3. Stage 1--Define and Create the Test Environment : Cluster Environment: Hardware and Software
Share this page                  
Cluster Environment: Hardware and Software
Generally speaking the hardware requirements of VectorH match the recommendations of the underlying Hadoop vendors: dual CPU commodity servers with sufficient RAM, disks, and 10 GbE networking.
For disks, we recommend using the higher end of the Hadoop vendor recommendations and opting for at least 16 physical drives and, if possible, choosing the drive type as SAS rather than SATA or Near Line SAS. Solid State Disk (SSD) is preferred over rotating drives for maximum performance.
Some sample reference information is provided here:
Hardware Recommendations For Apache Hadoop ( http://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.1.2/bk_cluster-planning-guide/content/ch_hardware-recommendations.html)
Select the Right Hardware for Your New Hadoop Cluster ( http://blog.cloudera.com/blog/2013/08/how-to-select-the-right-hardware-for-your-new-hadoop-cluster/)
Consult the vendor documentation on high performance DataNodes for distribution specific configuration details.
For the test cluster, ensure that the number of VectorH nodes is greater than the HDFS block replication factor. (As of version 4.2.2 a configuration item is available the sets the replication factor for VectorH independently of the HDFS default value.)