Continuuity Open Sources Loom Big Data Cluster Management Tool

Continuuity Inc. yesterday open sourced its Loom software for managing cloud-based Big Data clusters.

Following an increasingly familiar pattern, the company's in-house developers found existing tools to be too cumbersome and complex when they were working on internal software projects, so they created their own tools to smooth the workflow. Continuuity's main product is Reactor, described as "a runtime for scalable Big Data applications" that simplifies the complexities of Apache Hadoop.

Used internally with Reactor projects for about a year, Loom was refined, templated and released to GitHub under an Apache license.

"We needed a fast, self-service way to provision clusters because testing against real clusters is such a critical and frequent component of our development cycle," explained Jonathan Gray, co-founder and CEO. "Our vision was to enable all Continuuity developers to be able to quickly create a cluster with the push of a button."

The company found homegrown scripts combined with recipes from the Chef automation platform to be too complex, so it decided to build its own complete production system that developers could easily work with. "Today, Loom is used to provision Sandbox Reactors on the Continuuity Cloud and as an internal DevOps tool to test new and incremental features on an ongoing basis," the company said.

Loom includes predefined templates for clusters such as Hadoop and LAMP, features a REST API and "works out of the box with in-memory ZooKeeper and embedded DB," the company said.

Continuuity said Loom has been tested on the Mac OS and Linux, and a standalone version--the current "snapshot of the project--can be downloaded and run on Mac machines. "Loom has been tested on OSX Mavericks," the company said on a FAQ. "There are plans to add Windows support in a future release, but until then Windows users will have to use the usual workarounds, such as using Cygwin."

About the Author

David Ramel is an editor and writer for Converge360.