Some possible topics:
HPC in the Cloud
Latest Developments in GPU, CUDA for HPC
Whats going on in the world of FPGA’sa for HPC
Send us your ideas
DATE : Tuesday September 9th
TIME: 2:30 PM to 4:30 PM
LOCATION: WWT 4201 Forbes Blvd, Lanham Maryland 20706
PARKING: FREE
EVENT COST: FREE
Article by Joab Jackson the Technical Editor of Government Computer News on the event :
http://www.gcn.com/online/vol1_no1/47104-1.html
ABSTRACT:
Today faster more dense HPC systems require a much faster data I/O storage systems. These systems must be easy to maintain and simple to upgrade. To meet the requirement of the HPC community these storage systems must be highly cost effective. The presentation will focus on these problems and the practical solutions for meeting these challenges.
StorageWorks Enterprise NAS division’s Enterprise File Services, provides superior performance, linear scalability and mission critical availability over all other clustered files system solutions making it perfect for the HPC market. With demonstrable performance of over 3 GB’s per second in a 16 node cluster, the NAS solution has differentiated itself as a market leader. Incremental scalability – grow as you need to with capacity and performance independent of each other. Add storage on the fly as well as performance nodes without interruption. Drive high storage utilization while providing superior operational efficiency through unified backups.
Lincoln’s Bio:
Lincoln is a Federal Solutions Architect for Hewlett Packard’s Enterprise NAS Division. Lincoln has over 8 years of systems engineering and implementation experience and currently provides direction and support for a number of government installations. Lincoln and his family moved to the Columbia, MD in June of 2004.
The old Baltimore Washington Beowulf Users Group is now the New
HPC TECH FORUM. Under the new structure we will continue to provide some of the top speakers in the HPC community. We will expand the scope of the discussions to include pressing issues that were not problems a few years ago.
Infrastructure – Power and cooling issues
Management – HPC as a managed service and remote offsite system administration and support
Financial – With Moore’s Law working on all eight cylinders and replacement cycles contracting, is it best to own or lease HPC Systems?
Cost of Operation – What are the challenges that control the Total Cost of Ownership (TCO)?
Green HPC Systems – Is it practical to have Green HPC Systems?
Form Factor – Is the “pizza box” server in a “Death Spiral” soon to be replaced with high density blade servers which use less power and run cooler?
Interconnect – Will Infiniband be put on a “life support system” as it is being replaced with low latency 10 Gb
The Meetings will be held at World Wide Technology 4201 Forbes Blvd Lanham, MD 20706
Sep 9, 2008, Oct 14, 2008, Nov 11, 2008
Speakers and topic TBD
__________________________________________________________
We are pleased to announce that Kevin Kelly Terascala will be our special guest speaker. January 8, 2008 at Georgetown University from 2:30 to 5:00 PM. Please join us for a very informative talk on low cost very high speed storage for High Performance Computing.
Meeting Location: 3300 Whitehaven Street, Washington DC 20007. (this is NOT at the main Georgetown U campus. This is an off campus building one block from Wisconsin avenue).
If you have suggestions regarding speakers for the next meeting or meetings in the future please send them to beowulfUG@aol.com
NetEffect’s Terry Hulett to Present at the next HPC Cluster Users Group & BWBUG meeting May 8th at Georgetown University
April 19, 20072:30-5:00 PM
Meeting Location: 3300 Whitehaven Street, Washington DC 20007. (this is NOT at the main Georgetown U campus. This is an off campus building one block from Wisconsin avenue).
Dail-In information: 1-888-296-6500 guest code 435 588
NetEffect is addressing the needs of this market with a family of 10 Gbps iWARP Ethernet Channel Adapters that meet the low latency and high data throughput requirements of high performance cluster computing.
NetEffect’s full iWARP Ethernet implementation supports all the node-to-node interconnect performance enhancement and overhead reduction requirements including RDMA, TCP Offload, and OS Bypass
Terry Hulett- Vice President of Architecture and Silicon EngineeringTerry leads NetEffect’s innovation efforts, leveraging more than 23 years of semiconductor design experience. Prior to joining Banderacom and NetEffect, he held key positions at Advanced Micro Devices (AMD) and Motorola Semiconductor, leading several microprocessor development teams. At AMD, Terry led the development of over 15 chips and built AMD’s first high-volume card-based products. He joined Banderacom in 1999 where he successfully led the Company’s InfiniBand chip development program delivering the first product in 9 months. Terry also previously served as vice president of research and development for Austin-based Si Solutions, Inc. Terry received his BSEE at Oklahoma State University.
A Technical Briefing for Researchers and Scientists
April 18, 2007
1-5 PM
NIH, Bethesda, MD
Building 5, Room 127 This educational event will focus on the role that high-performance computing (HPC) systems play in genomic and proteomic discoveries that impact the treatment of disease. Attendees will learn strategies for implementing affordable, balanced HPC platforms that support the unique work flow requirements of life sciences professionals. Speakers will include:– Stephen Wheat Ph.D., Executive Director, HPC Platform Organization, Intel – Jack Helmly, President, GTSI Financial Services – Doug Brown Ph.D., Genetic Scientist, HP Please arrive at Room 127 at NIH Building 5 by 1:00 pm.
Today!
April 10th 2:30 to 5:00 PM Bill Silva of Microsoft will speak at Georgetown University.
Meeting Location: 3300 Whitehaven Street, Washington DC 20007. (this is NOT at the main Georgetown U campus. This is an off campus building one block from Wisconsin avenue).
The dial in number for the meeting today is 1-888-296-6500 your guest code is: 435 588
ABSTRACT:
Microsoft Windows Compute Cluster Server 2003
For customers solving complex computational problems, Microsoft Windows Compute Cluster Server 2003 accelerates time-to-insight by providing a High-Performance Computing (HPC) platform that is simple to deploy, operate, and integrate with existing infrastructure and tools. Prescriptive setup procedures simplify network configuration, remote loading of nodes using Remote Installation Services, node configuration, and security setup. The integrated Job Scheduler can be accessed via command-line interface, or through several API’s is provided for submitting and managing cluster workloads. The job console allows you to manage jobs in much the same fashion that you queue and manage print jobs today. Active Directory integration provides end-to-end user and security management while the Microsoft Management Console supports extensible snap-ins and integration with Microsoft Operations Manager.
SPEAKER BIOGRAPHY:
Bill is the manager of Microsoft’s Federal High Performance Computing business, based in the Washington, DC area. Bill’s High Performance Computing background includes executive level management positions ranging from venture capital start-ups such as the SIMD massively parallel pioneer Wavetracer, to Floating Point Systems and Bolt Beranek and Newman (BBN), to work on the development team at Digital Equipment Corporation on the VAX 9000 Vector Processor, as well as managing the technology transfer program at the DARPA sponsored Center for High Performance Computing, Worcester Polytechnic Institute
Our next class in January 2007. Registration is now open.
Georgetown University’s Advanced Research Computing, in cooperation with the Baltimore Washington Beowulf User Group, is pleased to announce “Introduction to Beowulf Administering,” the first course in the High Performance and High Throughput Computing Series. This introductory course is geared toward Linux/Unix system administrators who want to learn how to design, build, and maintain Beowulf Clusters. For more course details and registration information, see the URL below or contact Jess Cannata directly.
http://www.gridswatch.com/index.php?option=com_content&task=view&id=25&Itemid=16
December 5, The Ronald Reagan Building, Washington, DC
You are Invited to this Complimentary Executive Seminar
Traditionally, high-performance computing (HPC) has been the domain of the computer scientist, the research lab and a limited user group demanding unprecedented computational capability.
However, today the need to collect, process and analyze terabytes of data reach beyond the confines of real-time intelligence, modeling weather patterns, deciphering genomes and running flight simulators. An
increasing number of agencies have a need for the cutting-edge capabilities of scalable, ultra-fast computing.
Attend this Seminar and Learn
- How government organizations are evaluating and deploying scalable, ultra-fast computing to meet growing requirements for analysis and modeling that reach well beyond recognized high-performance computing requirements
- How the public sector is leveraging the price/performance potential of scalable, ultra-fast computing for data mining, customer relationship management (CRM), predictive and statistical analysis, business intelligence, large databases and more about the technology creating opportunities across government, including enhanced Commercial-of-the-Shelf (COTS) tools, which can be applied for faster, more energy-efficient computing
Take away best practices and lessons learned on how you can reconsider HPC strategies for your computing environment, with proven cases from a variety of government sectors.
Space is limited!
Reserve your seat today and enter the new year with a fresh perspective on ultra-fast, scalable computing to meet your critical agency mission and business requirements.
Your Registration Includes:
- Networking breakfast
- Admission to all sessions
- Executive Seminar materials
For more information, visit the website or view the schedule.
Space is limited.
Register Today!