BigData(Hadoop) Virtualization as a Service
This development POC was in an effort to break away from the traditional infrastructure as a service to a more service provider package for a BigData support solution. Also to develop a new standard for technology to support the multi-tenant and multi-distribution Hadoop as a service provider market. Hardware included in the design was Cisco servers, EMC isilon storage and Cisco switching. Software included VMware vSphere virtualization in combination with BDE for cluster automation that included multi-distributions that ranged from the base Apache, Hortonworks to Cloudera. Also OneFS using NFS and HDFS protocols was incorporated for the actual data analysis repository. Fusion IO technology was also used and combined with IO Turbine virtual appliances for a more efficient Data load times via caching. The POC was a successfully project that takes BigData from the traditional physical server combined with local SSD storage and moves it into the future and that of virtualization.