Benchmarking Datastax Cassandra cluster with YCSB on EC2

Start a Cassandra cluster

Closely following the Datastax instructions

I manage to run, but not benchmark, a C* cluster. Here is how I did it, what didn't work, an important notes on Cloud-init and a existential question.

1. Launch AMI ami-ada2b6c4
2. Use cassandra-cluster security group
3. Choose one of the recommended instance:

• Development and light production: m3.large
• Moderate production: m3.xlarge
• SSD production with light data: c3.2xlarge
• Largest heavy production: m3.2xlarge (PV) or i2.2xlarge (HVM)

4.Choose number of VM to run, and set user data (3 in the following example)

User data

--clustername name test-cluster
--totalnodes 3
--version community
--opscenter no

5. Add Storage
6. Launch

test machine

8. Create a EC2 Linux instance
9. follow the instructions to test the cluster. Make sure to use

strategy_options = {replication_factor:3};

to take advantage of the cluster replication replication factor of 3 is used by Google C* benchmark