Visit additional Tabor Communication Publications
April 30, 2012
This August, the IEEE is hosting its annual symposium on high-performance interconnects, known as Hot Interconnects (HOTI) . Now in its 20th year, the event focuses on the latest developments in the field, both in industry and academia, with a special emphasis on how the technology is advancing in the realm of supercomputing and large-scale datacenters. The event covers chip-to-chip interconnects as well as networking fabrics that bind whole systems and datacenters together.
In the world of high performance computing, the interconnect is second only to the microprocessor in terms of performance-critical components. As systems expand into every-widening arrays of processors and servers, the interconnect is the enabling technology that allows the underlying computation to scale efficiently. And with the size of these systems increasing, new developments in this area are generating even greater interest.
This year's HOTI event comes at a time when the latest commodity system networks, FDR InfiniBand and 10 GbE Ethernet, are making their way into some of the newest HPC systems. Meanwhile, custom-built interconnects are undergoing somewhat of a revival, with examples like Fujitsu's Tofu interconnect, IBM's Blue Gene/Q torus, and the Tianhe-1A's Galaxy network. As a backdrop to these developments are custom interconnects that are being acquired by commodity chipmakers -- SeaMicro's Freedom supercomputing fabric, by AMD, and more recently, Cray's Aries interconnect and related technology by Intel.
The HOTI meeting provides a forum that encompasses all these developments. This year the event will take place from August 22nd through the 24th at the Huawei North America Headquarters in Santa Clara, California. The first two days are the symposium proper, consisting of presentations by vendors and academicians, with the final day dedicated to tutorials. The event organizers are calling for paper submissions by May 13, 2012 for peer review by its technical committee.
HPCwire spoke with University of Illinois' Torsten Hoefler, HOTI's Technical Chair, and Myricom's Patrick Geoffray, the General Chair, about the conference and what's on tap for this year's event.
HPCwire: For our readers who are not familiar with the IEEE Hot Interconnects Symposium, can you briefly encapsulate what the event is about?
Patrick Geoffray: HOTI is a technical conference for the presentation of new developments and architectures in high-speed interconnects both within and between systems as well as the evolution of software to improve network performance, efficiency and configuration for virtualization, file systems and big data. Along with its sister conference, Hot Chips, HOTI aims to attract leading research work in academia as well as industry, interconnecting between these two communities worldwide.
HPCwire: Who has been attending the event in years past?
Torsten Hoefler: Because HOTI is usually in Silicon Valley, we have a great audience of network researchers, software engineers and product managers from nearly all the local major universities and companies – Stanford, Berkeley, MIT, Columbia, Cisco, HP, IBM, Oracle, Google, Facebook, etc. In addition technologists from IT organizations worldwide like Wall Street, Amazon, Boeing, and centers in Europe and Asia attend to find out the latest developments.
HPCwire: Such as?
Hoefler: Last year's HOTI featured talks on the TianHe-1A and Kei (Tofu) interconnects, powering the world's two fastest computers at that time. In addition the networks of IBM's Blue Gene/Q machine were first revealed as a keynote. Videos of the 2011 conference are available on the site.
HPCwire: How do you go about attracting research papers from academia?
Hoefler: The reviewers of the submissions are first rate and are strongly oriented to both research and the progress of data center infrastructure. As we are an IEEE-sponsored event all papers are published in the IEEE digital library which is open to members worldwide. Top papers and talks from HOTI usually appear in IEEE journals, for example, the January/February 2012 issue of IEEE Micro featured Petascale Network Architectures from HOTI11. In addition HOTI has an average impact factor of 10.5 as stated by MS Academic search.
HPCwire: Are you seeking paper submissions in particular areas?
Geoffray: Examples include chipset and SOC level processing and interfaces – electronic and optic, high bandwidth low latency I/O for networks, switches and storage, novel APIs and protocols for HPC, cloud, and big data, software definition of fabric and network configuration and performance.
HPCwire: What are some of the presentations on the schedule for this year's conference?
Hoefler: In our single track format we have four great keynotes by leaders in industry and academia scheduled. Bob Alverson, the lead architect of Cray's Aries network will discuss the evolution of networking at Cray in his talk titled "Cray High Speed Networking." John Roese will discuss Huawei's strategy towards networking in his talk "The Future Of Network Technology - What is Old, is New Again". Fuad Doany will talk about IBM's new break-through technology in optical networking and storage, the Holey Optochip, and Nick McKeown will once again challenge our status-quo thinking with the latest Developments In Software Defined Networking.
We also have an industrial panels dealing with the controversial issues of today and great practical tutorials for early adopters. It is now up to you and the community to contribute tutorials, papers and presentations to the conference. We want to encourage everybody to submit their best work to this year's HOTI. The technical paper submission is still open until May 13th!
This interview was prepared in collaboration with the HOTI’12 media chairs Bill Boas and Raj Channa.
May 23, 2013 |
The study of climate change is one of those scientific problems where it is almost essential to model the entire Earth to attain accurate results and make worthwhile predictions. In an attempt to make climate science more accessible to smaller research facilities, NASA introduced what they call ‘Climate in a Box,’ a system they note acts as a desktop supercomputer.
May 22, 2013 |
At some point in the not-too-distant future, building powerful, miniature computing systems will be considered a hobby for high schoolers, just as robotics or even Lego-building are today. That could be made possible through recent advancements made with the Raspberry Pi computers.
May 16, 2013 |
When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
May 15, 2013 |
Supercomputers at the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) have worked on important computational problems such as collapse of the atomic state, the optimization of chemical catalysts, and now modeling popping bubbles.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.
The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.