jackgoza / ClusterConfig

Fork of anask/ClusterConfig hoping to focus on automating TPC testing for multinode spark clusters

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

ClusterConfig

Prerequisite: upload your public key to CloudLab. See tutorial.

  1. Create a Mitaka cluster in Utah using type m400.
  2. Clone https://github.com/anask/ClusterConfig and cd ClusterConfig
  3. Get the public ips of the machines in the cluster. Save them to cluster-machines.txt. with the master machine on the first line.

Set up one password and ssh keys for a cluster of machines:
execute ./set-cluster-passwd.sh from your local machine.

Set up a Hadoop and Spark cluster on these machines:
execute ./instance-setup.sh on every machine.
 or
execute ./run-instance-setup.sh from your local machine.

Then follow the last notes printed on the master (or in CLSTR_README.txt).

Designed for use with CloudLab. Tested on OpenStack Mitaka.

Special thanks to debarron and hastimal for paving the way.

About

Fork of anask/ClusterConfig hoping to focus on automating TPC testing for multinode spark clusters


Languages

Language:Shell 100.0%