Group items tagged

Our community guide allows you to set up your own MPI (Message Passing Interface) based supercomputer cluster with the Playstation 3. This guide was co-written by Gaurav Khanna, based on his previous work on the Gravity Grid and is a current run-time environment for the research of co-author (Chris Poulin), based on his current work in distributed pattern recognition. As such, we currently utilize the Fedora Core for this infrastructure and illustrate a "how-to" below. NOTE: We focus on the Fedora 8 distribution, due to prevalence of Fedora and its Cell SDK (3.0) compatibility. Finally, this content should be considered open source, and here is the license.

With the following three steps, you can set up a Disco cluster in the Amazon's Elastic Computing Cloud. This will cost you a few dollars (or more, depending on your needs) but requires no resources on your side besides a single machine that you use to setup the cluster. In this setup Disco master and nodes run on EC2. Your Disco client can run either on the master node on EC2 or on a local machine.

We build here a repository of assembled transcript sequences from the contigation (Expressed Sequence Tag (known as EST) & mRNA) in order to discover new genes from already existing data. Publicly available EST & mRNA sequences are clusterised and then contigated with specific bioinformatic tools (see technology).

Hadoop comes bundled with launch scripts to simplify initializing an Amazon Elastic Compute Cloud (EC2) cloud for Hadoop. Once initialized, running CloudBurst is identical to running on a local cluster. If you use EC2 regularly with the same datasets (i.e. the human genome as a reference), you will probably want to copy the data once to Amazon Simple Storage Service (S3) so you can quickly copy the data from S3 to your cloud at low cost.