Dell, EMC, Dell Technologies, Cisco,

Showing posts with label Infiniband. Show all posts
Showing posts with label Infiniband. Show all posts

Monday, December 4, 2017

Mellanox and NEC Partner to Deliver Innovative High-Performance and Artificial Intelligence Platforms

#MellanoxTechnologies, Ltd. (NASDAQ: #MLNX), a leading supplier of high-performance, end-to-end smart interconnect solutions for #datacenter servers and storage systems, today announced in collaboration with #NEC Corporation support for the newly announced #SXAurora #TSUBASA systems with @Mellanox® #ConnectX® #InfiniBand adapters. A typical Aurora server platform includes from one to four InfiniBand adapters. The top-of-the-line Aurora platform utilizes 32 ConnectX adapters to support 64 vector engines in a single system. The NEC SX-Aurora TSUBASA systems leverage general-purpose vector-based NEC coprocessors and Mellanox in-network computing interconnect accelerators to achieve the highest application performance and scalability. “We appreciate the performance, efficiency and scalability advantages that Mellanox interconnect solutions bring to our platform,” said Shigeyuki Aino, assistant general manager system platform business unit, IT platform division, NEC Corporation. “The in-network computing and PeerDirect capabilities of InfiniBand are the perfect complement to the unique vector processing engine architecture we have designed for our SX-Aurora TSUBASA platform.” “Mellanox is proud to work with NEC to enable a next-generation computational platform for high-performance computing, machine learning, cloud and more,” said Gilad Shainer, vice president of marketing at Mellanox Technologies. “The combination of Mellanox ConnectX adapters, in-network computing, and acceleration engines, with NEC vector processing, provides our users with a world-leading compute platform that enables the highest application performance and the best return on investment.” Mellanox InfiniBand solutions deliver the highest efficiency for high performance, artificial intelligence, cloud, storage and more infrastructures. InfiniBand accelerates all of the compute architectures - X86, Power, GPU, ARM, FPGA and Vector-based compute and storage platforms - delivering highest flexibility and best return on investment

http://news.sys-con.com/node/4205297

Tuesday, November 14, 2017

Mellanox Propels NetApp to New Heights With 100Gb/s InfiniBand Connectivity

ConnectX® Network Adapters Enable High Performance NVMe-over-InfiniBand Connectivity for Two of NetApp’s Most Advanced Storage Platforms November 14, 2017 08:30 AM Eastern Standard Time SUNNYVALE, Calif. & YOKNEAM, Israel--(BUSINESS WIRE)-- #SUPERCOMPUTING 2017 – @Mellanox Technologies, Ltd. (NASDAQ: MLNX), a leading supplier of high-performance, end-to-end smart interconnect solutions for data center servers and storage systems, today announced that @NetApp has selected ConnectX #100Gb /s @InfiniBand adapters for two of its most advanced storage platforms: the EF570 and E5700. Both the EF570 all-flash system and the E5700 hybrid flash system now support 100Gb/s NVMe over Fabrics on InfiniBand with Mellanox ConnectX-4 based adapters. The platform delivers unmatched price-performance that accelerates database, technical computing, big data analytics, and video processing workloads. “NetApp is one of the first, if not the first, enterprise vendor to support NVMe over Fabrics with 100Gb/s InfiniBand. Given Mellanox’s historic leadership in high-speed networking, it’s no surprise that they are doing it with Mellanox-based adapters.” Tweet this NetApp’s EF570 is the storage industry’s price/performance leader as demonstrated by SPC-1™ and SPC-2 ™ benchmark tests. As the first available enterprise-class NVMe-over-InfiniBand systems, the solution breaks new ground with 1M+ sustained IOPs, 21GB/s throughput, and sub-100 microsecond latency. This performance is critical to perform real-time analytics to allow businesses to gain insight from an ever increasing amount of unstructured data, as well as to drive faster database, analytics, video processing, and technical computing performance. The EF5700 also offers 100Gb/s NVMe-over-InfiniBand in a hybrid flash system that is ideal for enterprise applications. The ConnectX family of Virtual Protocol Interconnect® adapters support both InfiniBand and Ethernet and offer unmatched RDMA (Remote Direct Memory Access) features and capabilities and future-proofs data center investments by supporting speeds of 10, 25, 40, 50, and 100Gb/s. With full support for RDMA over Converged Ethernet (RoCE), ConnectX-4 allows platforms to support the latest storage protocols including iSCSI, iSER, NVMe-over-Fabrics, and SRP (SCSI RDMA Protocol). “NetApp is an innovation leader when it comes to storage performance, connectivity, and efficiency,” said Eric Stoltman, vice president and general manager, Hyperscale Group at NetApp. “We are proud to work with Mellanox to deliver faster network connections, innovative new protocols, and faster performance to our customers in our E-series arrays.” “NVMe over Fabrics is the newest, most efficient block storage protocol and EDR InfiniBand is the fastest, lowest-latency storage fabric technology,” said Michael Kagan, CTO, Mellanox Technologies. “It’s very exciting to see NetApp support both with their EF570 and E5700 platforms.” “It’s clear that faster storage controllers and faster drives need faster network connections to realize their full value,” said Arun Taneja Founder & Consulting Analyst, Taneja Group. “NetApp is one of the first, if not the first, enterprise vendor to support NVMe over Fabrics with 100Gb/s InfiniBand. Given Mellanox’s historic leadership in high-speed networking, it’s no surprise that they are doing it with Mellanox-based adapters.”

http://www.businesswire.com/news/home/20171114005074/en/Mellanox-Propels-NetApp-New-Heights-100Gbs-InfiniBand

Tuesday, October 24, 2017

In-Network Computing and Next Generation HDR 200G InfiniBand

UThree #HPC practitioners stand at the dawn of the fourth industrial revolution. What do they call “ #BigData”? The humorous answer is of course, just “Data,” but the reality is that with the exponential growth of data that needs to be analyzed and the data resulting from ever-more complex workflows, the need for data movement has never been more challenging and critical to the worlds of High Performance Computing (HPC) and machine learning. @Mellanox Technologies, the leading global supplier of end-to-end @InfiniBand interconnect solutions and services for servers, storage, and #hyperconvergedinfrastructure, is once again moving the bar forward with the introduction of and end-to-end HDR 200G InfiniBand product portfolio. When Fast Isn’t Fast Enough In an age of digital transformation and big data analytics, the need for HPC and deep learning platforms to move and analyze data both in real-time and at faster speeds is ever increasing. Machine learning enables enterprises to leverage the vast amounts of data being generated today to make faster and more accurate decisions. Whether for brain mapping or for homeland security, the most demanding supercomputers and data center applications need to produce astounding achievements, often in real time! Until relatively recently, state-of-the-art applications analyzing automotive construction or weather simulations enjoyed the data performance and throughput speeds offered by 100G interconnect. Today’s HPC, machine learning, storage and hyperscale now require both faster interconnect solutions and more intelligent networks to analyze data and run complex simulations with greater speed and efficiency. With the expected volume of network data availability doubling by the end of 2019, it only makes sense to prepare for the eventuality of demand for HDR 200G interconnectivity capabilities.

https://www.hpcwire.com/2017/10/23/network-computing-next-generation-hdr-200g-infiniband/

Tuesday, August 8, 2017

Mellanox Announces BlueField Storage Solutions to Accelerate NVMe over Fabrics

SUNNYVALE, Calif. & YOKNEAM, Israel, Aug. 8, 2017 — #MellanoxTechnologies, Ltd. (NASDAQ: MLNX) today announced the availability of storage reference platforms based on its revolutionary #BlueField System-on-Chip ( #SoC ), combining a programmable multicore CPU, networking, storage, security, and virtualization acceleration engines into a single, highly integrated device. BlueField integrates all the technologies needed to connect #NVMe over Fabrics flash arrays, with the fastest performance available in the market. BlueField provides 200 Gb/s of throughput and more than 10 million IOPS in a single SoC device. In addition, the powerful on-board multicore ARM processor subsystem enables flexible programmability that allows vendors to differentiate their software-defined storage appliances with advanced capabilities. This makes BlueField the ideal chip to control and connect All Flash Arrays and Just-a-Bunch-Of-Flash (JBOF) systems to InfiniBand and Ethernet Storage fabrics. “BlueField is the most highly integrated NVMe over Fabrics solution,” said Michael Kagan, CTO of Mellanox. “By tightly integrating high-speed networking, programmable ARM cores, PCIe switching, cache, memory management, and smart offload technology all in one chip; the result is improved performance, power consumption, and affordability for flash storage arrays. BlueField is a key part of our Ethernet Storage Fabric solution, which is the most efficient way to network and share high-performance storage.” “As data takes over the world, networked storage takes over computing,” said Peter Burris, GM and Chief Research Officer of Wikibon. “The BlueField SoC controller is a leading example of the right technology at the right time.” BlueField and the BlueField flash array reference platform are being shown at Flash Memory Summit, Aug. 8-10, at the Santa Clara Convention Center, Mellanox booth, #138. BlueField samples and storage reference design systems will be available in Q4, 2017. Supporting Quotes: “We use Mellanox NICs and RDMA today, and are excited that BlueField integrates all the technologies needed to connect NVMe over Fabrics flash arrays,” said Yaniv Romem, CTO at Excelero. “We are looking forward to seeing the fastest performing SoC come to market in the form of Mellanox BlueField technology.” “Continuous innovation around NIC and RDMA solutions is why E8 Storage is proud to collaborate with Mellanox to deliver an end-to-end shared NVMe solution for high-performance enterprise storage applications,” said Ziv Serlin, VP System Architecture, E8 Storage. “With 200 Gb/s of throughput and up to 10 million IOPS in a single SoC device, BlueField will bring increased performance and flexible programmability to the industry.” “BlueField will power a new class of flash storage designs based on Mellanox’s robust SoC,” said Evan Chien, Director of Inventec. “We are pleased to see this advancement as BlueField enables a new level of flexible programmability that will enable us to differentiate our products with additional advanced capabilities.” “The Mellanox BlueField technology and ConnectX-5 adapters can pass storage traffic directly to our flash controllers and NVRAM devices, thereby greatly enhancing NVMe over Fabrics performance,” said Derek Dicker, vice president and business unit manager for performance storage at Microsemi. “Our SwitchTek PCIe switches and FlashTek SSD controllers are the ideal fit for flash storage systems using the BlueField SoC and we look forward to leveraging this technology.” “BlueField holds the promise of greatly enhancing our product offerings,” said Eugene McCabe, EVP, Sanmina. “As we continue to improve our offerings, BlueField will enable us to offer our customers increased performance and efficiency, thereby delivering a more robust system solution.” “BlueField is a brand new, highly integrated device,” said Peter Tung, Chief Operation Officer for Enterprise Business Group, Wistron. “The advanced performance and capabilities makes it an elegant solution for controlling and connecting our All Flash Arrays to Mellanox InfiniBand and Ethernet storage fabrics.”
https://www.hpcwire.com/off-the-wire/mellanox-announces-bluefield-storage-solutions-accelerate-nvme-fabrics/

Friday, July 22, 2016

Mellanox plans to SoC it to storage speed with Multi-ARM BlueField

#InfiniBand and Ethernet adaptor biz #Mellanox has storage acceleration SoCs coming to provide faster external array access across NVMe fabrics. Mellanox CEO @EyalWaldman talked about NVME over Fabrics ( #NVMeF ) in his company's second quarter 2016 earnings call. His company makes InfiniBand and Ethernet networking devices, and is doing very well with a fifth consecutive record quarter's revenues. NVMeF is a way of connecting an array of external NVMe flash drives to accessing servers over a network link that is so fast array data access are the same as ones to local, directly connected NVMe flash drives. Vendors like #EMC with DSSD and Mangstor have product available now. E8 will launch at next month's Flash Memory Summit. Kaminario and #Tegile have plans to adopt the technology, and both NetApp and Pure Storage are keeping a close eye on it. During the call Waldman said: "Integration efforts between #EZchip and Mellanox have been ongoing since the acquisition closed in late February. During the second quarter, we announced the first joint product, our multi-core system-on-a-chip solution named BlueField. BlueField combines multiple ARM cores interconnected by a coherent mesh network with the leading ConnectX-5 network adapter. Initial target markets include storage and network applications." In storage the multi-ARM core BlueField system-on-a-chip (SOC) could act as a controller for a group of NVMe flash drives, with access over 25/50/100 Gbit/sec Ethernet. Waldman then said; "Utilising offload accelerators within the embedded ConnectX-5 adapter, BlueField delivers a programmable intelligent network solution. Within storage markets, BlueField can be utilized in scale-out NVMe over fabric systems as storage controller."
http://www.theregister.co.uk/2016/07/22/mellanox_punting_multiarm_bluefield_into_greenfield/

Tuesday, July 12, 2016

Africa’s fastest computer built at CSIR

The Centre for High Performance Computing (CHPC) of the Council for Scientific and Industrial Research (CSIR) has created the fastest supercomputer on the continent, with processing speed capable of a thousand trillion floating point operations per second. Floating point operations (flops) are used in computing to calculate extremely long numbers. With over 40 000 cores, the new machine is the fastest computer in Africa owing to its speed of roughly 1 petaflop (1000 teraflops). That is 15 times faster than the system it succeeds, which was named Tsessebe (Setswana for antelope). Aptly, the new beast is called Lengau, which is the Setswana word for cheetah. Dr Thomas Auf der Heyde, Deputy Director-General: Research Development and Support at the Department of Science and Technology (DST), outlined the role high-performance computing plays in growing the economy. “For our country to grow at the required rate, as set out in the National Development Plan, it needs to change gear by building capacity in the production and dissemination of knowledge,” he explained. “The CHPC represents a deliberate move by this country to invest in modernising our research and development. High-performance computing and advanced data technologies are powerful tools in enhancing the competiveness of regions and nations.” Dr Happy Sithole, the director of CHPC, detailed the journey leading to the unveiling of the new PFLOPs machine: “When we started in 2007, we took inspiration from the fastest animals in the land and named our first high-performance computing system iQudu (Xhosa for kudu) which boasted 2,5 teraflops). “In 2009 there was increased demand for computational resources, and a new high-performance computing system dubbed the Tsessebe was launched. It boasted 24,9 teraflops and became number 311 on the TOP500 #supercomputer s, and ranked number one in the African continent. The system was later upgraded to 64,44 teraflops,” he said. The technology powering Lengau means that it is also smaller in footprint than the previous system. The #Dell HPC system is comprised of 1039 Dell #PowerEdge servers based on Intel Xeon processors, totalling 19 racks of compute nodes and storage. It has a total #DellStorage capacity of 5 petabytes, and uses #DellNetworking Ethernet switches and #Mellanox EDR #InfiniBand with a maximum interconnect speed of 56 GBps.

http://www.dataweek.co.za/8682a

Wednesday, December 2, 2015

Genomics and HPC

#Genomesequencing is a technology that can takes advantage of the growing capability of todays ‘ modern #HPC systems. #Dell is leading the charge in the area of personalized medicine by providing highly tuned systems to perform genomic sequencing and data management. The whitepaper, The InsideHPC Guide to Genomic is a overview of how Dell is providing state-of-the-art solutions to the life science industry.
There are many aspects of creating a solution that can manage tremendous amounts of data and perform fast sequencing. The system must be architected and tuned to use the right combination of CPUs, memory, network and storage to provide a balanced solution. By integrating the appropriate components, Dell has created an optimized system that has been tested for a variety of applications and storage requirements. The Dell Genomic Data Analysis Platform ( #GDAP ) includes the latest Intel Xeon processors, #InfiniBand and 10 GB Ethernet, as well as the latest #Intel Studio software.

http://insidehpc.com/2015/12/genomics-and-hpc/