Title:

Cluster Construction to Advance Analytics

High performance computing clusters are behind some of the most important research happening in the 21st century. These clusters are made up of many computers, or nodes, acting as single units for parallel processing of tasks assigned to the cluster. The goal of our project is to construct a high performance computing cluster using reclaimed hardware from the University of New Hampshire, as well as to document the cluster construction process for use by others creating similar clusters in the future. Our project utilizes six rack mounted servers, or nodes, reclaimed by our sponsor. These nodes were provisioned with a modern Linux operating system and networked together to form our computational cluster. The cluster is available to aid researchers looking to complete experiments involving intense computational components, whether they are students or faculty. In addition, the project’s resulting documentation includes cluster usage procedures and summaries of rationale for design decisions made during construction. This documentation will benefit the high performance community by providing an alternative perspective on cluster construction and implementation.

Name

Comment*

Comments are viewable only by submitter

Captcha*