Smart Cities Archives - Page 400 of 570 - SmartCitiesElectronics.com

Nvidia prepares for data centre growth with latest NIC

Mon, May 18, 2020

Claimed to be the industry’s first secure Ethernet smart network interface controller (NIC), the Nvidia Mellanox ConnectX-6 Lx SmartNIC is designed to meet the growth in enterprise and cloud scale-out workloads.

The 25/50Gbits per second Ethernet smart NIC is the 11th generation product in the ConnectX family. It is designed for data centres, where 25Gbits per second connections are becoming standard for handling demanding workflows, such as enterprise applications, artificial intelligence (AI) and real-time analytics.

The ConnectX-6 Lx SmartNIC SmartNIC extends accelerated computing by leveraging software-defined, hardware-accelerated engines to offload more security and network processing from CPUs, explained Nvidia.

Accelerated security features such as IPsec in-line cryptography and hardware root of trust, and a 10-fold performance improvement for connection tracking, enable zero trust security throughout the data centre.

According to the company, best-in-class remote direct memory access (RDMA) over converged Ethernet (RoCE), provides GPUDirect RDMA acceleration for NVMe over Fabrics (NVMe-oF) storage, to accelerate computing and high-speed video transfer applications and to offload networking. The zero touch RoCE (ZTR) affords scalable, easy-to-deploy, best-in-class RoCE without switch configuration, added Nvidia.

The Mellanox ConnectX-6 Lx provides two ports of 25Gbits per second, or a single port of 50Gbits per second and Ethernet connectivity with PCIe Gen 3.0/4.0 x8 host connectivity.

Accelerated switching and packet processing (ASAP2), with built-in SR-IOV and VirtIO hardware offloads for virtualisation and containerization, to accelerate software-defined networking and connection tracking for next-generation firewall services

ConnectX-6 Lx is sampling now, with general availability expected in Q3 2020.

Mellanox SmartNICs and software are used in cloud, hyperscale and enterprise data centers, high-performance computing, cybersecurity, and financial services systems around the world.

Nvidia is credited with inventing the graphics processing unit (GPU) in 1999. The GPU is credited with sparking the growth of the PC gaming market, redefined modern computer graphics, and revolutionised parallel computing. More recently, GPU deep learning ignited modern AI — the next era of computing — with the GPU acting as the brain of computers, robots, and self-driving cars that can perceive and understand the world.

(Pic credit: Vladimir_Timofeev)

http://nvidianews.nvidia.com

> Read More

Robust connectors provide high-speed transmission for industry, says Hirose

Mon, May 18, 2020

The ix Industrial series of I/O connectors have been developed by Hirose for use in space-constrained industrial applications requiring high-speed signal transmission up to 10Gbits per second.

The connector series reduces functional space by up to 75 per cent and footprint by two-thirds compared to RJ45 solutions, says the company. They also support CAT5e (1Gbits per second) and CAT6A (10Gbits per second) cabling.

The ix Industrial series I/O connectors’ EMI/ESD shielding design is optimised for secure data transmission up to 10Gbits per second.

The series is compliant with the standard IEC 61076-3-124. Models in the range have receptacles and plugs with two keying codes, differentiating Ethernet according to IEEE 802.3 from other applications.

There are two receptacle types within the variations. The upright right angle type can be mounted in parallel with a pitch distance of only 10mm to save space. The vertical type allows the mating plug to be mated from the top giving design flexibility.

A horizontal right angle type is currently being developed. This model is designed with a height of 5.7mm, less than half the height of Hirose’s RJ45 connector. This version is intended for use in devices where height is limited. It can also be used for thin-type control panels and FA control equipment.

The receptacle shell is mounted on the board via through-hole solder legs to enhance the retention to the board and to resist the wrenching of the cable assembly, protecting the surface-mount contacts.

The plugs are available in IDC and solder type. The wire-termination unit and cable clamp are integrated into a continuous metal shell, which prevents the connecting part of the cable assembly from being affected by the load on the cable.

Other design features are 10 contact, and a current rating of 1.5A or 3.0A if only fur pins are used.

The voltage rating is 50V AC / 60V DC. The robust connectors are tested to 5,000 mating cycles and to withstand an insertion/extraction force of 25N.

The connectors are RoHS-compliant and have an operating temperature range of -40 to +85 degrees C.

Target applications include factory automation controllers, servo amplifiers, industrial robotics, servers, programmable logic controllers and security systems.

http://www.hirose.com/eu

> Read More

Nvidia offers tools for data centre computing escalation

Fri, May 15, 2020

Following the acquisition of Mellonix, Nvidia is driving data centre changes as they become computing units. In what should have been a keynote at the company’s conference but which was altered to a virtual presentation due to the Covid-19 pandemic, CEO Jensen Huang introduced a GPU architecuture optimised for the new scale of data centre computing.

He announced the Nvidia A100 GPU, based on the Nvidia Ampere architecture, which is claimed to provide “the greatest generational performance leap” of Nvidia’s eight generations of GPUs. It is built for data analytics, scientific computing and cloud graphics, and is in full production and shipping to customers worldwide.

The A100 and the Nvidia Ampere architecture on which it is built, boosts performance by up to 20x over its predecessors. It is claimed to be the world’s largest 7nm processor with over 54bn transistors.

It also features third-generation Tensor Cores with TF32, a new math format that accelerates single-precision artificial intelligence (AI) training, and structural sparsity acceleration, an efficiency technique for AI maths. Another feature is multi-instance GPU (MIG), which allows a single A100 processor to be partitioned into up to seven independent GPUs, each with its own resources.

Also contributing to the increased performance is the third-generation NVLink technology, doubling high-speed connectivity between GPUs. The result is that A100 servers can act as one giant GPU, says Nvidia.

In the same announcement, Nvidia also announced a third generation of its Nvidia DGX AI system based on the A100. The Nvidia DGX A100 is believed to be the world’s first 5PetaFLOPS server. Each DGX A100 can be divided into as many as 56 applications, all running independently.

It allows a single server to either scale up to race through computationally intensive tasks such as AI training, or scale out, for AI deployment, or inference, Huang said.

The A100 will also be available for cloud and partner server makers as HGX A100.

A data centre powered by five DGX A100 systems for AI training and inference running on just 28kW can do the work of a typical data centre with 50 DGX-1 systems for AI training and 600 CPU systems consuming 630kW, Huang explained.

Another announcement was the next-generation DGX SuperPOD, powered by 140 DGX A100 systems and Mellanox networking technology. It offers 700PetaFLOPS of AI performance, the equivalent of one of the 20 fastest computers in the world.

The next-generation DGX SuperPOD delivers 700PetaFLOPS of AI performance.

http://www.nvidia.com

> Read More

Vicor’s latest ChiP-set addresses AI accelerator cards

Fri, May 15, 2020

Designed for high performance graphics processing units (GPUs), central processing units (CPUs) and application specific integrated circuit (ASIC) or XPU processors powered directly from 48V, the 4609 ChiP-set has been added to Vicor’s Power-on-Package portfolio.

A driver, MCD4609, and a pair of MCM4609 current multiplier modules supply up to 650A continuous and 1,200A peak. Vicor reports that the small footprint and low profile (45.7 x 8.6 x 3.2mm) enable the current multipliers to be placed close to the processor, enabling reduced power distribution network (PDN) losses and higher power system efficiency.

Powering GPU and Open Compute Project (OCP) accelerator module (OAM) artificial intelligent (AI) cards, the 4609 ChiP-set is in mass production and available on Vicor’s Hydra II evaluation board.

The 4609 ChiP-set is part of Vicor’s Power-on-Package portfolio of Lateral Power Delivery (LPD) solutions.

Vicor has also revealed that its Vertical Power Delivery (VPD) will soon enable much higher current density. The VPD system delivers current from power converters vertically stacked under a processor through a capacitor network geared to a processor-specific pin-map. A Geared Current Multiplier (GCM) is a specialised VPD device, incorporating a gearbox capacitor network as a layer in the vertical stack. By supplying current directly under the processor and eliminating PDN losses, GCMs will soon facilitate current densities reaching up to 2A/mm², says Vicor.

Vicor IP on the critical path to Power-on-Package LPD and VPD is claimed to enable unparalleled current density and efficient current delivery for advanced processors in applications including AI accelerator cards, AI high density clusters and high-speed networking.

Vicor designs, develops, manufactures and markets modular power components and complete power systems based on a portfolio of patented technologies.

The company is headquartered in Andover, Massachusetts, USA and sells its products to the power systems market, including enterprise and high-performance computing, industrial equipment and automation, telecommunications and network infrastructure, vehicles and transportation, aerospace and defence.

http://www.vicorpower.com

> Read More

Nvidia prepares for data centre growth with latest NIC

Robust connectors provide high-speed transmission for industry, says Hirose

Nvidia offers tools for data centre computing escalation

Vicor’s latest ChiP-set addresses AI accelerator cards

About Smart Cities