Best Practices for Deploying Hadoop on SUSE Linux Enterprise Server

The number of data sources available to a typical organization that can be mined for business benefit has exploded. At the same time, declining server and storage costs and the emergence of software such as Hadoop and NoSQL have made it relatively inexpensive for organizations to store and analyze this data. Businesses are looking to accelerate the speed with which they can gain insight to this large volume of disparate data to answer critical business questions and make better, fact-based decisions.

Planning, installing, configuring, operating and maintaining Hadoop clusters are non-trivial tasks that require a lot of special knowledge. SUSE has published a new white paper that details best practices when deploying SUSE Linux Enterprise Server with Hadoop. This paper gives guidance on installing Hadoop on SUSE Linux Enterprise Server and on configuring and optimizing the operating system for Hadoop. It also includes a comparison between virtualized and bare-metal Hadoop workloads. This white paper and the underlying technical workshops were closely developed with SUSE OpenStack partner B1 Systems.

While providing guidance on general infrastructure considerations like hardware, network services, software infrastructure and facilities, the paper focuses on the Linux operating system installation and optimization for a Hadoop environment including tips for installing and configuring Cloudera CDH and Hortonworks HDP on SUSE Linux Enterprise Server, and testing methodologies for benchmarking your Hadoop cluster.

SUSE Linux Enterprise Server is the best foundation for big data deployments. Tested and certified to work with the latest releases from leading big data partners like Cloudera (CDH5 certified) and Hortonworks (HDP2 certified), SUSE Linux Enterprise has been validated against a comprehensive suite of integration test cases for big data deployments, benchmarked for scale under varied workloads and comprehensively documented. SUSE has a 22-year legacy in enterprise Linux and leadership in performance-intensive environments such as high performance computing and Linux on the mainframe and is today the cornerstone operating system of market leading big data solutions such as SAP HANA and Teradata. Need further proof of SUSE Linux Enterprise Server as the standard for Hadoop and big data? The artificially intelligent computer system IBM Watson, which IBM calls “one of the most significant inventions in IBM’s history”, runs Hadoop on SUSE Linux Enterprise Server. Now that’s a pretty solid reference.

If you’re an IT architect who is responsible for scoping, planning and executing the deployment of a Hadoop cluster, or a system administrator charged with maintaining, operating and possibly extending existing Hadoop clusters, download this paper now to gain some valuable insight into maximizing the performance of your Hadoop cluster on SUSE Linux Enterprise Server.