October 30, 2013
In the latest of an interesting series of Q&As on Argonne National Laboratory's website, Susan Coghlan, Deputy Division Director at the Argonne Leadership Computing Facility, opens up about her role working with Mira, the fifth-fastest supercomputer in the world.
In addition to her position as Deputy Division Director at the Argonne Leadership Computing Facility (ALCF), Coghlan is also the project director for the large supercomputing systems that are essential to Argonne's mission. Overseeing these multimillion-dollar systems installations has in many ways become her primary role and as such she has a unique insight into what it takes to go from planning stage to implementation for these leadership-class systems.
With Mira, which occupies the fifth position on the TOP500 list of fastest computers, now in full production mode and open for scientific research, Coghlan addresses what it took to get to this point.
"Typically, it takes about five years, from start to finish, to complete a project like this – to go from preparing a budget and figuring out what type of system is possible to the delivery and deployment of the system," she states. "So, we actually started the planning and documentation process for Mira in 2008. In this case, we worked very closely with our vendor, IBM, and also had a close collaborative R&D relationship with Lawrence Livermore National Laboratory and IBM. I was responsible for making sure we had the right people and appropriate experts at Argonne reviewing the designs, providing the feedback to IBM, approving design choices and looking at applications and their performance on the different designs as they moved forward."
Coghlan continues: "The delivery of the first racks for Mira started in April 2012, and that process of delivery, then build-out, went on until July 2012. After delivery, the system moved into the acceptance preparation phase, and finally through the acceptance tests. Mira was accepted in December 2012. Initially, the research being run on Mira was part of the Early Science Program (ESP), and now that Mira is fully online, it is open to all research, including work being done that is part of the Innovative & Novel Computational Impact on Theory and Experiment (INCITE) program."
Now Mira is up and running and tackling some of the grand challenges of our time with a focus on sustainable energy, a healthy environment and the security of the country. Current projects include developing better batteries, wind turbine advances and studies into water and earthquake research.
Read the full interview here.
Check out this time-lapsed video of the installation of Mira below:
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?