 |
|
 |
       |
 |
|
- Cluster Interconnects: Single Points of Performance (pdf)- The most common approach for comparing different interconnect solutions is the “single-points” approach. It is difficult, and sometimes misleading, to predict real-time application performance with just single-points of data. In order to determine the system’s performance and the interconnect of choice, one should take into consideration a set of metrics including the single-point of performance (bandwidth, latency etc.), architecture characteristics (CPU utilization, overlap capabilities of computations and communications scalability etc.), applications results, field proven experience and hardware reliability.
2007 OpenFabrics Sonoma Workshop - The Lodge at Sonoma
April 29 - May 2, 2007
The OpenFabrics Alliance is hosting the 3rd Annual International Sonoma Workshop from April 29 - May 2, 2007. The event is being held at The Lodge at Sonoma. Anyone interested in the work of the Alliance is welcome to attend. Membership in the Alliance is not necessary.
This year's workshop will be the most content-rich event so far. We are broadening the topics to include updates from industry on expected evolution of Server, Blade System, Storage Networking, Telecommunications, Bus Architecture, InfiniBand and Ethernet technologies in 2007 and 2008. We are particularly interested in participation from early adopting System Suppliers, OEMs, ISVs, Integrators and Customers from all parts of the world to give us feedback on their experience deploying either InfiniBand or Ethernet with the OpenFabrics Enterprise Distribution software.
At the Workshop the Alliance will host its Annual General meeting which provides the opportunity for attendees to get up-to-date with our work including:
- the development, quality assurance and feature set of the next Enterprise Distribution of the OpenFabrics software stack (OFED 1.2)
- the status of the Alliance’s Interoperability Working Group with its Logo program in collaboration with the University of New Hampshire’s Laboratory
- input from the High Speed Interconnect Roundtable and the Alliance’s user community
- reports from the Alliance’s Development, Legal and Marketing Working Groups
- collaboration with the InfiniBand Trade Association and the Ethernet Alliance to define capabilities (e.g. IB routing, problem diagnostics, etc.) that can aid in RDMA technology adoption and deployment as well as standards advances that will enable data center adoption, higher speeds and longer-distance transports
Joining forces to power the data center and high-performance computing clusters
Mellanox is dedicated to building a world-class marketing alliance program by working closely with our chosen partners and customers to ensure the best total solution is available to end-customers. We are committed to cultivating long-term, marketing relationships with best-in-class original equipment manufacturers (OEMs) as well as strategic technology suppliers and independent software vendors (ISVs).
Member Profile for February 2007
Hewlett Packard is a technology solutions provider to consumers, businesses and institutions globally. The company’s offerings span IT infrastructure, global services, business and home computing, and imaging and printing. The HP BladeSystem c-Class portfolio is a key part of HP’s Adaptive Infrastructure offering, which helps customers move toward automated, “lights-out” computing environments that lower the cost of IT operations and deliver a higher quality of service.
- Hewlett Packard Website
- InfiniBand for HP BladeSystem c-Class
- Press Release: Mellanox InfiniBand Supercharges HP Blade Server
Mellanox Cluster Center offers an environment for developing, testing, benchmarking and optimizing products based on InfiniBand technology. The center, located in Santa Clara, California, provides on-site technical support and can be accessed through secure sessions onsite or remotely.
The Cluster Center provides a unique ability to access the latest InfiniBand technology, even before it is generally available to the public. It provides a development testing and tuning environment for applications to take advantages of Mellanox’s superior characteristics.
Visit Mellanox at these upcoming 2007 events!
- Storage Networking World - Spring
Manchester Grand Hyatt
San Diego, CA
April 16-19, 2007
Booth #PP15 - IDF - Beijing
Beijing International Convention Center
Beijing, PRC
April 17-18, 2007 - Linux/OpenSource on Wall Street
Roosevelt Hotel
New York, NY
April 23, 2007
Booth #212
- 22 Feb 07
FalconStor Focuses on China - 1 Feb 07
Cluster Interconnects: Single Points of Performance
- 28 Feb 07
NCSA Adds Two Clusters, 110 Teraflops of Computing Power - 26 Feb 07
Cisco Pushes InfiniBand for High-End DW - 26 Feb 07
Department of Defense to Use NX5010 Platform from Network Equipment Technologies - 21 Feb 07
Zarlink's Innovative ZLynx Optical Cable Products Simplify Data Center and Computer Cluster Interconnects - 20 Feb 07
Datallegro Unveils All-Commodity Data Appliance - 16 Feb 07
Ripples in Spacetime - Solved! - 14 Feb 07
FalconStor Tunes Up InfiniBand - 13 Feb 07
Supercomputer launched in Tomsk Polyctechnic - 13 Feb 07
Dynamic Network Factory's New High-Performance Enterprise F12-HA RAID Subsystem Ensures Fault Tolerance With Active-Active RAID Controllers - 8 Feb 07
High-performance Linux cluster in operation in Chemnitz
 |
|