Today Tachyum announced its participation and support for the Open Euro High Performance Computing Project (OEUHPC). OEUHPC is a consortium of innovators, developers and users who are being brought together to create reference designs for high-performance, power-efficient, flexible and scalable Data Center/AI/HPC solutions building on the scalability of an Open Compute model.
The capability to simulate the human brain in real time is no longer a ‘Can we do it?’ question but rather a ‘How soon can we get it done?’ challenge,” said Dr. Radoslav Danilak, Co-founder and CEO of Tachyum. “We believe that the goals of the OEUHPC Project – to create a new class of Universal Computing Centers (UCCs) throughout Europe and around the world that are AI and HPC enabled – dovetails with Tachyum’s interests in bringing to market a new class of “Universal Processor”, capable of efficiently handling the convergence of data center, AI, and HPC workloads. As a member of the OEUHPC Project, together we can ascertain how to best extract maximum performance from underlying hardware fabrics, to deliver vastly more computationally powerful and energy efficient data centers which can change the face of the industry.”
The OEUHPC Project is dedicated to re-inventing high-performance computing from expensive proprietary supercomputers towards low cost open systems, making it accessible, ultra-low power and more flexible with unparalleled scalability, far beyond conventional supercomputers. Success of the project will facilitate the democratization of HPC and AI, enabling broad, cost effective access to cloud-based state-of-the-art HPC resources, not only by governments and corporations, but also by small to medium enterprises and even individual entrepreneurs.
Tachyum is supporting the OEUHPC Project due to its ability to satisfy the need for both scalable and converged computing, which, in part, will ultimately enable users to cost effectively simulate, in real-time, human brain-sized Neural Networks. Tachyum has been working to develop its ultra-low power Prodigy Universal Processor Chip to allow system integrators to build a 32 Tensor ExaFLOPS AI supercomputer in 2020, well ahead of the scheduled EU goal to achieve 1 ExaFLOPS in 2028.
Tachyum has built strategic relationships with partners that will deliver supercomputer and cloud systems to enable the EU to exploit the unprecedented capabilities of the Prodigy Universal Processor Chip, the smallest and fastest general purpose, 64-core processor developed to date, requiring 10x less power and reducing processor cost by 3x. Tachyum’s Prodigy processor reduces data center server power by an order of magnitude, through a disruptive hardware architecture and a smart compiler that has made many parts of the hardware found in a typical processor redundant. Fewer wires and shorter wires, due to a smaller, simpler core, translates into much greater speed and power efficiency for the processor.
By unleashing the EU’s corporate, SME, and academic prowess, focused on a common goal – the development and demonstration of Europe’s Universal Computing Center (UCC) model – the OEUHPC Project can serve as a focal point and a critical enabler in the resurgence of EU leadership in cloud computing infrastructure and AI/HPC capabilities.