Visit additional Tabor Communication Publications
January 13, 2011
The Weekly Top Five features the five biggest HPC stories of the week, condensed for your reading pleasure. This week, we cover the Intel-NVIDIA cross-licensing agreement, the arrival of a Cray supercomputer at Colorado State, advancements in the understanding of storage materials, the latest batch of AAAS Fellows, and UW-Madison's new HPC cluster.
Intel, NVIDIA Ink Cross-Licensing Agreement
In what is arguably the biggest story this week, representatives from Intel and NVIDIA have hammered out a six-year cross-licensing deal. Intel has agreed to pay NVIDIA $1.5 billion in licensing fees divided into five annual payments, the first due Jan. 18, 2011. The two chipmakers have also agreed to drop all outstanding legal disputes.
In a prepared statement, Jen-Hsun Huang, NVIDIA's president and chief executive officer, said: "This agreement signals a new era for NVIDIA. Our cross license with Intel reflects the substantial value of our visual and parallel computing technologies. It also underscores the importance of our inventions to the future of personal computing, as well as the expanding markets for mobile and cloud computing."
For an in-depth explanation, look no farther than HPCwire Editor Michael Feldman's coverage.
For the time-starved, here's the meaty bit:
In a nutshell, the agreement provides cross-licensing access to each other's patents. However, it's not a license to repurpose one another's chip designs; rather its an understanding not to sue each other when they bump up against their competitor's patents. This is important because both NVIDIA and Intel own rich patent portfolios that apply to many areas of computing. Without such an understanding, it's nearly impossible for engineers to design anything without inadvertently stepping into someone else's territory. It gives both parties the freedom to build CPUs, GPUs, and everything in between without having to worry about who came up with the original ideas.
Cray Provides CSU with New Supercomputer
Colorado State University has a shiny new Cray supercomputer, thanks to a National Science Foundation grant, worth $627,326. The ISTeC High Performance Computer, made possible by stimulus funding, will provide university researchers with platform for advanced modeling, simulation and analysis at higher levels than were previously available.
According to the official announcement, "the system will support much larger and more complex problems in science and engineering, especially for data intensive applications; add greater physical fidelity to existing models; facilitate application of computing to new areas of research and discovery; and support training to attract new researchers to computational science, engineering and mathematics."
The midrange Cray XT6m supercomputer has 1,248 cores, 1.6 terabytes of main memory, and 32 terabytes of disk storage. Colorado State researchers plan to use the system for a diverse assortment of data- and compute-intensive applications, among them ultraviolet laser design, weather forecasting, bioinformatics, atmospheric modeling, network traffic analysis, and robotics.
A reception will take place Friday to celebrate the supercomputer's debut.
DVD Storage Mechanism Revealed
Scientists at Forschungszentrum Jülich, working in tandem with researchers from Finland and Japan, have solved a DVD mystery -- not some murder mystery, but the mysterious workings of the DVD storage format itself. The physical basis for the storage mechanism was not previously understood in detail despite the disk's ubiquity. The team's findings, published in the current issue of the journal Nature Materials, provide insight into the read and write processes in a DVD.
Using the JUGENE supercomputer as well as the Japanese synchrotron SPring-8, the world's most powerful x-ray source, the researchers were able to determine the structures of both storage phases for the first time and develop a model that explained the rapid phase change.
From the announcement:
Some 4,000 processors of the Jülich supercomputer JUGENE were used for over four months in order to obtain the necessary precision. In addition to sheer computing power, however, experience in scientific computing and the simulation of condensed matter is essential. [Dr. Robert] Jones [of Forschungszentrum Jülich] notes: "Forschungszentrum Jülich is one of the few places where all these aspects come together."
The new knowledge is expected to lead to storage media with longer life, larger capacity, or shorter access times.
AAAS Announces New Fellows
This week, the American Association for the Advancement of Science (AAAS) made public its yearly selection of fellows. A total of 503 recipients were named from more than 220 institutions worldwide, including 16 designated with an "Information, Computing, and Communication" affiliation. The honor recognizes individuals who have made significant contributions to the advancement of science and technology. The newly-inducted fellows will be presented with a certificate and a blue and gold rosette pin at the Fellows Forum on Feb. 19 2011, held during the AAAS Annual Meeting in Washington, D.C.
Many academic institutions have released their own announcements, naming faculty members who have been hand-picked for this prestigious group. Among them are the University of Tennessee, Knoxville, the Pacific Northwest National Laboratory, and Louisiana State University. You can read about the selection process here.
UW-Madison Cluster Enlisted to Fight Pollution
Several University of Wisconsin-Madison departments banded together to bring a new HPC cluster to campus. The Euclid cluster, now the largest at UW-Madison, harnesses the power of many computers at once in order to run large-scale computing jobs more quickly. It can also move large datasets and files at high speeds among the cluster's individual servers.
Euclid was the result of over nine months of planning involving a partnership of several campus departments with vendor assistance coming from Dell, Cisco, Chelsio and APC. The cluster has 261 servers, almost 2,100 Intel Nehalem computer cores, and 13 terabytes of central storage. A peak theoretical performance of 19 teraflops gives Euclid the power of 1,000 average desktop computers. The system's high-bandwidth, low-latency 10 Gigabit Ethernet interconnect allows for efficient communication between the various servers.
While Euclid was designed to tackle the usual array of HPC applications, such as weather modeling, high energy physics, bioinformatics, and materials design, it is primarily being used for materials science, specifically in the design of novel catalytics, under the direction of professor Manos Mavrikakis. The professor's research group uses computational chemistry approaches to improve engineering practice in a variety of areas, including chemical processing, alternative energy and pollution prevention. The group is part of worldwide effort to uncover the next generation of catalytic materials and has published its findings in Science, 329, 1633 (2010).
Additional background information is available here.
Jun 19, 2013 |
Supercomputer architectures have evolved considerably over the last 20 years, particularly in the number of processors that are linked together. One aspect of HPC architecture that hasn't changed is the MPI programming model.
Jun 18, 2013 |
The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Jun 18, 2013 |
Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Jun 17, 2013 |
The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Jun 14, 2013 |
For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
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.
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?
Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.