In this video from the 2013 HPC User Forum, Scott Schultz from Mellanox presents an overview of Mellanox and HPC.
Download the slides (PDF) or check out the HPC User Forum Video Gallery.
Video: An Overview of Mellanox in HPC
Mellanox to Acquire Kotura
Today Mellanox announced plans to acquire photonics leader Kotura, Inc. for approximately $82 million. The acquisition is expected to expand Mellanox’s ability to deliver cost-effective, high-speed networks with next generation optical connectivity, allowing data center customers to meet the growing demands of high-performance, Web 2.0, cloud, data center, database, financial services and storage applications. Mellanox believes that the Kotura acquisition will enhance its ability to provide leading technologies for high speed, scalable and efficient end-to-end interconnect solutions.
Operating networks at 100 Gigabit per second rates and higher requires careful integration between all parts of the network. We believe that silicon photonics is an important component in the development of 100 Gigabit InfiniBand and Ethernet solutions, and that owning and controlling the technology will allow us to develop the best, most reliable solution for our customers,” said Eyal Waldman, president, CEO and chairman of Mellanox Technologies. “We expect that the proposed acquisition of Kotura’s technology and the additional development team will better position us to produce 100Gb/s and faster interconnect solutions with higher-density optical connectivity at a lower cost. We welcome the great talent from Kotura and look forward to their contribution to Mellanox’s continued growth.”
Read the Full Story.
Video: Massive I/O Requirements for the SKA Telescope
In this video from the 2013 Open Fabrics Developer Workshop, Bill Boas from Cray presents: Massive I/O Requirements for the SKA Telescope.
Processing the vast quantities of data produced by the SKA will require very high performance central supercomputers capable of 100 petaflops per second processing power. This is about 50 times more powerful than the most powerful supercomputer in 2010 and equivalent to the processing power of about one hundred million PCs.
Download the slides (PDF) or check out more OFA videos in our Open Fabrics Worshop Video Gallery.
Upgrading the Grid for the LHC
Over at International Science Grid This Week, Katie Kahlie writes that the current maintenance hiatus for the Large Hadron Collider provides a window to upgrade one of the most ambitious data grids on the planet.
The challenges for the Grid were three-fold. The main one was to understand how best to manage the LHC data and use the Grid’s heterogeneous environment in a way that physicists could concern themselves with analysis without needing to know where their data were. A distributed system is more complex and demanding to master than the usual batch-processing farms, so the physicists required continuous education on how to use the system. The Grid needs to be fully operational at all times (24/7, 365 days/year) and should “never sleep”, meaning that important upgrades of the Grid middleware in all data centres must be done on a regular basis. For the latter, the success can be attributed in part to the excellent quality of the middleware itself (supplied by various common projects, such as WLCG/EGEE/EMI in Europe and OSG in the US, see box) and to the administrators of the computing centres (coordinated by EGI in Europe and OSG in North America), who keep the computing fabric running continuously.
Read the Full Story.
RDMA and Storage at a Distance
Over at Forbes, Tom Coughlin writes that RDMA extends the capability of fast direct access to memory between computers in a cluster to greater distances, within a Metropolitan Area Network (MAN ) or even in a Wide Area Network (WAN) that can span continents.
RDMA over a WAN allows some very useful capabilities that can increase the overall power of a clustered computer system. It can provide remote collaboration with a remote file system allowing access as though it were local, enabling apparent real-time collaboration. RDMA also allows very efficient file transfer over a WAN. This direct data placement is accomplished with little impact on the processors on either end of the file transport. These features are very useful for working with large data files such as those common in many HPC applications. Storage at a Distance will not directly impact conventional client computing since these devices typically don’t have access to dedicated high-speed Internet connections. However with the growth of on-line (cloud) services the use of RDMA could accelerate many background processes within a given data center and between data centers. This could improve overall cloud performance and provide services such as fast backups and replications of data to provide data recovery. Thus Storage at a Distance could have a great impact on the overall performance and capabilities available over the Cloud.
Read the Full Story or see Coughlin’s recent Open Fabrics presentation over at inside-Cloud.
Video: Infinetics – Innovative Topologies
In this video from the 2013 Open Fabrics Developer Workshop, Harry Quackenboss and Ratko Tomic from Infinetics present: Innovative Topologies.
We have invented a unique approach to building a fabric across a large number of Ethernet switches, and built a comprehensive technology platform based on this Flexible Radix Switching (FRS) technique. This innovation enables transparent integration with existing data center solutions and big improvements to networks supporting cloud, virtualization, and big data applications. These data center network solutions are superior in terms of cost, performance, robustness and ease of use.
Download the slides (PDF). You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
Intel MPI Library for Intel Xeon Phi-based Clusters & MVAPICH2 for Intel MIC
In this video from the 2013 Open Fabrics Developer Workshop, Bill Magro from Intel presents: Intel MPI Library: Implementation for Intel Xeon Phi Based Clusters. Download the slides (PDF).
In this follow-up session, DK Panda from Ohio State University presents: MVAPICH2 for Intel MIC.
You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
Video: Network Direct v2 and WinOFED
In this video from the 2013 Open Fabrics Developer Workshop, Fab Tillier from Microsoft presents: Network Direct v2 and WinOFED.
You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
Evolving OFS – Teaching Sharks to Swim, from the Top Down
In this video from the 2013 Open Fabrics Developer Workshop, Paul Grun from OFA presents: Evolving OFS – Teaching Sharks to Swim, from the Top Down.
This talk is a follow-on to Grun’s OFA workshop opening presentation. You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
Interview: Open Fabrics Alliance Looks Ahead
In this video, Jim Ryan and Paul Grun from OFA discuss where Open Fabrics is headed in the future.
Open Fabrics began as a way of providing a memory-to-memory messaging service from application to application. So, fundamentally, that’s what sets Open Fabrics apart from any other network. Typically networks are about delivering packets from one platform to another. But Open Fabrics takes a perspective that its about the way the applications communicate, it’s truly about memory to memory.
You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
Video: InfiniBand Fabric Administration for Linux
In this video from the 2013 Open Fabrics Developer Workshop, Rupert Dance from Software Forge presents: InfiniBand Fabric Administration for Linux.
You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
High Performance RDMA-based Design for Big Data and Web 2.0 memcached
In this video from the 2013 Open Fabrics Developer Workshop, D.K. Panda from Ohio State University presents: High Performance RDMA-based Design for Big Data and Web 2.0 memcached.
Check out more presentation videos at our Open Fabrics Workshop Video Gallery. Most of the slides from the Workshop have been posted as well.
Video: Update on Lustre, OpenSFS, and FastForward InfiniBand
In this video from the 2013 Open Fabrics Developer Workshop, Doug Oucharek from Intel presents an Update on Lustre, OpenSFS, and FastForward InfiniBand.
Download the slides (PDF). You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
In related news, slides from LUG 2013 are now posted.
Video: Ethernet over InfiniBand (EoIB)
In this video from the 2013 Open Fabrics Developer Workshop, Ali Ayoub from Mellanox presents: Ethernet over InfiniBand (EoIB).
Download the slides (PDF). You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
Video: RSOCKETS – RDMA for Dummies
In this video from the 2013 Open Fabrics Developer Workshop, Sean Hefty from Intel presents: RSOCKETS – RDMA for Dummies.
You can check out more OFA videos at our Open Fabrics Workshop Video Gallery.
