StackIQ Update Rocks Big Data
This week StackIQ released the latest version of their automation and cluster management suite with the announcement of Rocks+ 6. While often considered a pure high performance computing geared company, they say Rocks+ 6 is tailored for the needs of massive systems handling big analytics workloads.
According to the team behind the release, data-intensive computing environments require “powerful management software that turns loosely coupled commodity hardware and open source software into tightly coupled enterprise-grade appliances.” This is the stated goal of other cluster management suite providers, and so far, all of the major ones are hopping on the big data management bandwagon.
Like other cluster management solution providers, StackIQ is touting their ability to handle the major distributions of Hadoop, including the raw Apache version and distros from Cloudera and MapR. According to the company, the new release will provide “automated provisioning and management of MapReduce, HDFS, HBase, Zookeeper and Hive.” They point to new developments in beta-land for supporting both Cassandra and MongoDB.
The concept behind this release for customers with what StackIQ is calling “big infrastructure” is that it is possible to quickly obtain a complete stack that wraps up everything from the OS to the middleware to the MapReduce and HDFS layers.
StackIQ says that one of the keys to fine-tuning of Rocks+ for big data systems lies in their database-driven design. They said that this “ensures consistent deployments without requiring a team of engineers writing custom scripts.” For those who want to further tune their system to the application needs, there are optional modules that can be integrated for more control.
While they’re not naming names at this point, StackIQ says that they have use cases that point to their ability to help customers spin up “big infrastructure” from bare metal in minutes. They point to one “hyperscale web company” that used their software to deploy their custom Hadoop distribution across a large data center, and that they “successfully used Rocks+ to upgrade and redeploy the entire cluster from bare metal in less than 30 minutes.”
One of the other improvements the team points to are refinements in the GUI, which sports a new visualization feature that provides real-time information about the bare metal, parallel installation process. They are also throwing in support the newest upgrades across a number of operating systems that are now in their sixth incarnations—Red Hat, CentOS and Oracle Linux.
In addition to offering Rocks+ 6 through their own site, it is also available through AWS as StackIQ is an Amazon Web Services Solution Provider. Partners, including Dell and HP will be carried along with the update. Rocks+ 6 is free for systems that are at the 16-node level. More details can be found at the Rocks+ 6 fact sheet.