Everyone knows applications drive the HPC boat. It is one thing to run benchmarks and burn-in programs, but when it is time for production work, applications take over. Fortunately, there are many applications that can take advantage of clusters. These applications can be divided into three oversimplified categories.
"This project," said Sergiu Sanielevici, PSC director of scientific applications and user support, who also leads user support and services for the TeraGrid, "exemplifies how powerful systems like Pople can open doors to data-mining and data-centric research in fields not traditionally associated with HPC, such as the social sciences, and make it possible to get answers that would otherwise be impractical or impossible." PSC supported this project through the NSF TeraGrid program, which allocates large-scale computing resources free to researchers at U.S. universities on a peer-review proposal basis.
LexisNexis has built its business on bringing together billions of different records from many different sources. Its data and tools allow customers to query those data to find out everything from which middle-aged soccer moms bought white wine last month, to the names of everyone who registered a car last week in San Diego with a license plate that has an "O" and an "H" in it. Recently the company has been working with Sandia National Laboratories to understand whether the LexisNexis data tools might help researchers manage and understand the flood of data coming from the supercomputers and high resolution scientific instruments that drive discovery tod
"High-performance computing is transforming physics research," said Ralph Roskies, co-scientific director of the Pittsburgh Supercomputing Center (PSC), during a presentation on Friday, March 20, at the American Physical Society Meeting, held in Pittsburgh, March 16-20. "The Impact of NSF's TeraGrid on Physics Research" was the topic of his talk, which led off a panel of physicists who have made major strides in their work through the TeraGrid, the National Science Foundation's cyberinfrastructure program. "These world-class facilities," said Roskies, "on a much larger scale than ever before, present major new opportunities for physics researchers to carry out computations that would have been infeasible just a few years ago."
Last year, researchers at Indiana University's Cryo-Transmission Electron Microscopy Facility (cryoEM) acquired a powerful new microscope capable of electron cryomicroscopy, a method of analyzing the structure of proteins at really low temperatures. However, the process often damages samples so researchers have to use a large number to ensure accurate results. This in turn means multiple images from hundreds of thousands of protein particles which then need to be made into composite images, requiring thousands of hours of compute time. So the analysis, movement, and management of all these image files quickly became an IT headache almost as soon as they flipped the on switch
It looks like selling baby supercomputers based on a blade design and running the HPC variants of Windows and Linux is not as easy as Cray had hoped - which is why Cray has announced a new lower-end baby super, the CX1-LC
I recently gave a presentation entitled "Cyberinfrastrucutre and its Role in Science" at IAI International Wireless Sensor Networks Summer School held at the University of Alberta on July 6th, 2009. This presentation examines some of the challenges scientists face and describes various cyberinfrastructure technologies that help address these challenges. Example projects employing cyberinfrastructure technologies that we have worked on at the Grid Research Centre, including the GeoChronos project, are also presented. Cyberinfrastructure and its Role in Science
In a sharply worded speech to the Security Council this week, Russian President Dmitry Medvedev warned that Russia is "significantly behind" other countries in producing powerful supercomputers, and said the lag hurts Russia's competitiveness and its ability to defend itself.
The San Diego Supercomputer Center (SDSC) at the University of California, San Diego, has officially launched the Triton Resource, an integrated, data-intensive computing system primarily designed to support UC San Diego and UC researchers. The Triton Resource -- which features some of the most extensive data analysis power available commercially or at any research institution in the country because of its unique large-memory nodes -- also will be available to researchers throughout the larger academic community, as well as private industry and government-funded organizations. Plans for the new system were first announced last fall, as SDSC formally opened a new building and data center that doubled the size of the existing supercomputer center to 160,000 square feet.
Linux magazine HPC Editor Douglas Eadline had a chance recently to discuss the current state of HPC clusters with Beowulf pioneer Don Becker, Founder and Chief Technical Officer, Scyld Software (now Part of Penguin Computing). For those that may have come to the HPC party late, Don was a co-founder of the original Beowulf project, which is the cornerstone for commodity-based high-performance cluster computing. Don’s work in parallel and distributed computing began in 1983 at MIT’s Real Time Systems group. He is known throughout the international community of operating system developers for his contributions to networking software and as the driving force behind beowulf.org.
Personal supercomputers that lash together a stack of graphics processing units and can sit on a desktop are becoming popular with researchers. These machines can be used to run calculations by the desk instead of waiting for time on one of the national supercomputers.
Before specialized graphics-processing chips existed, pioneers in the field of visualization used multicore supercomputers to realize data in three dimensions. Today, however, the speed at which supercomputers can process data is rapidly outstripping the speed at which they can input and output that data. Graphics-processing clusters are becoming obsolete.
Virtual Machines and Types of Service for TeraGrid Computing Foundational capabilities we provide in TeraGrid, such as "roaming" access and a "coordinated" software environment, open new possibilities in terms of more specialized services, or to allow the TeraGrid, as a system, to respond to supply and demand. For example, a resource provider might elect to increase the "price" of a queue in order to improve turnaround time by reducing demand, or decrease the price to increase demand (and thus utilization).
Traditionally, large scale-up servers used cache-coherent buses for inter-processor communications. These proprietary buses and servers are very costly and power-hungry. Today’s powerful x86 servers replace proprietary scale-up architectures with low-cost machines connected through high-speed, low-latency clustered interconnects. This article will take an in-depth view of their cost and power benefits compared to scale-up architectures, and explain that Ethernet can be tunneled through a PCI Express (PCIe) fabric to provide a very-high-performance, low-cost cluster interconnect suitable for storage IO.
Last week I moderated a webinar entitled Optimizing Performance for HPC: Part 2 - Interconnect with InfiniBand. It was a great presentation with a lot of practical information and good questions. If you missed it, it will be available for a few months, so you still have a chance to check it out. As part of the webinar, Vallard Benincosa of IBM, mentioned that the speed of light was a becoming an issue in network design. In engineering terms, that is refered to as a hard limit.
raditional supercomputer vendors pushing miniaturized versions of their big machines, like Cray with its CX1, or NEC with its SX-9, have definitely been endorsed by pockets of life sciences researchers, but certainly not on any widespread scale. GPU chip maker NVIDIA even has its own personal supercomputer offering in an effort to capitalize on the growing use of graphics chips in scientific computing. And according to market research firms like IDC's High Performance Computing group, personal supercomputers that cluster together GPUs and CPUs are a definite boon to pharmaceutical research shops. And with the steadily climbing growth of workgroup systems selling for less than $100,000, what's the problem?
Multi-million dollar supercomputers take up most of the headlines, but many organizations are now considering the addition of smaller, personal supercomputers to their desktop fleet. Despite some strong global sales, find out why the idea still hasn’t taken off at most companies
I was doing some work and thought, "Wouldn't it be nice to have my own cluster?" I'm guessing not many people have those types of revelations, and probably fewer that decide they should go ahead and solve the problem. I wanted a cheap, small, easy to pack, light, quiet, low-power cluster that I could sit on my desk, and not even think about it.
As computational scientists are confronted with increasingly massive datasets from supercomputing simulations and experiments, one of the biggest challenges is having the right tools to gain scientific insight from the data. One common method for gaining insight is to use scientific visualization, which transforms abstract data into more readily comprehensible images using advanced computer software and computer graphics. But the ever-growing size of scientific datasets presents a significant challenge to modern scientific visualization tools. As a result, there is a great deal of motivation to explore use of large, parallel resources, such as those at the U.S. Department of Energy's (DOE) supercomputing centers, to take advantage of their vast computational processing power, I/O bandwidth and large memory footprint.
The San Diego Supercomputer Center has taken a significant step forward for scientific processing by developing the first of its kind High-Performance Computing (HPC) system which utilizes flash memory. Commonly used in household electronics such as digital cameras and cell phones, flash is generally considered a faster storage medium than traditional hard drives due to the fact that there are no moving parts, as opposed to the traditional drive which stores information on magnetic plates which must be individually accessed.
Philip, a new supercomputer-- named after one of the first Boyd Professors (A Boyd Professorship is the highest and most prestigious academic rank LSU can confer on a professor) at LSU -- chemistry professor Philip W. West, is a 3.5 TFlops Peak Performance 37 compute node cluster running the Red Hat Enterprise Linux 5 operating system. Each node contains two latest Quad Core Nehalem Xeon 64-bit processors operating at a core frequency of 2.93 GHz. Philip was delivered to LSU in May, 2009 and is to be open for general use to LSU users.
IBM's future Power7 chip may be just about done as far as the engineering is concerned, and its server designs might also be more or less completed as well. But there is plenty of time yet to tweak the boxes, and I doubt very much that the final packaging and pricing for the future Power7 machinery is anywhere close to being set. Which is a pity, really.
PelicanHPC is a distribution of GNU/Linux that runs as a "live CD" (or it can be put on a USB device, or it can be used as a virtualized OS). If the ISO image file is burnt to a CD, the resulting CD can be used to boot a computer. The computer on which PelicanHPC is booted is referred to as the "frontend node". It is the computer with which the user interacts. Once PelicanHPC is running, a script - "pelican_setup" - may be run. This script configures the frontend node as a netboot server. After this has been done, other computers can boot copies of PelicanHPC over the network. These other computers are referred to as "compute nodes". PelicanHPC configures the cluster made up of the frontend node and the compute nodes so that MPI-based parallel computing may be done.
"For a while now, IBM has had multiple and competing tools for managing AIX and Linux clusters for its supercomputer customers and yet another set of tools that were used for other HPC setups with a slightly more commercial bent to them. But Big Blue has now cleaned house, killing off its closed-source Cluster Systems Management (CSM) tool and tapping its own open source Extreme Cluster Administration Toolkit (known as xCAT) as its replacement."
The Ohio Supercomputer Center provides supercomputing, research and educational resources to a diverse state and national community, including education, academic research, industry and state government. At the Ohio Supercomputer Center, our duty is to empower our clients, partner strategically to develop new research and business opportunities, and lead Ohio's knowledge economy.
Building and Promoting a Linux-based Operating System to Support Virtual Organizations for Next Generation Grids (2006-2010). The emergence of Grids enables the sharing of a wide range of resources to solve large-scale computational and data intensive problems in science, engineering and commerce. While much has been done to build Grid middleware on top of existing operating systems, little has been done to extend the underlying operating systems to enablee and facilitate Grid computing, for example by embedding important functionalities directly into the operating system kernel.
InfiniBand is a switched fabric communications link primarily used in high-performance computing. Its features include quality of service and failover, and it is designed to be scalable. The InfiniBand architecture specification defines a connection between processor nodes and high performance I/O nodes such as storage devices. InfiniBand forms a superset of the Virtual Interface Architecture.
Numascale's SMP Adapter is an HTX card made to be used with commodity servers with AMD processors that feature an HTX connector to its HyperTransport interconnect.
FUSION1200® is a scalable 8 to 32-processor SMP system for the High Performance Technical Computing (HPTC) market. Available in both deskside and 19-inch rack-mount design, the FUSION1200® is a scalable alternative to traditional RISC based servers. FUSION1200® Series is an enterprise-class system for an IT department looking to leverage the benefits of Intel® standards in a data center. With the flexibility to grow from 8 to 32 Intel® Xeon® processors (quad or dual core), the FUSION1200® Series scales beyond conventional Intel® based platforms while delivering superior price-performance compared to traditional high-end servers. The SMP operational model of the FUSION1200® provides reduced management costs compared to clusters. This Intel® Xeon® processor based server, supporting Intel® Extended Memory 64 Technology and the ScaleMP® vSMP architecture, is the ideal platform for clients with applications that require high processor count and large shared memory.
Engineers, scientists, researchers and other workstation users are continually challenged with more complex problems and shorter deadlines in which to solve them. The Cray CX1-iWS™ solution addresses both of these issues by allowing larger models and simulations to be worked within the workstation environment, eliminating the need to move the problem to an external shared resource cluster, and providing an easy to setup solution.
Giving users more flexibility in how they configure systems to attack various workloads was a big thread running through SC09 last year. At the show, we took at look at three different companies who are, in one way or another, providing large system images. (Click to see our posts on ScaleMP, 3Leaf, and SGI.)
an Miller joined Cray in February 2008 and currently heads up Cray’s Productivity Solutions group with the recently introduced Cray CX1 desk side supercomputer – an Intel Cluster Ready product. Mr. Miller also leads Cray’s corporate marketing organization. Prior to joining Cray, he served as the Vice President of Polyserve Software at HP and as Vice President of Worldwide Sales for PolyServe prior to its acquisition by HP. Before joining PolyServe, Mr. Miller was Vice President of Worldwide sales for IBM High End xSeries Servers where he worked for both IBM xSeries and pSeries organizations, with a particular focus on marketing and sales for high end Intel based systems. Prior to IBM, he was Vice President of Global Marketing for Sequent Computer Systems, and Vice President Asia Pacific. Miller has also worked for Software AG as Senior Vice President Asia Pacific, and for Unisys in many capacities, ending as General Manager for Asia South. Mr. Miller is a Graduate of London University.
ScaleMP, a maker of virtualization and aggregation software that allows a cluster of x64 servers to look like a big, bad, symmetric multiprocessing (SMP) shared-memory system to operating systems and selected classes of applications, is going downstream to target SMBs and upstream to chase cloud infrastructure providers.
Rocks adds a vastly expanded solutions layer (Rocks HPC, Rocks Cloud, Rocks Rolls) and enterprise-class support, which transforms the leading open source cluster distribution into a production-ready cluster operating environment suitable for data centers of all shapes and sizes. Clustercorp also partners with the industry's leading workload management providers to offer Rocks MOAB, Rocks LSF, and Rocks SGE. Purchase turnkey Rocks clusters from a long list of reliable hardware partners including HP, Dell, Cray, Silicon Mechanics, and more.
Computerworld - High-performance computing (HPC) has almost always required a supercomputer — one of those room-size monoliths you find at government research labs and universities. And while those systems aren’t going away, some of the applications traditionally handled by the biggest of Big Iron are heading to the desktop. One reason is that processing that took an hour on a standard PC about eight years ago now takes six seconds, according to Ed Martin, a manager in the automotive unit at computer-aided design software maker Autodesk Inc. Monumental improvements in desktop processing power, graphics processing unit (GPU) performance, network bandwidth and solid-state drive speed combined with 64-bit throughput have made the desktop increasingly viable for large-scale computing projects.
OSCAR allows users, regardless of their experience level with a *nix environment, to install a Beowulf type high performance computing cluster. It also contains everything needed to administer and program this type of HPC cluster. OSCAR's flexible package management system has a rich set of pre-packaged applications and utilities which means you can get up and running without laboriously installing and configuring complex cluster administration and communication packages. It also lets administrators create customized packages for any kind of distributed application or utility, and to distribute those packages from an online package repository, either on or off site.
Welcome To The PS3cluster Guide Our community guide allows you to set up your own MPI (Message Passing Interface) based supercomputer cluster with the Playstation 3. This guide was co-written by Gaurav Khanna, based on his previous work on the Gravity Grid and is a current run-time environment for the research of co-author (Chris Poulin), based on his current work in distributed pattern recognition. As such, we currently utilize the Fedora Core for this infrastructure and illustrate a "how-to" below. NOTE: We focus on the Fedora 8 distribution, due to prevalence of Fedora and its Cell SDK (3.0) compatibility. Finally, this content should be considered open source, and here is the license.
J. Eberle, M. Schwinger, und H. Zwenzner. Proceedings of the 2021 conference on Big Data from Space, Seite 89--92. Publications Office of the European Union, (Mai 2021)