OSG August Newsletter, 2009

 

OSG Science Reports

Recently we were asked to give two science accomplishments benefitting from the OSG. I have included them here for your interest, together with an interesting editorial from Frank Würthwein.

Physics results from the CDF and D0 Tevatron experiments benefit from sharing processing and storage resources that are made accessible to the OSG infrastructure, especially those from their competitor experiments waiting for data from the LHC. These results include increased constraints on the mass of the Higgs particle—excluding the range between 160 and 170 GeV/c2, whereas previous experiments predicted a mass between 114 and 185 GeV/c2. (*See Frank's note below.) This is the first improvement in the Higgs mass prediction over several years. The configuration of opportunistically available processing and storage allows this to happen naturally as use by other scientific communities that own resources fluctuates. The OSG will continue to contribute to the “scientific return on investment” of the Tevatron as data analyses continue during the ramp-up of the LHC. Read Fermilab press release for more information.

Scientific results from non-physics OSG users have increased as a result of improved usability and robustness of the infrastructure, advances in the “just in time” job scheduling and processing technologies, and software from the OSG’s VDT. During 2008-2009 the number of non-physics users of the OSG increased with publications of results from molecular dynamics, mathematics, protein prediction and computer science.

~Ruth Pordes

*Basically, the easiest way to find the Higgs at both the Tevatron and the LHC is in its decay H->WW -> 2 leptons and 2 neutrinos. This final state is most sensitive—slightly above twice the W-mass, or 165GeV. The Tevatron sees no evidence of Higgs at that mass, and thus rules out a mass around that value. As the Tevatron experiments take more data, and assuming nature chose a Higgs mass far away from 165GeV, then the Tevatron exclusion region will grow. The top mass measurements from the Tevatron, when combined with electroweak (EWK) data (mostly from CERN’s LEP collider) constrains the Higgs from above. The LEP search constrains the Higgs mass from below. The new Tevatron results constrain the Higgs mass in the middle of the region allowed by the previous two constraints. For the Tevatron to make this exclusion is a big deal because it keeps the Tevatron in the Higgs business for longer.

~Frank Würthwein

____________________________________________________________________________________________________________________________

Superlink-Online

Knowing which genetic mutation within a family causes a disease can lead to lifestyle changes or to the creation of disease treatments. But the type of computing analysis needed to identify the mutations can take the equivalent of years to complete on a single computer.

Geneticists use the statistical method, genetic linkage analysis, to determine the location of these mutations on the chromosome. Based on a given genealogy and its members’ genetic makeup, the analysis is mapped onto a probabilistic graphical model that represents the likelihood of a genetic marker being linked to the disease. In large families with many genetic markers, these computations are extremely compute-intensive.

Superlink-online, developed at the Technion-Israel Institute of Technology, helps researchers perform their analyses in a matter of days. Geneticists submit their data through the web portal , while behind the scenes, computations are split into hundreds of thousands of independent jobs, invoked by available resources, and assembled back into a single data set.

Technion computer scientist Mark Silberstein and colleagues released the first version of Superlink-online in early 2006. Originally using UW-Madison’s and their own Condor pool with about 100 CPUs, the team eventually made use of additional resources from the OSG, EGEE, and the Superlink@Technion community grid.

During a 3 month period, over 25,000 non-dedicated hosts actively participated in the computations, reaching maximum effective throughput roughly equal to that of a dedicated cluster of up to 8,000 cores.

Increased throughput has resulted in the discovery of several rare-disease-causing mutations, including those causing Hereditary Motor and Sensory Neuropathy and Ichthyosis.

A new version is being finalized that will access resources over all aforementioned grids and Tokyo Tech’s Tsubame supercomputer. A new resource management system, GridBot, unifies these into a single scheduling framework. (GridBot will be presented at Supercomputing in November.)

Read more about Superlink-online.

~Marcia Teckenbrock

____________________________________________________________________________________________________________________________

OSG Site Administrators Workshop

The OSG Site Administrators workshop took place August 6th and 7th. Based on experience and feedback from previous events, the workshop was “hands-on,” with a single plenary session to introduce OSG to the new Tier-3 administrators, to describe what’s new in version 1.2, and to present the parallel sessions.

Forty-five site administrators participated, taking advantage of the expertise of the instructors and other site administrators. Meeting topics included installation, security, network troubleshooting, virtual machines, and monitoring.

“Install-fest” sessions successfully accelerated the deployment of OSG 1.2, released one week before and especially important for the LHC experiments. Before the meeting, only one production resource had version 1.2, but several services were installed during the meeting, including at least six GUMS servers, six computing elements and five storage elements. One day after the workshop, eight OSG 1.2 production sites were “green” in MyOSG.

The material produced, is available for those who were unable to attend.
This includes: a series of tutorials, presentations , and some recordings of the sessions .

~Marco Mambelli

____________________________________________________________________________________________________________________________

OSG Sponsored Secure Programming Workshop

The workshop, “Secure Programming and Vulnerability Assessment of Distributed Computing Middleware,” will be co-located with OGF 27 in Banff, Canada on October 12th, 2009. Organized by the OSG security team and researchers from the University of Wisconsin-Madison, Professor Bart Miller and his team from UW-Madison will be presenting.

This workshop is relevant to those who want to learn about analyzing software for security flaws and for developers wishing to minimize vulnerabilities. A variety of topics will be covered, from offensive security matters (how to find problems by using proactive vulnerability assessment) to defensive security (how to prevent problems by seeing many types of vulnerabilities that occur in code and learning what techniques can prevent them). Real-life case studies and a security assessment of the Condor project will also be presented.

All OSG software providers are encouraged to apply for participation at the workshop web site OGF 27 . Please note the same workshop will also be given during EGEE09 conference (still unconfirmed by EGEE09 organization committee). We strongly encourage anyone who cannot make it to Banff to attend the workshop at EGEE09.

~ Mine Altunay

____________________________________________________________________________________________________________________________

OSG Enterprise Document Repository

The OSG enterprise document repository, also known as DocDB , includes the official governance documents of the OSG. This summer we completed a project to recategorize documents in the repository. We hope that this will make it easier to find documents of interest. While the OSG twiki contains technical documentation, the repository is the place for official documents, such as OSG-wide policies and blueprints, outreach materials, and OSG published papers.

We encourage you to upload documents of this nature to the repository. To help you properly categorize your documents, we have some topic guidelines posted. DocDB requires your IGTF certificate. The first time you access the non-public database to update or post a document, you will be prompted to apply for access. This is a quick and easy one-time process that requires only your email address and the access group/s you need membership in.
(We will assist you if you are unsure of which group/s you need.)

~Marcia Teckenbrock

____________________________________________________________________________________________________________________________+

Meet the new iSGTW U.S. Editor

Grids are made great by collaboration. It’s a quality shared by the publication I edit, International Science Grid This Week.

iSGTW is the fruit of a longstanding collaboration between Europe and the United States. Within the U.S., a variety of organizations are contributing money and effort to produce the publication. Open Science Grid is a major contributor, and without OSG’s support, I would not be here today.

I come to iSGTW with a background in physics and journalism. I’ve published articles on topics ranging from mathematical medicine to high-energy cosmic rays, and although my work has appeared in newspapers and magazines, most recently I have focused my attention on online journalism. It is a mixture of experience which I hope will be valuable for iSGTW in the years to come.

Every week, iSGTW brings you interesting stories about science empowered by distributed computing. And because we are supported by such a wide variety of organizations, our readers can trust that we choose our stories regardless of which grid or computing centers participated in the research.

Every week, scientists read about distributed computing-powered research in a wide variety of disciplines. Since an astrophysicist may not know much about the work of a microbiologist, the reading level is meant to appeal to an educated general audience.

Every week, journalists and representatives of funding agencies read iSGTW. They read the newsletter because of the accessible reading level, and the wide spectrum of science topics we cover.

That’s why I invite you all to email me about your research at miriam.boon@isgtw.org. Help me keep that spectrum of science topics broad, and in the process, get some exposure for your research. I look forward to talking to you!

~Miriam Boon

____________________________________________________________________________________________________________________________

OSG Staff Retreat

We held our annual staff retreat the last week of July and are busy planning for next year’s work. Information generated from the retreat and plans we’ve made since then are available off the meeting agenda page, noted under the attachments link.

Please contact any of the area coordinators if there are activities and/or items your organization would like to collaborate on, whether user, site, or software. As a reminder, here is the list of people working on OSG activity plans for next year:

1.0 OSG Project – Ruth, Miron
1.0.1 Software Tools Group – Alain Roy, Mine Altunay
1.0.2 Production Coordination – Dan Fraser
1.0.3 Biology & Biomedical – Piotr Sliz
1.0.4 LIGO - Patrick Brady, Chander Sehgal
1.0.5 CMS - Lothar Bauerdick, Ian Fisk, Burt Holzman, Miron Livny, Frank Wuerthwein
1.0.6 ATLAS - Michael Ernst, Rob Gardner, Ruth Pordes, Jim Shank, Torre Wenaus
1.1 Software – Alain Roy & Tanya Levshina
1.2 Operations – Rob Quick
1.3 Integration and Sites – Rob Gardner
1.4 VOs Group – Abhishek Rana
1.5 Engagement – John McGee
1.6 Campus Grids – Sebastien Goasguen
1.7 Security – Mine Altunay
1.8 Training & Content – Jim Weichel
2.1 Usability and Scalability – Igor Sfiligoi
2.2 Work Load Management – Maxim Potekhin
2.4 Internet2 – Rich Carlson
3.1 Metrics – Brian Bockelman
3.2 Communication & Education – David Ritchie
3.3 Project Management – Chander Sehgal

~Ruth Pordes

 

OSG Newsletter Archive