May 20, 2008 14:00-18:00
BMC Auditorium, Habarzel 6, Ramat Hachayal. Tel Aviv
To reserve your place, please send your contact details to: info@grid.org.il
Agenda
14:00 – 14:20
Welcome
Avner Algom, IGT GM
Assaf Marron, PhD, Corporate Architect, BMC Software
14:20 – 15:00
The Condor Road Map & the Red Hat agreement
Prof. Miron Livny, Computer Sciences Department, University of Wisconsin
The Head of the Condor project
Red Hat MRG Grid is based on the Condor Project started and hosted by the University of Wisconsin-Madison. Red Hat and the University of Wisconsin have signed a strategic partnership to release Condor under an OSI License (making it possible to be included in open source distributions), and co-develop together to bring innovation from the research community to the enterprise. Condor has been in active development and used by a wide community since 1988. Now, based on their joint vision of advancing open source software, Red Hat and the University of Wisconsin are collaborating to add enhanced enterprise stability and functionality to Condor, add high throughput computing capabilities to Linux, and ultimately advance and strengthen the Condor project and community.
15:00 – 15:30
Grid and Cloud computing for e-Science, Microsoft Research
Dr. Fabrizio Gagliardi is Europe, Middle East, Africa and Latin America director for scientific and technical computing in the Advanced Strategies and Policy Division, at Microsoft Corporation.
The electronic infrastructure is an integral part of the new science environment and high performance computing is an essential instrument for doing research. The new science environment which is referred to as electronic Science (e-Science) uses simulation techniques based on software modeling which run on distributed computing infrastructures. In addition, it makes use of huge amounts of distributed and shared data captured by instruments or sensors and/or stored in databases, analyzed to provide new results for science. This distributed HPC and data environment allows sharing the acquired knowledge, accessing remote resources and enabling world-wide scientific collaboration.
In the research and academic arena, the EU Datagrid project was one of the first to adopt the Grid computing paradigm, which was followed by the EGEE (Enabling Grids for E-science) project integrating more than 55.000 CPUs in Europe and beyond, and 25 Peta Bytes (millions of GigaBytes) of storage. EGEE is currently serving multiple application communities including HEP, Bioinformatics, Astrophysics, Computational Chemistry, Earth Sciences, and Fusion. Some business/industrial applications are also adopting the distributed HPC computing such as the automotive, finance, multimedia, and there a few examples of e-Government ones such as in the civil protection area. In this way, Grid computing has delivered an affordable high performance computing infrastructure to scientists all over the world to solve intense computing problems within constrained research budget. Furthermore it has promoted the use of advanced research networks and catalyzed effective international collaborations.
On the other hand, industry is also moving ahead in this direction, either as an internal user of the technology or as an infrastructure and services provider. In the first case, intra grids, i.e. grids of computers composed of the facilities of a single company’s sites, is the main implementation of Grids in industry. As an example Novartis, a huge pharmaceutical consortium spread around the world is using an intra Grid for their computing needs. In the latter case, we have seen recently enormous interest from companies to become “cloud computing” service providers. Amazon Elastic Computing Cloud (EC2) offers CPU virtual instances for a low cost per hour ($0,10) and it is quite easy to use. In addition, it offers data transfer services in and out of their systems ($0,10 per GB for data in and $0,18 for data transfer out. For datasets which are hosted on the Amazon system, one has to use the Amazon Simple Storage Service (S3), with a similar cost (around $0,15 per GB per month). Other major stakeholders in the market such as Google, IBM and SUN are moving in the same direction towards offering similar services of on-demand computing, storage, hosting and an application development environment. Google recently announced a free service, but with limited sign-ups, as a preview release of Google App Engine, which enables developers to build and run their web applications on the Google infrastructure. As announced “every Google App Engine application will have enough CPU, bandwidth, and storage to serve around 5 million monthly page views”. IBM in Haifa has launched an EC funded project called RESERVOIR - Resources and Services Virtualization without Barriers, will explore the deployment and management of IT services across different administrative domains, IT platforms and geographies. This cloud computing project aims to develop technologies to support a service-based online economy, where resources and services are transparently provisioned and managed. SUN, who was one of first to offer utility computing, is offering its Sun Grid Compute Utility, which was recently extended to other countries beyond the US. It offers its Sun servers (Sunfire dual processor Opteron-based) with Solaris and Sun Grid engine for 1$ per CPU-hr. In parallel, many multi-cores and CPU accelerators promise potential breakthroughs, without needing to rely on computer clusters and the grid.
All the above constitute some of the emerging trends and dynamics of on-demand cloud, utility and pay-per-use computing which might soon change the IT environment. Microsoft is actively investigating this field, and the Technical Computing activity in Microsoft Research, is supporting e-Science initiatives in collaboration with leading scientists around the world to enable easier and better scientific discovery. We need to advance in making computing easy to use for the scientists to concentrate their energy in real science and not the computing tools.
15:30 – 16:00
The importance of research collaborations between Industry and Universities
Dr. Avi Mendelson, Intel Israel
Examples from Intel experience and in particular about Intel-Microsoft experience.
Avi Mendelson is a principal engineer in Intel’s Mobile Platform Group in Haifa, Israel, and adjunct professor in the CS and EE departments, Technion – Israel Institute of Technology. He received his B.Sc. and M.S.c degrees from the Technion, Israel Institute of Technology and his Ph.D from the University of Massachusetts at Amherst. Avi has been with Intel for 7 years. He started as senior researcher in Intel Labs, later he moved to the Microprocessor group where he served as the CMP architect of Intel Core Duo. Avi's work and research interests are in computer architecture, low power design, parallel systems, OS related issues and virtualization.
16:00 – 16:30 Coffee Break
16:30 – 17:00 (WebEx)
Using MRG to look at and understand the trend in HPC, distributed and Grid computing
Carl Trieloff, Senior Consulting Software Engineer, Red Hat
Carl Trieloff, Senior Consulting Software Engineer at Red Hat, has over 16 years of enterprise engineering experience in Trading Exchanges, Middleware, Messaging technologies, distributed systems (SOA), including work on many mission critical control systems.
This presentation will be combination of industry trends and lessons learned in the implementation of the Red Hat MRG platform, which comprises technologies from Messaging (AMQP), Linux Realtime and Grid computing using the Condor project from the University of Wisconsin. Demands are ever increasing to provide compute for more critical business function, with higher determinism and lower cost, out of which some innovative techniques & solutions are being born in the industry.
17:00 – 17:30
Hunting for disease genes using thousands of computers
Mark Silberstein, PhD candidate, Computer Science Department, Technion
Genetic linkage analysis is a statistical tool used by geneticists for mapping disease-susceptibility genes in the study of genetic diseases. However such analysis is often beyond the capabilities of a single computer. We present a distributed system for faster analysis of genetic data, called Superlink-online. The system achieves high performance through parallel execution of linkage analysis tasks over thousands of computational resources residing in multiple opportunistic computing environments, aka Grids. It utilizes resources in all the available grids, unifying thousands CPUs over campus grids in the Technion and the University of Wisconsin in Madison, EGEE grids in Europe, Open Science Grid in the USA and Community Computing Grid Superlink@Technion. Notably, the system is available online, which allows geneticists to perform computationally intensive analyses with no need for either installation of software, or maintenance of a complicated distributed environment. It is being extensively used by medical centers worldwide, running over 15,000 interactive genetic analysis tasks since 2006. While the grids potentially provide enormous amount of computing power, we also explore an alternative approach of using Graphics Processing Units (GPUs) to accelerate the genetic linkage computations. We achieve up to two orders of magnitude speedups on average, and up to three order of magnitude speedups on some particularly complex problem instances versus the optimized application performance on a single CPU. The use of GPUs is particularly appealing in the context of Community Grids, considering the number of high performance GPUs available worldwide. In this talk we will describe various aspects of the system architecture which drives Superlink-online, as well as our recent GPU-related results.
The talk is self-contained and does not require any prior knowledge in human genetics or distributed computing.
17:30 – 18:00
GWiQ-P: An Efficient Decentralized Grid-Wide Quota Enforcement Protocol
Kfir Karmon, Liran Liss and Assaf Schuster (Technion)
Mega-grids span several continents and may consist of millions of nodes and billions of tasks executing at any point in time. This setup calls for a scalable and highly available resource utilization control that adapts itself to dynamic changes in the grid environment as they occur. In this paper, we address the problem of enforcing upper bounds on the consumption of grid resources. We propose a grid-wide quota enforcement system, called GWiQ-P. GWiQ-P is light-weight and in practice infinitely scalable, satisfying concurrently any number of resource demands, all within the limits of a global quota assigned to each user. GWiQ-P adapts to dynamic changes in the grid as they occur, improving future performance by means of improved locality. This improved performance does not impair the system's ability to respond to current requests, tolerate failures, or maintain the allotted quota levels.
To reserve your place, please send your contact details to: info@grid.org.il
