Monday, November 16, 2015

SC15: Intel Shows HPC Scalable System Framework and Omni-Path Fabric

At this week's SC15 in Austin, Intel is showcasing several advancements to its Scalable System Framework (Intel SSF) that promise to bring high performance computing (HPC) capabilities and benefits to more industries and new workloads.

One of the foundational elements of the Intel SSF is its Omni-Path Architecture, a new HPC-optimized fabric technology that makes the performance of HPC clusters more accessible to a broader variety of users.

Some highlights:

  • Intel will provide Intel SSF reference architectures, designs and validation tools. These technical system specifications will include hardware and software bill of materials for Intel SSF validated systems.
  • Companies planning to launch systems based on SSF include Colfax, Cray, Dell, Fujitsu Systems Europe, HPE, Inspur, Lenovo, Penguin Computing, SGI, Sugon and Supermicro
  • Intel OPA’s 48-port switch enables up to 26 percent more servers than InfiniBand EDR within the same budget and up to 60 percent lower power consumption for a more efficient switch and system infrastructure.
  • Intel OPA is currently being used at several large sites, including the Texas Advanced Computing Center and the Pittsburgh Supercomputer Center. Colfax, Cray, Dell, Fujitsu Systems Europe, Hitachi*, Huawei*, HPE, Inspur, Lenovo, NEC, SGI, Sugon, Supermicro and other system vendors starting to announce Intel OPA-based switches and server platforms, with volume shipments ramping in the first quarter of next year.
  • Intel announced that preproduction Intel Xeon Phi processors (code-named Knights Landing) are running in several supercomputing-class systems. Cray has a system currently running multiple customer applications in preparation for the supercomputer deployments at Los Alamos (Trinity system) and NERSC (Cori system). Systems are also installed at CEA (the French Alternative Energies and Atomic Energy Commission) by Atos and Sandia National Laboratories by Penguin Computing. Intel expects more than 50 system providers to have Intel Xeon Phi product family-based systems in the market at launch.
  • Intel is among more than 30 founding members of the OpenHPC Collaborative Project, a new community-led organization focused on developing a comprehensive and cohesive open source HPC system software stack to drive broader adoption. Intel-supported versions of the open source HPC system software stack are expected to be available next year.
  • Intel expands HPC ecosystem support and investments with the creation of five new Intel Parallel Computing Centers focused on Lustre software development; through a new long-term strategic partnership with the Alan Turing Institute to further the role of high performance computing and data sciences in the U.K.; and with an expanded collaboration with the Barcelona Supercomputing Center, focused on advancing the exascale capabilities of the Intel SSF.
  • Intel announced a multiyear, multimillion-dollar program to increase diversity within the HPC ecosystem consisting of a new scholarship program designed to attract women and underrepresented minorities to pursue graduate degrees in computational and data science programs, as well as a new internship program to increase the diversity of Intel’s HPC workforce.

“We’re entering a new era in which supercomputing is being transformed from a tool for a specific problem to a general tool for many,” said Charlie Wuischpard, vice president and general manager of HPC Platform Group at Intel. “System-level innovations in processing, memory, software and fabric technologies are enabling system capabilities to be designed and optimized for different usages, from traditional HPC to the emerging world of big data analytics and everything in between. We believe the Intel Scalable System Framework is the path forward for designing and delivering the next generation of systems for the ‘HPC everywhere’ era.”

http://supercomputing_intel.tri-digital.com

Intel Unveils Knights Landing Processor and Omni Scale I/O Fabric

Intel announced microarchitecture and memory details of its next-generation Intel Xeon Phi processor (code-named Knights Landing) and an end-to-end interconnect fabric optimized for fast data transfers, both aimed at powering High-Performance Computing (HPC) systems.

Knights Landing, which is expected to debut in the second half of 2015, will include up to16GB high-bandwidth, on-package memory at launch – designed in partnership with Micron – to deliver five times better bandwidth compared to DDR4 memory, five times better energy efficiency and three times more density than current GDDR-based memory. Knights Landing will be available in a PCIe-based card option, or mounted directly on the motherboard socket to remove the bandwidth bottlenecks of data transfer over PCIe, common in GPU and accelerator solutions.

Powered by more than 60 HPC-enhanced Silvermont architecture-based cores, Knights Landing is expected to deliver more than 3 TFLOPS of double-precision performance and three times the single-threaded performance compared with the current generation.

Intel's new Omni Scale fabric, which is built upon a combination of enhanced acquired IP from Cray and QLogic along with tis own in-house innovations, addresses the I/O requirements of the HPC sector.  Intel Omni Scale Fabric is an end-to-end I/O solution encompassing adapters, edge switches, director switch systems, cables and management software. Intel intends to replace traditional electrical transceivers in the director switches with silicon photonics, enabling increased port density, simplified cabling and reduced costs. Intel Silicon Photonics-based cabling and transceiver solutions may also be used with Intel Omni Scale-based processors, adapter cards and edge switches.

Intel noted continued market gains in powering the HPC sector.  The company said its current generation of Xeon processors and Xeon Phi coprocessors powers the top-rated system in the world – the 35 PFLOPS "Milky Way 2" in China, and its designs account for 85 percent of all supercomputers on the 43rd edition of the TOP500 list.

http://newsroom.intel.com/community/intel_newsroom/blog/2014/06/23/intel-re-architects-the-fundamental-building-block-for-high-performance-computing

SC15: CenturyLink and Infinera Deliver 1.5 Tbps of Super-Channel DWDM

CenturyLink is delivering 1.5 terabits per second (Tbps) of super-channel transmission capacity to support SCinet at this week's International Conference for High Performance Computing, Networking, Storage and Analysis (SC15)p in Austin.

The 1.5 Tb/s of line-side capacity will connect the Austin Convention Center to CenturyLink's global fiber optic network, utilizing the Infinera DTN-X platform.

The terabit deployment will demonstrate CenturyLink's rapid provisioning of 100 Gigabit Ethernet (GbE) services from the conference to CenturyLink's point of presence in Austin and to six major U.S. cities across the country.

"This deployment demonstrates the power of CenturyLink's network and its ability to scale and quickly provide super-channel transmission capacity for real-time high-performance computing," said Pieter Poll, senior vice president of national and international network planning at CenturyLink. "Providing SCinet with 1.5 terabit of transmission complements the work we are doing to transform our network and keep up with increasing bandwidth demands now and into the future."

http://www.infinera.com
http://www centurylink.com

Nokia Networks joins Open Compute Project

Nokia Networks has joined the Open Compute Project (OCP), demonstrating its commitment to openness and collaboration in offering leading cloud-based solutions.

Nokia Networks said it will work with the OCP Foundation to build an OCP variant of the Nokia AirFrame Data Center Solution.

The vision is to deliver modular and flexible cloud architectures and delivering low latency and telco-grade reliability for all types of workloads.
 
Nokia plans to support the full range of its own Virtual Network Functions (VNFs), as well as third-party VNFs. A variety of configurations, from hyperscale to smaller and distributed data centers needed to support specific services, will begin to be supported. Added enhancements made available through OCP will include improved cooling as well as greater Power Usage Effectiveness (PUE), one of the key metrics for all data center deployments.

http://www.nokianetworks.com


http://networks.nokia.com/Nokia Expands its Telco Cloud Portfolio
Nokia Networks announced a number of improvements and enhancements in its telco cloud portfolio, including the new Nokia OSS Office for Telco Cloud, the Nokia AirFrame Data Center Solution (now also available in a container), the intelligent Nokia Service Chaining and Nokia cloud wise Care Services.

Some highlights:

  • The Nokia OSS Office for Telco Cloudsolution is a service to help operators plan their best path for transforming their network and service operation centers for the telco cloud.  This includes planning from high level strategy to processes, tools and use cases. 
  • The Nokia AirFrame Data Center Solution is now available as a containerized solution with an efficient power and cooling system built-in. It can simply be dropped into place to cater for peak demand or to provide localized telco and IT services. A software-defined storage (SDS) module has also been introduced to provide flexible data storage pools within the data center environment.
  • Nokia Service Chaining provides greater efficiency and management through a virtualized service environment for the delivery of network services. Dynamic service functions, such as firewalls and media optimizers, can easily address peak loads and intelligently steer traffic. These scalable and best-of-breed functions can be implemented within the telco cloud at a fraction of the cost required for conventional networks. Nokia also launched third party appliance certification as well as implementation and integration services for service chaining.
  • Nokia cloud wise Care Services: two new service packages to help operators run and maintain their deployed telco cloud networks. Nokia Resolution Prime Services provides operators with a single point of contact and service management for resolving VNF faults within a complex multivendor telco cloud network. Nokia TotalCare for VMWare, a newly launched care package, supports operators’ VMWare deployments with an extended SLA covered by the company.

http://networks.nokia.com/news-events/press-room/press-releases/nokia-networks-launches-help-operators-execute-strategic-decisions-in-cloud-transformation

NEC's UNIVERGE BLUE Delivers UC

NEC Corporation of America (NEC) introduced its UNIVERGE BLUE Business Cloud Services (UNIVERGE BLUE) for cloud unified communication (UC) services. UNIVERGE BLUE delivers business-class voice, mobility, messaging, presence, web conferencing and contact centers.

The solution consists of NEC Fault Tolerance servers, software-defined networking (SDN), software, and data centers.

Some highlights:

  • Architected for high availability, 99.999 percent
  • Multi-tenant, enterprise-grade security and encryption in the cloud
  • 24/7 support and system monitoring
  • Mission-critical UC applications: voice, mobility, messaging, presence, conferencing and contact center
  • Supports private, hybrid and public cloud UC deployments


“If an enterprise is building a private, hybrid or public cloud service, the underlying architecture has a big impact on what can be done,” said Jay Krauser, director, Business Development Cloud Services, NEC Corporation of America. “Not all cloud solutions are built the same, so it’s also important for businesses to examine the bottom line of their cloud investments and to understand the underlying technology. UNIVERGE BLUE is built from the same reliable software foundation as UNIVERGE 3C™, ensuring mission-critical communications.”

http://www.necam.com
http://www.univergeblue.com

Procera Intros Virtual Probe for Subscriber Intelligence

Procera Networks is launching a Virtual Experience Probe designed to deliver intelligence to real-time analytics systems.

Procera's "eVolution" is an all-virtual data enablement solution that empowers fixed, cable, mobile and WiFi operators to enhance the subscriber experience, by providing contextual intelligence they can act upon immediately. The eVolution data probe runs in VMware or KVM virtualization environments on COTS hardware. Procera said eVolution is differentiated by:

  • Infrastructure and Vendor agnostic data acquisition
  • Real-time virtualized software with highly optimized performance on COTS hardware
  • Deployment flexibility for a broad ecosystem of customers: carriers, enterprises, system integrators, OEMs, channel partners, and third party big data providers
  • Flexible data and KPI formats for use by big data repositories
  • Best cost per subscriber value designed to augment or displace cumbersome and expensive passive probing solutions

As NFV orchestration is deployed, eVolution instances will be able to be created, sized, and deployed in real time, and relocated as needed through the network by the operator using ETSI NFV specifications.  The base eVolution probe delivers access to real-time network forensics leveraging Procera's unique Dynamic LiveView. Software license add-ons deliver real-time data feeds to Big Data systems using IPFix or historical data through ODBC database integration, enabling integration into a wide variety of Real-time and Big Data ecosystems that leverage Hadoop and other standard database technologies.

"Network operators need a turnkey virtual probing solution that provides better real-time data to feed their analytics and customer relationship systems in order to deliver a high quality broadband experience to their subscribers," said Lyn Cantor, President and CEO of Procera Networks. "eVolution from Procera represents a game-changer; the most economical cost per bit data enablement solution for subscriber-based analytics on the market today, and we look forward to the market place unlocking new value from this disruptive solution."

http://www.proceranetworks.com

Alcatel-Lucent Intros Radio Interface for DAS

Alcatel-Lucent introduced a radio frequency module for distributed antenna systems (DAS) for connecting subscribers at large public venues.

The Distributed Antenna System Radio Frequency Module is a wideband low-power LTE interface card which removes the need for bulky radio technology in a public installation. It connects to Alcatel-Lucent's 9926 digital baseband unit, working directly with the analog DAS through RF signals that consume just one-eighth of the power and heat dissipation of an average remote radio head, reducing space requirements and optimizing costs. The 9926 digital baseband unit, connecting into a number of distributed radios, can divert capacity where needed when there are spikes in traffic, redistributing capacity to other radios when no longer required.

Up to 40 radio sectors with the associated baseband processing, can be fitted into a standard 19 inch indoor cabinet. The Alcatel-Lucent DAS RFM is offered in two wideband hardware versions, one for low band and one for high band radios. Each radio covers between 6 and 10 bands. Based on the latest Alcatel-Lucent RF platform it leverages common hardware and software and can quickly be adapted to various market frequency demands using pluggable cards.

Based on a study from Bell Labs, the company calculates its new DAS RFM offers up to 30 percent cost savings in terms of the wireless and DAS equipment required, up to 81 percent cost savings in terms of power and cooling, up to 91 percent cost savings in terms of construction,  and up to 78 percent cost savings in terms of energy.

"Alcatel-Lucent has been working closely with operators on how to improve LTE deployment in dense urban environments by leveraging their DAS infrastructure. This product was developed on the latest leading edge RF Platform that operators use for their Macro today allowing them full benefit and feature parity with their Macro sites. It is designed for use with any vendors' analog DAS," said Glenn Booth, Senior Vice President and General Manger of Alcatel-Lucent's LTE Business Unit said: .

CableLabs Schedules Conference for April 2016

CableLabs will host a new Inform[ed]: conference series beginning on April 13, 2016 in New York City.

  • Phil McKinney, President and CEO, CableLabs
  • Prof. Theodore S. Rappaport, NYU (founding director of NYU WIRELESS)
  • Neville Meijers, VP of Small Cell, Qualcomm
  • Jeremy Bye, VP of Carrier and Wholesale, Cox Business
  • Paul de Sa, Bernstein Research
  • Rob Howald, VP of Network Architecture, Comcast

"As wireless technology continues to advance, it’s imperative that the wireless industry fully explores both the opportunities and challenges of developing innovative wireless technologies that benefit consumers and drive technology forward,” said Phil McKinney, president and chief executive officer of CableLabs. "CableLabs is continually expanding its testing and research in a variety of areas in the wireless space. This conference gives us the opportunity to bring together the industry’s top minds to tackle the most prominent issues in wireless in new and inventive ways.”

http://www.cablelabs.com
http://www.cablenet.org

PMC-Sierra Reiterates Commitment to Skyworks Deal

PMC-Sierra's board of directors unanimously determined that the proposal received on November 9, 2015 from Microsemi Corporation to acquire all of the outstanding shares of PMC common stock is not superior to PMC’s existing agreement with Skyworks Solutions, Inc..

The company's board said the Skyworks all cash proposal provides more value certainty to PMC stockholders than the stock and cash consideration provided in the Microsemi proposal.

http://www.pmc-sierra.com

Ericsson Wins 5-Year Deal with Ooredoo Group

Ooredoo Group awarded a five-year frame agreement to Ericsson for 2G, 3G and 4G radio, core and transmissions solutions. Financial terms were not disclosed.

The group frame agreement includes radio, core and transmission equipment as well as related services. The agreement also covers Ericsson's Software model with its predefined software value packages which ensures better performance and reduces time to market for new functionalities. Using these solutions, Ooredoo's network performance will improve to meet growing end-user expectations.

http://www.ericsson.com


See also