In this podcast, the Radio Free HPC team goes over a Trip Report from Rich Brueckner from insideHPC, who’s been on the road for the past month at a series of HPC conferences:
HPC User Forum in Broomfiled, CO. The event featured two full days of talks. Here we discuss a couple of standouts that you can watch on insideHPC:
- Processing 1 Exabyte per Day for the SKA Radio Telescope. Peter Braam from Cambridge University presented this talk in the Disruptive Technologies session. “The Square Kilometre Array is an international effort to investigate and develop technologies which will enable us to build an enormous radio astronomy telescope with a million square meters of collecting area.”
- ExaNeSt Technology: Targeting Exascale in 2018. Peter Hopton from Iceotope presented this talk at the HPC User Forum. “ExaNeSt will develop, evaluate, and prototype the physical platform and architectural solution for a unified Communication and Storage Interconnect and the physical rack and environmental structures required to deliver European Exascale Systems. The consortium brings technology, skills, and knowledge across the entire value chain from computing IP to packaging and system deployment; and from operating systems, storage, and communication to HPC with big data management, algorithms, applications, and frameworks. Building on a decade of advanced R&D, ExaNeSt will deliver the solution that can support exascale deployment in the follow-up industrial commercialization phases.”
- User Agency Panel Discussion on the NSCI Initiative. In this video (with transcript) from the 2015 HPC User Forum in Broomfield, Bob Sorenson from IDC moderates a User Agency panel discussion on the NSCI initiative. “You all have seen that usable statement inside the NSCI, and we are all about trying to figure out how to make usable machines. That is a key critical component as far, as we’re concerned. But the thing that I think we’re really seeing, we talked about the fact that a single thread performance is not increasing, and so what we’re doing is we’re simply increasing the parallelism and then the physics limitations, if you will, of how you cool and distribute power among the parts that are there. That really is leading to a paradigm shift from something that’s based on how fast you can crunch the numbers to how fast you can feed the chips with data. It’s really that paradigm shift, I think, more than anything else that’s really going to change the way that we have to do our computing.”
PBS Works User Group. Job schedulers like PBS Works are key to keeping supercomputers running efficiently. Notable talks included:
- HPC Across the Enterprise: How HPC Transforms the Corporate IT Ecosystem. Thomas Leung from the GE Global Research Center presented this talk at the PBS Works User Group. “The commercial world uses significant HPC resources for simulation and product design. An increasing number of HPC systems are deployed in the commercial space and their scale is getting larger and larger. These advanced systems push limits in every aspect of Enterprise IT. Accommodating such systems within the enterprise is a challenge, and there have been many recent changes to enterprise IT infrastructures and architectures resulting from the need to support HPC.”
- Video: HPC Technology Panel at the PBS Works User Group. Rich Brueckner from insideHPC moderated this panel discussion on current trends in HPC. “President Obama’s Executive Order establishing the National Strategic Computing Initiative (NSCI) will set the stage for a new chapter in leadership computing for the United States. In this panel discussion, thought leaders from leading supercomputing vendors share their perspectives on current HPC trends and the way forward.”
HPC Advisory Council Spain Conference. This day-long event in Barcelona featured a set of great talks, but here we discuss two notable presentations:
- Communication Frameworks for HPC and Big Data. DK Panda from Ohio State University presented this talk at the HPC Advisory Council Spain Conference. “Dr. Panda and his research group members have been doing extensive research on modern networking technologies including InfiniBand and 10-40GE/iWARP. His research group is currently collaborating with National Laboratories and leading InfiniBand and 10-40GE/iWARP companies on designing various subsystems of next generation high-end systems.”
- The Road to Exascale. Rich Graham from Mellanox looks at how interconnects are evolving to get us to the next grand challenge of supercomputing performance.
- On the Role of Flash in Large-Scale Storage Systems. Nathan Rutman presented this talk. “So why is a spinning disk company talking about Flash? Last year, Seagate acquired Avago LSI’s flash division. We now have an array of flash-based storage. So I have nothing against Flash. This presentation is really on: Where does Flash make sense? I also have a personal agenda because I hate the term “Burst Buffer.” Everyone says “Burst Buffer” instead of saying “Flash.” It drives me crazy. So I’m going to explain what a Burst Buffer is and what it is not.”
- Video: Lustre 2.9 and Beyond. Andreas Dilger from Intel presents an overview on the features currently under development for the upcoming Lustre 2.8 and 2.9 releases. This includes Layout Enhancement, Progressive File Layouts, Data-on-MDT, and improved single-client metadata and IO performance. In addition, several Lustre-specific ZFS improvements are also under development that will be available in this timeframe.
- Lustre Development Update from LAD’15. Chris Morrone from OpenSFS/LLNL gave this talk at LAD’15. “The first thing I want to talk about is that Lustre.org is now managed collectively by OpenSFS and EOFS. So the Lustre.org site is now managed by the community. One of the major benefits of this is that the major mailing lists (lustre-announce, lustre-discuss, and lustre-devel) are now under our control, so we want everyone to come back to those mailing lists and start using them again.”