Nvidia acquires Mellanox for $6.9bn

Nvidia is to buy Mellanox in a $6.9bn deal that values each share at $125. Together, Nvidia’s computing platform and Mellanox’s interconnects power more than 250 of the world’s TOP500 supercomputers and can boast every major cloud service provider and computer maker as customers. ‘The emergence of AI and data science,

as well as billions of simultaneous computer users, is fuelling skyrocketing demand on the world’s datacentres,’ said Jensen Huang, founder and CEO of Nvidia. ‘Addressing this demand will require holistic architectures that connect vast numbers of fast computing nodes over intelligent networking fabrics, to form a giant datacentre-scale compute engine. ‘We’re excited to unite Nvidia’s

accelerated computing platform with Mellanox’s world-renowned accelerated networking platform under one roof, to create next-generation datacenter-scale computing solutions. I am particularly thrilled to work closely with the visionary leaders of Mellanox and their amazing people, to invent the computers of tomorrow.’ With Mellanox, Nvidia will optimise

datacenter-scale workloads across the entire computing, networking and storage stack to achieve higher performance, greater utilisation and lower operating


Altair adds Cetim’s Quilted Stratum Design to its partner alliance

Altair has announced that Cetim’s Quilted Stratum Design (QSD) software has been added to the Altair Partner Alliance. ‘We are excited to work with Cetim to provide a powerful, yet easy-to-use solution for evaluating composite layup strategies. QSD offers Altair customers an optimisation approach for designers to rapidly design high-performing and cost-controlled composite parts,’ said James Dagg, Altair’s CTO of design and

simulation solutions. QSD offers designers an

intuitive workflow within the Altair HyperMesh environment that benefits from Altair OptiStruct advanced technologies for a smart composite design with optimum thicknesses, fiber orientation and reduced material scraps. QSD assists in designing efficient structural parts made from hot-pressed, multi-thickness preforms of thermoplastic plies. The QSD method works | @scwmagazine

with all ‘tailored preform’ processes (tape placement, QSP, hand lay-up.) ‘With QSD, Cetim

provides access to the best practices of composite optimisation developed during our research project with ONERA, the French Aerospace Lab. The partnership with Altair gives HyperWorks users access to best-in-class software which is already used in numerous design offices,’ said Damien Guillon, composite R&D manager at Cetim.

Jensen Huang

Optalysys launch optical processing system

Optalysys, known for developing light-speed optical AI systems, has announced the first optical co-processor system, the FT:X2000. Optical processing comes

cost for customers. ‘We share the same vision for accelerated computing as Nvidia,’ said Eyal Waldman, founder and CEO of Mellanox. ‘Combining our two companies comes as a natural extension of our longstanding partnership and is a great fit given our common performance- driven cultures. This combination will foster the creation of powerful technology and fantastic opportunities for our people.’ The companies have a long history

of collaboration and joint innovation, reflected in their recent contributions in building the world’s two fastest supercomputers, Sierra and Summit, operated by the US Department of Energy. Many of the world’s top cloud service providers also use both Nvidia GPUs and Mellanox interconnects. Nvidia and Mellanox share a common

performance-centric culture, that the firms say will enable seamless integration.

at a pivotal time of change in computing. Demand for AI is exploding just as silicon- based processing is facing the fundamental problem of Moore’s Law breaking down. Optalysys is addressing this problem with its revolutionary optical processing technology – enabling new levels of AI performance for high- resolution image and video-based applications. CEO and founder Dr Nick

New, said: ‘It is a really exciting time in optical computing. As we approach the commercial launch of our main optical co-processor systems, we are seeing a surge in interest in optical methods, which are needed to provide the next level of processing capability across multiple industry sectors. We are on the verge of an optical computing revolution and it’s fantastic to be leading the way.’ The technology is being

developed to accelerate some of the most demanding processor- intensive tasks at a fraction of the energy consumption of silicon processors. It can be programmed through the API or Tensorflow interface to perform optical correlation and convolution functions to unlock new levels of AI and pattern recognition capability. The FT:X2000 is an entry-level system that operates at up to 2,400fps with a 2,048x1,536 resolution. The Optalysys technology

also uses photons in place of electrons, but crucially performs high-resolution calculations that enable large image/pattern-based data to be processed at speeds far faster than in silicon. By exploiting the properties of diffractive optics, the Optalysys technology can offer something unique in the AI space – a scalable processor that can perform end-to-end, full-resolution processing of multi-megapixel image and video data, or contextually pre-process data for boosting the performance of existing Convolutional Neural Network (CNN)-type models for high-resolution data applications.

April/May 2019 Scientific Computing World 27

Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32