Share |

Content about Infrastructure

August 29, 2007
iSGTW Feature - Joining the dots: creating interoperable grids

  Feature - Joining the dots: creating interoperable grids Interoperability is now making grids bigger and more able than ever before.Images courtesy of OSG (top) and Marijke Unger (bottom)Pooling computers into grids is enabling scientific discoveries that would not otherwise be possible. Always looking to the future, grid developers are now asking: is there a compelling case for connecting these grids together?As grid usage grows and more researchers discover grid power, many predict that scientific appetite for this power will also grow. Already satisfying large appetites as individual ventures, two U.S.-based grids—Open Science Grid and TeraGrid—agree that interoperability between their infrastructures is the next step.The interoperative idealTo a scientist, grid interoperability should mean nothing beyond reliable and continuous access to highly responsive computing and data storage resources. These resources, provided by one or any number of computing grids, could ideally respond interchangeably to user requ

August 29, 2007

  Feature - Thailand goes tera: scientists connect to the power of ThaiGrid ThaiGrid powers this tsunami simulation, run by a team from Chulalongkorn University, Thailand, and focusing on the potential effects of tsunamis in the Indian Ocean.Image courtesy of ThaiGrid Scientists across the globe are cranking up their computing requirements: they want more processing power to tackle ever more challenging problems. The wishes of Thai scientists recently came true when their first teraflop computing system went online as part of the Thai National Grid Project.“Our demand for computational power and storage is increasing rapidly,” says Putchong Uthayopas, director of the Thai National Grid Project. “We had a need for a national computational infrastructure that would allow researchers to discover and explore more challenging problems,” he says. “The Thai government decided that now was the right time to explore grids as a key infrastructure enabler.”80 systems across 15 sites Comprising 1

August 29, 2007
iSGTW Opinion - Grids can be green

  Opinion - Grids can be green Piles of cables and computer waste wait to be scrapped or reused in Guiyu, China.© Greenpeace, Guide to Green Electronics David Wallom is the technical manager of the Oxford e-Research Centre at the University of Oxford and chair of the UK e-Science Engineering Task Force. He is currently working on a JISC-funded Low Carbon Information and Communications Technology project. Can grids really be green? Grids have been designed to provide collaborating researchers with resources from many different physically disparate organisations. This has the advantage of ensuring that high-cost resources—in terms of initial purchase price and running costs—are operated as near to maximum capacity as possible. This means that, through correct management of related resources, grids have the potential to effectively lower an organisation’s overall carbon footprint.Disposing with pretenceHowever, we should also recognize the natural resources consumed to build these computing systems

August 22, 2007
iSGTW Technology - Weka4WS: distributed data mining using web services

  Technology - Weka4WS: distributed data mining using web services The Waikato Environment for Knowledge Analysis, or WEKA, is software developed at the University of Waikato, New Zealand. It gets its four-letter acronym from New Zealand’s native weka, flightless brown birds about the size of a chicken. Stock image from sxc.hu Released in June 2007, Weka4WS is a new tool designed to open the way for worldwide use of data mining services. Developed at the University of Calabria Grid Computing Lab, Weka4WS extends the open source Weka toolkit for supporting distributed data mining on grid environments. The original Weka provides a large collection of machine learning algorithms, written in Java, for data pre-processing, classification, clustering, association rules and visualization, which can be invoked through a common Graphical User Interface. In Weka, the overall data mining process takes place on a single machine, since the algorithms can be executed only locally. Weka4WS extends Weka to support remote grid execution

August 1, 2007
iSGTW Technology - GridWay: interoperability without the headache

  Technology - GridWay: interoperability without the headache The GridWay metascheduler can schedule jobs across several grids, including Open Science Grid, EGEE and TeraGrid.Image courtesy of GridWay What looks and feels like your Local Resource Management system, but lets you submit jobs to multiple heterogeneous grids? Sounds like the beginning of a bad joke or pipe dream? You might need to upgrade your metascheduler.So says Tino Vazquez Blanco, a grid technology engineer working on the GridWay metascheduler project.“GridWay allows you to do much more than your local LRM system. You can of course use GridWay to submit, monitor, and control your jobs; the added bonus is that you can do this across several different grids.”GridWay is a scheduler for other schedulers, Blanco explains. “It simulates a familiar environment, so it’s as if you’re working on a local cluster, but in fact you’re working across grids formed by heterogeneous clusters. You don’t need to worry about where the cl

July 25, 2007
iSGTW Feature - Data for the people: how to share the data love

  Feature - Data for the people: how to fast-track your network and share the data love “You don’t have to have a PhD to be interested in data and to want to analyze it. ” Bob Grossman is working to make large data sets easily accessible and publicly available, even to children.Stock image from sxc.hu So you’ve used a grid to split up your job, process it faster, then return your results. You now have a nice chunky terabyte of data. What do you do with it?Bob Grossman, Director of the National Center for Data Mining at the University of Illinois, Chicago, U.S., says the answer is share, share, share.“In terms of impact on society, the ability to use transparently other people’s data is going to be transforming,” Grossman says. “It is about ‘network effects’,” he continues. “In the same way that a network becomes more interesting as more people join it, you can draw more interesting conclusions about your own data if you put it into the context of other peo

June 27, 2007

  Feature - Adaptive fault tolerance for improved reliability Image generated using cosmology modeling software Enzo. FT-Pro is designed to help programs like Enzo run for longer by adaptively selecting a best‐fit action in response to failure prediction. Image courtesy of Mike Norman at University of California, San Diego and Greg Bryan at Columbia University, New York, U.S.  So there’s the good news. And then there’s the bad news. The good news is that high performance computing systems are getting bigger. And the bad news? As system size increases, Mean Time Before Failure is dramatically reduced: the number of hours you can run your application before everything grinds to a halt just keeps getting smaller. Yawei Li and Zhiling Lan of the Illinois Institute of Technology, U.S., want to change all that. They have developed an adaptive fault management scheme, called FT-Pro, which has already improved the robustness of several real-world applications run on the TeraGrid, including

June 6, 2007
iSGTW Feature - Going Mobile: Extending Grid to the Lower End

  Feature - Going Mobile: Extending Grid to the Lower End Community-based applications may offer a ready doorway to the mass popularization of grid-related technologies.Image courtesy of Stavros IsaiadisThink your cell phone has nothing to contribute to grid services?  Think again. Stavros Isaiadis of the University of Westminster is working on the so-called “lower end” of the grid performance spectrum, as part of CoreGRID, and the university’s Distributed and Intelligent Systems Group, led by Vladimir Getov.“Grid usually revolves around high performance computing,” explains Isaiadis, “which means you need raw resources—things like CPU and memory. Framed like this, mobile devices can only make a very limited contribution.” But, Isaiadis asks, what about flexibility, agility, mobility? “Mobile devices can extend grids into areas where static grids can not go. You can leverage resources from many devices, like from WiFi hotspots or conference rooms, to create an ad ho

June 6, 2007

Image of the Week - Science on Top of the World  Scared of heights? Don't look down. And don't get too excited about all those gamma rays and start hyperventilating: the partial pressure of oxygen at this laboratory is about half that at sea level.Image courtesy of EUChinaGRIDBeing 4,300 meters above sea level brings you closer not only to the stars, but also to the things stars spit out: gamma rays. For researchers eager to detect these gamma rays before the rays are absorbed by the atmosphere, this is a good place to be, or at least, it is a good place to put your equipment. In Tibet, 90 kilometers north of Lhasa, sits the High Altitude Cosmic Ray Laboratory. The lab, run by a Chinese-Italian collaboration, will use the EUChinaGRID infrastructure to store and transfer data so it can be accessed by collaborators around the world.EUChinaGRID expects to transfer 250 terabytes a year to Beijing, and then on to Italy, at the steady rate of around 10 megabytes per second.

April 25, 2007

Statistic of the Week - 250 CERN computer center operator at an IBM 370 in the early 1980s.Image copyright CERNIn 1990 the CERN Computer Center contained less than 250 CERN Computing Units. The CERN Unit of processing power was a standard proposed in 1985 by Eric McIntosh of the CERN computing division, then called “DD” for Data-handling Division. One CERN Unit (CU) corresponded to the power of a single CPU of an IBM 370 Model 168, or to that of a DEC Vax 8600. It was based on execution speed of a standard set of HEP code assembled by McIntosh. This code set was later incorporated as part of the SpecINT standard, a world measure for CPU power which is in general use today.The CERN Unit was equivalent to about 4 Million floating point operations per second (i.e. 4 Mips), or roughly 2 units of the more modern SpecINT2000 standard.The approximate 250 CERN Units in 1990 came primarily from the IBM complex (about 150 units), the Cray XMP-48 (about 50) and the biggest DEC Vax, the Model 9000 (contributing about 10).

April 18, 2007

Image of the Week - India’s National Grid Computing InitiativeGARUDA will link 45 institutions in 17 cities via a high-speed network to promote science and computing in India. Image courtesy of GARUDAGARUDA —which takes its name from a large, bird-like creature in Hindu and Buddhist mythology—is a collaboration of researchers and experimenters aiming to establish a nationwide grid in India. Currently in its proof of concept phase, GARUDA will link centers for computation, mass storage and scientific instruments to enable data and computing intensive science in India for the 21st century. The GARUDA high-speed network will connect 45 institutions in 17 cities at 10 to 100 megabits per second bandwidth. An interactive version of this map is available online. The GARUDA project is coordinated by Center for Development of Advanced Computing. C-DAC is a partner of EUIndiaGrid project. Interoperability between GARUDA and EGEE infrastructure is a main goal of EUIndiagrid.

April 11, 2007

Feature - Pacific Rim Researchers Collaborate to Study Avian Flu Nations with confirmed cases of the H5N1 avian influenza as of February 2007.Source http://pandemicflu.govResearchers at the University of California, San Diego and the University of Hawaii will use bioinformatics, grid computing and networking infrastructure, as well as collaborative ties to Asian institutions, to learn more about avian flu, in hopes of helping to head off a much-feared pandemic in the region of the world where the disease has already cost human lives.“We will use modern high-throughput biology to annotate the biological structures of different subtypes of the avian influenza virus, at the same time as we study their variations,” said principal investigator Peter Arzberger, director of Life Science Initiatives at UC San Diego. “We will also construct a grid infrastructure to support avian flu research—an infrastructure that could one day handle research on other infectious diseases as well.”“Fighting a pandemic will

April 4, 2007
iSGTW Feature - NYS Grid: Spreading the Word

Feature - NYS Grid: Spreading the WordImages of a cholestosome model, generated with NYS Grid resources.Image courtesy of Mary McCourt, Niagara UniversityLast summer a collection of universities and labs in New York State gathered with the goal of creating a cyberinfrastructure initiative to make it easy for regional scientists to manage data and use imaging software. This seven-month-old cyberinfrastructure, with New York State Grid as its foundation, may be in its infancy but it is growing, with lofty goals. “We are trying hard to get users in New York state to take the plunge and use the Grid,” says Russ Miller, professor at SUNY-Buffalo and executive director of the cyberinfrastructure initiative. NYS Grid uses the Open Science Grid software stack; Miller and his team are developing middleware to run on top of OSG software to support applications in biomedical computing, molecular structure studies and structural biology. Mary McCourt, Niagara University’s chemistry department chairperson, uses NYS Grid in her

March 21, 2007
iSGTW Feature - Grids Aid Battle Against Cancer

Feature - Grids Aid Battle Against Cancer A digital image such as this one, showing lesions in a patient's liver, can now be distributed among participating hospitals through a grid. Image courtesy of Ignacio BlanquerSpanish medical staff in the region of Valencia have a new weapon in the arsenal against cancer. In January, a team of medical and computing researchers launched a grid-enabled infrastructure that shares oncological studies, digital images and reports among five hospitals in the Valencia region.“In the field of oncology, quantification is very important for diagnosis,” says Ignacio Blanquer, associate professor of computer science and researcher in the High Performance Computing and Networking Group, at Universidad Politécnica de Valencia in Spain. “The more information the radiologist has at hand, the more objective and accurate the diagnosis.” On March 22, the team Blanquer works with, led by Vicente Hernandez, will present the deployment of CVIMO, the Valencian Cyberi

March 14, 2007

Feature - Europe looks forward to the European Grid Initiative A video about EGI has been created by the EGEE project. Image courtesy of EGEE. The European grid community is thinking about the future. Today many of the grid projects that are co-funded by the European Commission are limited-term national projects. For stability, the research community using these grids needs a long-term body to oversee European grid computing infrastructure. To this end, a workshop on the proposed European Grid Initiative was held in Munich on February 26-27, to discuss a EGI design project proposal to be submitted to the EU. The workshop was hosted by Deutsche Forschungsnetz, or DFN, the German research and education network, as part of their contribution to the EGEE project.“EGI most likely will be a body whose constituency is the National Grid Initiatives currently in participating countries,” said Klaus Ullmann, General Manager of DFN and host of the workshop. “Close to thirty countries were represented at the workshop. Th

February 21, 2007

Feature - Ensuring a Grid Without Boundaries The EUGridPMA members at the CCLRC Coseners House in Abingdon, England.Image courtesy of PPARC.  With more than 30 national Grids in Europe alone, the European Policy Management Authority for Grid Authentication must ensure that electronic identities will work across any boundaries. At last month’s meeting, its first ever held in the UK, the growing EUGridPMA considered bids for accreditation from Europe and Africa.“I was very pleased to welcome the EUGridPMA to the UK for the first time, especially as the PMA expanded its horizons,” said David Kelsey of CCLRC Rutherford Appleton Laboratory and GridPP in the UK. “We are now closer to enabling access to the grid to all scientists wherever they are located.”Kelsey was an original organizer of efforts to build international trust among grids. EUGridPMA is one of three regional PMA’s within the International Grid Trust Federation. IGTF establishes requirements and best practices for identity manageme

February 21, 2007

Feature - Widening the Gateway to India The “Gateway to India” in Mumbai. By typing in a simple command on Monday 12 February 2007, Gavin McCance initiated the first file transfer between CERN and India on the Worldwide LHC Computing Grid. Connection via the WLCG will give Indian researchers the same level of access to Large Hadron Collider particle physics experiments as researchers in European or North American countries. “It’s a good feeling to be involved in a collaboration like this,” says McCance,  of the WLCG and CERN. The file transfer, containing one gigabyte of test data, reached India in less then seven minutes and represents the first drop of an imminent deluge on the WLCG transfer service. Once the CERN Large Hadron Collider is generating particle physics data, the WLCG, which runs on EGEE and Open Science Grid infrastructures, will be needed to manage and process the files. Since the amount of data generated will be larger than one institution could handle, the task of storing a

February 7, 2007

Feature: San Diego Supercomputer Experts HelpNavajos Build “An Internet to the Hogan” Leonard Tsosie (center), a Navajo senator and a leader of the Internet to the Hogan Project, uses a laptop to explain the project's benefits to friends at their traditional dwelling, or Hogan. Supercomputer experts from UC San Diego will help end the "digital divide" for many in the Navajo Nation in the Southwest.  Image courtesy of SDSC.Navajos in the American Southwest, many of whom have never had access to a personal telephone, will soon make a significant leap into the Internet Age, thanks in part to resources and expertise provided by the San Diego Supercomputer Center at the University of California, San Diego.The Navajos, who refer to themselves as the “Dine” (dee-nay), celebrated “An Internet to the Hogan and Dine Grid Event” on Monday, January 29, at Navajo Technical College in Crownpoint, New Mexico. Highlights of the event include their official acceptance of a “Little Fe&rdquo

December 20, 2006

Feature: Building Japan’s Grid The Institute for Molecular Science helps nanoscience researchers use NAREGI grid technology. This image shows a chemical reaction between solute molecules in water. Institute for Molecular Science NAREGI, Japan’s National Research Grid Initiative, started in 2003 as a project to develop grid middleware and prove the technology’s usefulness to scientific communities and the business world. More recently, NAREGI’s goals have expanded to include supporting two nationwide computing infrastructures. “NAREGI’s grid middleware is one of the key component technologies for the Japanese Cyber Science Infrastructure framework,” explains NAREGI Project Leader Kenichi Miura. “We will also provide the infrastructure for access and data sharing within the Petascale computing system when it becomes available in 2010.” Both the Cyber Science Infrastruct

December 20, 2006

Feature: Government to grid Aleksandar Belic (left) with EC Commissioner for Science & Research Janez Potočnik (center) during a visit to the Belgrade Institute of Physics. It is well known that countries like the UK and Germany have invested heavily in grid technology, but they are not alone. Countries in South Eastern Europe have also enthusiastically embraced grid technologies. Aleksandar Belic, head of the Scientific Computing Laboratory at the Institute of Physics in Belgrade, has been one of the leaders in bringing grid infrastructures to the region. Unusually for a scientist, Belic has governmental experience that has proved invaluable in his work with projects such as Enabling Grids for E-sciencE. “I spent three years as Assistant Minister of Science for Serbia, working on international collaborations,” says Belic. “I was very proud to have helped to pave the way for Serbian participation in the

November 29, 2006

Feature: Digital Libraries Go Dynamic
Sites of the DILIGENT Pre Production Service, including gLite sites (green) and DILIGENT Hosting Node Managers (yellow).
Image courtesy DILIGENT
WhatÂ’s digital, can manage any information you might have, but doesnÂ’t exist until you ask for it? Digital libraries, according to Pasquale Pagano from the Digital Library Infrastructure on Grid Enabled Technologies project.
“DILIGENT wants to go beyond static digital libraries to provide an infrastructure to create and manage dynamic, virtual digital libraries using grid technology,” explains Pagano, the project’s technical director.
A prototype system for creating such dynamic digital libraries has been achieved by deploying the DILIGENT tools and services on top of a basic grid infrastructure layer. The infrastructure comprises sites from DILIGENT and from the Enabling Grids for E-sciencE projectÂ’s Pre-Production Service.
"DILIGEN