Topics » Interconnects

Features

Our Series Continues: Considering Cluster Lifecycle Management

Oct 22, 2014 |

In the first cluster management column, I discussed the importance of performing a needs assessment to ensure that an organization implementing its first HPC cluster designs a system that fulfills the requirements of its users and ultimately provides a positive ROI for the enterprise. It’s important to keep in mind there are no one-size-fits-all solutions Read more…

Austrian HPC Consortium Meets Demanding Inter-node Communication Challenge with Intel True Scale Fabric

Aug 6, 2014 |

The Vienna Scientific Cluster (VSC) refers to a collaboration of high-performance computing resources designed to support a consortium of multiple institutions in Austria consisting of the University of Vienna (UNVIE), the Vienna University of Technology (TUVIE), the University of Natural Resources and Applied Life Sciences Vienna (BOKU), the Graz University of Technology (TU Graz), and Read more…

Low Effort Porting Gives Great Scaling of Seismic Code on Numascale Shared Memory Cluster

Aug 4, 2014 |

Numascale and Statoil have initiated a collaboration to optimize OPM for Numascale systems. The original OPM code showed poor scalability even within a single node. With very limited effort, the collaboration was able to obtain almost linear scaling across an entire Numascale system. The configuration enables simulation of entire oilfields with fine grid resolution. The Read more…

This Week in HPC News

Jun 12, 2014 |

If middleware was the theme from last week, then this week is all about hardware, from networking/interconnects (Arista) to systems (HP) to storage (Panasas). Let’s look at little closer at this week’s top stories, starting with HP. HP’s rebooting of the Apollo brand this week was well received. This new family of high performance computing systems Read more…

What Power8 and OpenPOWER Might Mean for HPC

Apr 23, 2014 |

IBM is making a big play in hybrid computing, seeking to marry its POWER8 processors with various kinds of accelerators and high-speed networking and opening up its chip and system software through the OpenPOWER Foundation. At the Open Innovation Summit in San Francisco today, IBM and its foundation partners talked about how their collaboration would deliver Read more…

Short Takes

DOE Funds Exascale Interconnect R&D

Nov 15, 2013 |

The Department of Energy (DOE) continues to prime the R&D pump for the next-generation of supercomputers, exascale machines 50 times more powerful than today’s leading number-crunchers. In order to bring this goal to fruition, the Department of Energy’s Office of Science together with the National Nuclear Security Administration (NNSA) have awarded $25.4 million in research Read more…

Tiny Laser Transmits Data at Record Speeds

Nov 8, 2013 |

In the age of big data and cloud computing, fast and accurate data transfer is more important than ever. University of Illinois scientists address the network bottleneck. Scientists at the University of Illinois have figured out a way to transmit error-free data over fiber optic networks at 40 gigabits per second, a US record according Read more…

Chip-Level Optical Interconnect Market Primed for Growth

Oct 16, 2013 |

Analyst firm Communications Industry Researchers (CIR) has released a new report covering the latest developments in optical interconnection at the chip level and the progress in this area that is being made by research teams throughout the world. With the current era of multicore processors and 3D chips, computing speeds are limited by how fast Read more…

Infinera, DANTE Achieve Guinness Record

Aug 30, 2013 |

Earlier this month, Infinera and DANTE (Delivery of Advanced Network Technology to Europe) set a record for fastest provisioning of long haul optical transmission capacity. In less than 20 minutes, the network specialists enabled enough capacity to support the simultaneous streaming of 1.6 million HD films in both directions.

Considerations in HPC Cluster Network Selection

Jul 17, 2013 |

There’s a lot going on in the networks of HPC clusters, and selecting the right network fabric, equipment, and topology is important to ensuring good performance for given applications. A “one size fits all” approach rarely works, and architects will do well to tailor the network to the needs of the application.

Off the Wire

IEEE Photonics Society’s Optical Interconnects Conference to Explore Next-Gen Supercomputers

Mar 20, 2015 |

SAN DIEGO, Calif., March 20 — The IEEE Photonics Society will hold the fourth annual Optical Interconnects Conference on 20 – 22 April 2015 at the Wyndham San Diego Bayside Hotel in San Diego, CA. Established more than 25 years ago as the Workshop on Interconnections within High Speed Digital Systems, the conference will bring together the optical Read more…

IBTA Announces New Release of InfiniBand Architecture Specification

Mar 10, 2015 |

SUNNYVALE, Calif., March 10 — The InfiniBand Trade Association (IBTA), a global organization dedicated to maintaining and furthering the InfiniBand Architecture, today announced the public availability of Release 1.3 of Volume 1 of the InfiniBand Architecture Specification. The new release defines new capabilities that will enable computer systems to keep up with the requirements for increased Read more…

Avago Technologies Limited to Acquire Emulex

Feb 25, 2015 |

Combination of Emulex’s connectivity solutions with Avago’s Server Storage Connectivity and Fiber Optic products creates one of the industry’s broadest portfolios for Enterprise Storage Emulex to operate as a business unit within Avago’s enterprise storage segment Expected to be immediately accretive to Avago’s EPS, on a non-GAAP basis SINGAPORE and COSTA MESA, Calif., Feb. 25 Read more…

IDT Develops Supercomputing Platform with NVIDIA and Orange Silicon Valley

Feb 25, 2015 |

SAN JOSE, Calif., Feb. 25 — Integrated Device Technology, Inc. (IDT) announced today that it has developed with NVIDIA and Orange Silicon Valley a supercomputing platform that can analyze 4G to 5G base station bandwidth data in real time, enabling network operators to monetize the voluminous data flowing through their communications systems. By connecting clusters of Read more…

Mellanox ConnectX-4 100Gb/s Interconnect Adapter Delivers Record Performance Results

Feb 25, 2015 |

SUNNYVALE, Calif. & YOKNEAM, Israel, Feb. 25 — Mellanox Technologies, Ltd. a leading supplier of high-performance, end-to-end interconnect solutions, announced today that its industry-leading ConnectX-4 EDR 100Gb/s InfiniBand adapter attained world-record performance results. Achieving InfiniBand throughput of 100Gb/s, bi-directional throughput of 195Gb/s, applications latency level of 610 nanoseconds and message rate of 149.5 million messages per Read more…