Today the first meeting for the Beowulf Cluster took place. We planned out our goal of creating a computing cluster that would be able to be used by anyone on campus (including Mathematics and Data Analytics Graduate Students). The group decided that it would be good to first use a prexisting framework for an initial cluster and then work on our own system to be able to understand distributed computing further.
Baha gave a high level understanding on why we need distributed computing. Since we have a limitation of resources both with hardware and funding for high performance computing, we can use the computers in the ITL to be able to cheaply make a high performance machine.
Meetings are going to be continuing every Tuesday at 6:00pm. That said, for anyone who was at the meeting, or is still interested and were unable to make it, it would be reasonable for people to look into the idea of distributed computing. At the end of this post, I am including things such as a paper on MapReduce and a link to Apache Hadoop, a service we can be able to use for the cluster. Also, if anyone finds anything that would be useful for the project, if you have a name for the cluster, or if you have any other questions, email Benjamin Lannon with the beginning of the subject be either CDC or Computing Cluster.
See everyone next Tuesday!