HPCwire: STEP'09 Demonstrates LHC Readiness

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Translation Disclaimer

Subscribe | Sign In

Visit additional Tabor Communication Publications

Leading HPC
Solution Providers

July 01, 2009

STEP'09 Demonstrates LHC Readiness

GENEVA, July 1 -- Preparations are under way for the restart of the Large Hadron Collider (LHC), the world's most powerful particle accelerator. One of the most important systems needed to support the experiments that will utilise this great machine is the global computing grid: the worldwide LHC Computing Grid (WLCG). After months of preparation and two intensive weeks of 24 x 7 operation, the LHC experiments are celebrating the achievement of a new set of goals aimed at demonstrating full readiness for the LHC data taking run expected to start later this year.

Whilst there have been several large-scale data-processing tests in recent years, this was the first production demonstration involving all of the key elements from data taking through to analysis. Records of all sorts were established: data taking throughput, data import and export rates between the various grid sites, as well as huge numbers of analysis, simulation and reprocessing jobs -- ATLAS alone running close to 1M analysis jobs and achieving 6GB/s, of "grid traffic," the equivalent of a DVD worth of data a second, sustained over long periods.

This result is particularly timely as it coincides with the transition of grids into long-term sustainable e-infrastructures, clearly of fundamental importance to projects of the lifetime of the LHC. With the restart of the LHC only months away, one can expect a large increase in the number of grid users: from several hundred unique users today to several thousand when data taking and analysis commences. This can only happen through significant streamlining of operations and the simplification of end-users' interaction with the grid. Scale Testing for the Experiment Programme '09 (STEP'09) included massive-scale testing of end-user analysis scenarios, including "community-support" infrastructures, whereby the community is trained and enabled to be largely self-supporting, backed by a core of grid and application experts.

WLCG combines the IT power of more than 140 computer centres, the result of collaboration between 33 countries.

Sergio Bertolucci, director of research and computing at CERN, said: "The four LHC experiments -- ATLAS, CMS, ALICE and LHCb -- have demonstrated their ability to manage their nominal data rates concurrently. For the first time all aspects of the experiments' computing were exercised simultaneously: simulation, data processing and analysis. This gives them the confidence that they will be able to efficiently analyze the first data from the LHC later this year."

Bob Jones, director of the EGEE project, remarked: "Such a significant achievement is also a valuable testament to the state of maturity of the EGEE infrastructure and its ability to interoperate with major grid infrastructures in other parts of the world. Ensuring that this level of service continues uninterrupted as we transition from EGEE to EGI is clearly essential to our users, including flagship communities such as high energy physics."

"This is another significant step to demonstrating shared infrastructures can be used by multiple high throughput science communities simultaneously," said Ruth Pordes, executive director of the Open Science Grid consortium. "ATLAS and CMS are not only proving the usability of OSG, but contributing to maturing national distributed facilities in the US for other sciences."

David Britton, the GridPP Project Leader, reported: "In the UK, STEP09 ran very smoothly at the majority of sites, which allowed the focus to be on understanding the performance and tuning the infrastructure. The RAL Tier-1 performed exceedingly well with only a single out-of-hours call out over the two week period. Valuable information was obtained on the performance of tape-drives under realistic workflows; the OPN network was tested by laying down additional UDP traffic on top of the STEP09 data; and the fairshare system was successfully tuned to balance the load between experiments."

Gonzalo Merino, manager of the Tier1 centre in Barcelona, wrote: "The Spanish WLCG sites met the STEP09 targets. It has been a very valuable exercise since many of the experiment workflows have been tested simultaneously at unprecedented scale, well above the nominal values for LHC data taking. The Tier-1 at PIC has provided a very stable and reliable service at record breaking levels: exchanging up to 80 terabytes per day with other WLCG sites and processing data at more than 2 GBytes per second. This gives us confidence that the Spanish WLCG sites are ready for data taking."

David Foster, head of the LHC Optical Private Network activity, concluded: "The LHC Optical Private Network transporting data between the sites has proven its capability both in terms of performance and resiliency during STEP'09. New capabilities emerging in the 40Gbps and 100Gbps range should enable us to keep up with the anticipated data distribution needs of the LHC experiments."

About the Large Hadron Collider

The LHC, located at CERN near Geneva, Switzerland, is the world's largest particle accelerator. For thousands of physicists, analysing LHC data using the LHC Computing Grid will be like sifting for digital gold. Their search is predicted to unearth evidence of new fundamental particles that will provide clues to the ultimate nature of matter and the origins of our Universe.

About grid computing

Grid computing connects computers distributed over a wide geographic area. Just as the World Wide Web enables access to information, computing grids enable access to computing resources. These resources include data storage capacity, processing power, sensors, visualisation tools and more. Grids can combine the resources of thousands of different computers to create a massively powerful computing resource, accessible from the comfort of a personal computer and useful for multiple applications in science, business and beyond.

About CERN

CERN, the European Organization for Nuclear Research, is the world's leading laboratory for particle physics. It has its headquarters in Geneva. At present, its Member States are Austria, Belgium, Bulgaria, the Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Italy, Netherlands, Norway, Poland, Portugal, Slovakia, Spain, Sweden, Switzerland and the United Kingdom. India, Israel, Japan, the Russian Federation, the United States of America, Turkey, the European Commission and UNESCO have Observer status.

-----

Source: CERN

Share Options

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Subscribe to All Content

All

May 23, 2012

May 22, 2012

May 21, 2012

May 18, 2012

May 17, 2012

May 16, 2012

May 15, 2012

May 14, 2012

May 11, 2012

Feature Articles

NVIDIA Works On CPU Co-Dependency Issues with Kepler GPU

NVIDIA is telling everyone that the GK110, its new Kepler GPU aimed at supercomputing, is all about improving performance per watt. But the other driving theme behind the new architecture is reducing the GPU's reliance on its CPU host. How well it accomplishes both these goals areas could determine the success of the new chip in high performance computing.
Read more...

OpenACC Starts to Gather Developer Mindshare

PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...

NVIDIA Launches Kepler Into HPC

NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...