do something with the client...

Or using elasticsearch.yml file as shown in Node Client

The client allows to sniff the rest of the cluster, and add those into its list of machines to use. In this case, note that the IP addresses used will be the ones that the other nodes were started with (the "publish" address). In order to enable it, set the client.transport.sniff to true:

Settings settings = ImmutableSettings.settingsBuilder()
        .put("client.transport.sniff", true).build();
TransportClient client = new TransportClient(settings);

Other transport client level settings include: