Tachyum demonstrates native KVM virtualization on Prodigy emulation
LAS VEGAS–(BUSINESS WIRE)–Tachyum™ today announced the latest advancements in its Prodigy software ecosystem with KVM hypervisor virtualization that enables data center providers to run multiple instances of a bare metal operating system on the same machine with minimum overload. This Day One capability dramatically reduces the cost and power requirements of data center, AI, and HPC workloads.
KVM (for Kernel-based Virtual Machine) is a complete virtualization solution for Linux on x86 hardware containing virtualization extensions. It consists of a loadable kernel module, kvm.ko, which provides the basic virtualization infrastructure, and a processor-specific module, kvm-intel.ko or kvm-amd.k. Tachyum Prodigy processors support Type 2 (i.e. hosted) hypervisors via a Linux kernel in the KVM module to mediate all interactions between hardware and virtual machines.
Along with virtualization, Prodigy enables resource partitioning and management, device virtualization (eg, a shared USB port), and symmetric multiprocessing (SMP) allowing a single virtual machine to simultaneously use two or more processors. Tachyum additionally provides advanced security to hosted virtual machines. This latest advancement in KVM virtualization joins Tachyum’s IOMMU and other hardware-proven hardware virtualizations.
“Our software team is once again advancing the state of Prodigy by adding features that will allow future customers to benefit from it at launch,” said Dr. Radoslav Danilak, Founder and CEO of Tachyum. “By enabling KVM virtualization, we are providing an attractive option for data center providers looking to leverage the power, performance and economic benefits of Prodigy. This achievement is one of many software enhancements, including compilers and libraries, bootloaders and monitors, operating system kernels and more, that we plan to implement in Prodigy in 2022. »
Tachyum’s Prodigy processor can run HPC, AI, ML, and DL applications, as well as normal data center workloads, on a single, homogeneous processor platform, using existing standard programming models. Without Prodigy, HPC centers must choose between single-threaded performance or high throughput, building systems that combine disparate processors, GPUs, and/or other acceleration hardware for these different workloads, which creates inefficiency, expense and complexity of separate provisioning 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 difficult programming, support, and maintenance. Prodigy’s ability to seamlessly switch between these different workloads using existing programming models and legacy code support, while delivering high performance, will dramatically change the competitive landscape and economics. data centers.
KVM virtualization is available for early adopters to test now through the Tachyum Customer and Partner Portal. Testing the full virtualization stack before going out of tape ensures that production chips perform as expected when they leave the factory.
A video demonstration of KVM running on a Prodigy emulation platform is currently available at https://youtu.be/uFIQaTcwfSA.
Tachyum is transforming the AI, HPC, and 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 purpose and purpose computing general. When Prodigy processors are delivered in a large-scale data center, they enable all AI, HPC, and general-purpose applications to run on hardware infrastructure, saving businesses billions of dollars. 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 more than 3% of the planet’s electricity, up from 10% by 2025, the ultra-low-power Prodigy Universal Processor is essential if we are to continue doubling global data center capacity every four years. years. Tachyum, co-founded by Dr. Radoslav Danilak with its flagship product Prodigy, is heading for tape release and chip sampling in 2022, with software emulations and an FPGA-based emulator running native Linux available for early adopters . The company is building the world’s fastest 64 exaflops AI 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/.