This section describes how to run YCSB on Elasticsearch running locally.

Clone the YCSB git repository and compile:

git clone git://github.com/brianfrankcooper/YCSB.git
cd YCSB
mvn clean package

Now you are ready to run! First, load the data:

./bin/ycsb load elasticsearch -s -P workloads/workloada -p path.home=<path>

Then, run the workload:

./bin/ycsb run elasticsearch -s -P workloads/workloada -p path.home=<path>

Note that the <path> specified in each execution should be the same.

The Elasticsearch binding has two modes of operation, embedded mode and remote mode. In embedded mode, the client creates an embedded instance of Elasticsearch that uses the specified <path> to persist data between executions.

In remote mode, the client will hit a standalone instance of Elasticsearch. To use remote mode, add the flags -p es.remote=true and specify a hosts list via -p es.hosts.list=<hostname1:port1>,...,<hostnamen:portn>.

./bin/ycsb run elasticsearch -s -P workloads/workloada -p es.remote=true \
-p es.hosts.list=<hostname1:port1>,...,<hostnamen:portn>`

Note that es.hosts.list defaults to localhost:9300. For further configuration see below:

The default setting for the Elasticsearch node that is created is as follows:

cluster.name=es.ycsb.cluster
es.index.key=es.ycsb
es.number_of_shards=1
es.number_of_replicas=0
es.remote=false
es.newdb=false
es.hosts.list=localhost:9300 (only applies if es.remote=true)

If you wish to customize the settings used to create the Elasticsearch node you can created a new property file that contains your desired Elasticsearch node settings and pass it in via the parameter to 'bin/ycsb' script. Note that the default properties will be kept if you don't explicitly overwrite them.

Assuming that we have a properties file named "myproperties.data" that contains custom Elasticsearch node configuration you can execute the following to pass it into the Elasticsearch client:

./bin/ycsb run elasticsearch -P workloads/workloada -P myproperties.data -s

If you wish to change the default index name you can set the following property:

es.index.key=my_index_key

If you wish to run against a remote cluster you can set the following property:

es.remote=true

By default this will use localhost:9300 as a seed node to discover the cluster. You can also specify

es.hosts.list=(\w+:\d+)+

(a comma-separated list of host/port pairs) to change this.