With the exponential growth of data and the increase in businesses that takes advantage of real-time data processing for high performance computing (HPC), data analytics, machine learning, national security and 'Internet of Things' applications, the market needs not only the fastest interconnect available, but also interconnect intelligence that can perform data algorithms as the data moves throughout the data centre.
The new intelligent ConnectX-5 100G adapter enables the most advanced real-time in-network computing engines to unleash business opportunities and new technological developments.
"The new ConnectX-5 100G adapter further enables high performance, data analytics, deep learning, storage, Web 2.0 and more applications to perform data-related algorithms on the network to achieve the highest system performance and utilization", stated Gilad Shainer, vice president, marketing at Mellanox Technologies. "Today, scalable computer and storage systems suffer from data bottlenecks that limit research, product development, and constrain application services. ConnectX-5 will help unleash business potential with faster, more effective, real-time data processing and analytics. With its smart offloading, ConnectX-5 will enable dramatic increases in CPU, GPU and FPGA performance that will enhance effectiveness and maximize the return on data centres' investment."
ConnectX-5 enables greater HPC performance with new Message Passing Interface (MPI) offloads, such as MPI Tag Matching and MPI AlltoAll operations, advanced dynamic routing, and new capabilities to perform various data algorithms. ConnectX-5 delivers the highest available message rate of 200 million messages per second, which is 33 percent higher than the Mellanox ConnectX-4 adapter and nearly 2x compared to competitive products.
ConnectX-5 is the first interconnect adapter to support PCI Express 3.0 and 4.0 connectivity options, and includes an integrated PCIe switch. For upcoming PCI Express 4.0 enabled systems, ConnectX-5 will deliver an aggregated throughput of 200Gb/s.
The new ConnectX-5 Accelerated Switching and Packet Processing (ASAP2) technology enhances Open V-Switch (OVS) offloading, which results in significantly higher data transfer performance without overloading the CPU. Together with native RDMA and RoCE support, ConnectX-5 will dramatically improve Cloud and NFV platform efficiency.
For storage infrastructures, ConnectX-5 introduces new acceleration engines for NVM Express (NVMe). NVMe over Fabrics (NVMf) enables end-users to connect remote subsystems with flash appliances, leveraging RDMA technology to achieve faster application response times and better scalability across virtual data centres. Storage applications will see improved performance and lower latency with the advanced NVMf target offloads that ConnectX-5 delivers.
"IDC research continues to show that InfiniBand is a network of choice for the world's fastest HPCs. Indeed, in a recent IDC study, 30 percent of HPC users surveyed indicated that InfiniBand was the interconnect used in their fastest HPC", stated Bob Sorensen, research vice president, IDC. "Demands for growing HPC performance dictate that more efficient, capable, and intelligent networks will soon be needed to help manage the complex flow of data within a processor/storage fabric, relieve the computational processors from the overheard of data communications housekeeping, and add additional processing and network dataflow support within the network where and when it is needed. Mellanox is taking the first critical steps in formulating and realizing that vision, and the company is signaling that it will remain committed to adding new and higher performance functionality to its smart interconnect product lines."
"With the move to off-premise Cloud services and adoption of hybrid Cloud architectures, enterprises have access to more data and new data analytics possibilities, and we expect to see a significant increase in adoption by organisations looking to do more with their data", stated Cliff Grossner, Ph.D, senior research director of Data Center, Cloud, and SDN at IHS. "There is now a clear growing need for more data processing to be accomplished within the network in order to achieve the performance levels required by artificial intelligence, machine learning and the Internet of Things algorithms. Mellanox's ConnectX-5 features in-network computing capability, which allows algorithms and other tasks to be offloaded from server CPUs, and thus making requisite performance levels possible, is an important advancement for the market."
"The rapid expansion of data makes it imperative that the data moves both as fast as possible and as little as possible, or else it can overwhelm a network", stated Addison Snell, CEO, Intersect360 Research. "Having a high-bandwidth network that is also intelligent and can offload data algorithms, as well as network-related tasks, keeps the rest of the system running efficiently without the network becoming a bottleneck. In-network computing makes applications run faster, which means quicker analysis and time to discovery."
"Dell and Mellanox are long-time partners, delivering Dell HPC Systems utilizing Mellanox InfiniBand, including the recently announced one petaflops system at the Center for High Performance Computing in South Africa", stated Jim Ganthier, vice president and general manager, Engineered Solutions, HPC and Cloud, Dell. "With the announcement of Connect-X5, Mellanox is raising the bar, accelerating HPC performance and results and enabling new levels of innovation for our mutual customers."
"To succeed in today's increasingly competitive market, organisations must harness the latest advancements in compute and interconnect fabric technologies to fuel new applications and extract profit-making insight from their data", stated Scott Misage, vice president and general manager, High Performance Computing, Hewlett Packard Enterprise. "Our industry-leading compute platforms, combined with Mellanox's intelligent ConnectX-5 adapters, can provide customers a significant boost in real-time data processing, improve analytics and usher in a new wave of innovation."
"IBM and the OpenPOWER Foundation are deeply committed to acceleration technologies as they represent a critical piece of the new high performance computing designs emerging today", stated Sumit Gupta, vice president, High Performance Computing and Data Analytics, IBM. "Mellanox's ConnectX-5 technology continues to push the envelope to maximize system throughput for HPC, deep learning, and data analytics applications."
"Bringing more value to high performance computing and the data center is what ConnetX-5 is all about", noted Mr. Leijun Hu, VP of Inspur Group. "We believe Mellanox's innovative product will drive application performance to new levels and vastly improve ROI in the data centre."
"With the advent of high performance applications like Machine Learning and real-time analytics, data centre customers are looking for technologies that have the underpinnings to support next-generation workloads. EDR gave us a great jump in bandwidth, and now Mellanox adds a substantial boost in message rates with ConnectX-5", stated Martin W. Hiegl, WW HPC sales leader at Lenovo.
"SGI continuously focuses on delivering application performance to our HPC users. With our partnership with Mellanox we continue to drive application performance at scale with increasingly larger data workloads", stated Gabriel Broner, vice president and general manager of HPC, SGI. "The ConnectX-5 technology with increased message rates enables us to deliver overall increased system and application performance, offering real and tangible value to our customers."
"Future proofing the data centre is essential to our success and we are pleased to see Mellanox introduce a product aimed at achieving just that", stated Mr. Chaoqun Sha, SVP of Product and Technology of Sugon. "This new technology will enable new levels of interconnect speed while deploying deep interconnect intelligence that lays the foundation for a new era of HPC solutions."