PelicanHPC is an iso-hybrid (CD or USB) image that let’s you set up a high performance computing cluster in a few minutes. A Pelican cluster allows you to do parallel computing using MPI. You can run Pelican on a single multiple core machine to use all cores to solve a problem, or you can network multiple computers together to make a cluster. The frontend node (either a real computer or a virtual machine) boots from the image. The compute nodes boot by PXE, using the frontend node as the server. All of the nodes of the cluster get their filesystems from the same image, so it is guaranteed that all nodes run the the same software. Packages can be added to all nodes using apt-get, thanks to aufs. The bootable image is created by running a single script, which takes advantage of the Debian Liveinfrastructure. It is very easy to create a custom version with new packages by adding package names to the script and then running it. PelicanHPC is creates a fairly bare-bones cluster, with few applications other than those relevant to the author’s research and teaching. To use PelicanHPC for serious work in other disciplines, it will be necessary to create a customized version. The tutorial explains how to do this. It is not very difficult. Two examples of specialized distros that add to the PelicanHPC base are MOLA and birgHPC. These are very good examples of how it is intended for the make_pelican script to be used.
• Pelican is created using tools from the Debian Live project. To make your own version you only need live-build (also deboostrap and rsync) and the make_pelican script, which is provided below. All of this is also available on the PelicanHPC images, so you can build a custom version while running PelicanHPC.
• Contains the latest stable version of the OpenMPI implementation of MPI.
• Available for 64 bit CPUs. An older version is available for 32 bit CPUs.
• Has the Linpack HPL benchmark.
• Pelican releases and all testing is done using the stable (Wheezy) version of Debian GNU Linux as the base.
Downloading Mirror: mirror.fiba-solution.com