読むのに 4 分
プレスリリース
LAS VEGAS, Nevada, March 22, 2022 – Tachyum™ today announced the latest progress achieved in its Prodigy software ecosystem with KVM hypervisor virtualization that empowers data center providers with the ability to execute several instances of a bare-metal OS on the same machine with minimal overhead. This Day One functionality massively reduces the cost and the power requirements of data center, AI and HPC workloads.
KVM (for Kernel-based Virtual Machine) is a full virtualization solution for Linux on x86 hardware containing virtualization extensions. It consists of a loadable kernel module, kvm.ko, that provides the core virtualization infrastructure and a processor specific module, kvm-intel.ko or kvm-amd.k. Tachyum Prodigy processors support Type 2 (aka hosted) hypervisors through a Linux kernel within the KVM module to mediate all interactions between hardware and virtual machines.
With virtualization, Prodigy enables resource partitioning and management, device virtualization (e.g. shared USB port), and symmetric multiprocessing (SMP) allowing a single VM to use two or more processors simultaneously. Tachyum additionally provides advanced security to hosted virtual machines. This latest advancement in KVM virtualization is in addition to Tachyum’s IOMMU and other hardware virtualization that is being proven in hardware.
“Our software team once again advances the state of Prodigy by adding functionality that will allow future customers to benefit at the time of launch,” said Dr. Radoslav Danilak, founder and CEO of Tachyum. “By enabling KVM virtualization, we offer an appealing option for data center providers looking to leverage the power, performance and cost benefits of Prodigy. This achievement is one of many software enhancements, including compilers and libraries, boot loaders and monitors, OS kernels and more, we plan to implement to Prodigy in 2022.”
Tachyum’s Prodigy processor can run HPC, AI, ML and DL applications, plus normal data center workloads, on a single homogeneous processor platform, using existing standard programming models. Without Prodigy, HPC centers must choose either single thread performance or high throughput, building systems combining disparate CPU, GPU and/or other accelerator hardware for these different workloads, creating inefficiency, expense, and the complexity of separate supply and maintenance infrastructures. Using specific hardware dedicated to each type of workload (e.g. data center, AI, HPC), results in underutilization of hardware resources, and more challenging programming, support, and maintenance. Prodigy’s ability to seamlessly switch among these various workloads using existing programming models, and legacy code support, while providing high performance, will dramatically change the competitive landscape and the economics of data centers.
KVM virtualization is available to early adopters for testing now through the Tachyum customer and partner portal. Testing of the full virtualization stack before tape-out ensures that production chips function as intended when they come out of manufacturing.
A video demonstration of KVM running on a Prodigy emulation platform is currently available for viewing below.
https://www.linkedin.com/company/tachyum
https://www.facebook.com/Tachyum/
Tachyum is transforming AI, HPC, public and private cloud data center markets with Prodigy, the world’s first Universal Processor that delivers industry-leading performance, cost, and power efficiency for both specialty and general-purpose computing. When Prodigy processors are provisioned in a hyperscale data center, they enable all AI, HPC, and general-purpose applications to run on one hardware infrastructure, saving companies billions of dollars per year. A fully functional Prodigy emulation system is currently available to select customers and partners for early testing and software development. With data centers currently consuming over 3% of the planet’s electricity, predicted to be 10% by 2025, the ultra-low power Prodigy Universal Processor is critical, if we want to continue doubling worldwide data center capacity every four years. Tachyum, Co-founded by Dr. Radoslav Danilak with its flagship product Prodigy, is marching towards tape out and chip sampling in 2022, with software emulations and an FPGA-based emulator running native Linux available to early adopters. The company is building the world’s fastest 64 AI exaflops supercomputer in 2022 in the EU with Prodigy chips. Tachyum has offices in the United States and Slovakia. For more information, visit https://www.tachyum.com/.