4 minutes to read
SANTA CLARA, Calif., June 30, 2020 – Tachyum™ Inc. today announced that it has joined the Software for Open Networking in the Cloud (SONiC) Open Compute Project as part of the company’s efforts to help redesign traditional IT infrastructure technology to better serve the technical needs of modern data centers.
SONiC is an open-source Network Operating System (NOS) community project of the OCP. Originally developed to improve network switch operations, Microsoft contributed SONiC to the OCP community as a way to further its development and extend its reach beyond the large web-scale companies and cloud providers that had the means to deploy and support it within their own ecosystems. As an active member of OCP, Tachyum is contributing resources to SONiC that enable resulting hardware and software solutions as part of the project to be able to overcome IT limitations regarding performance, power efficiency and cost.
Tachyum’s contribution to SONiC are an extension of the progress it has made on its 64-core flagship Prodigy Universal Processor, which is scheduled for volume production in 2021. It outperforms the fastest Xeon processors at 10x lower power (core vs. core) on data center workloads, as well as outperforming NVIDIA’s fastest A100 GPU on neural net AI training and inference. Due to its high computational density and I/O bandwidth, networks of Prodigy processors comprising just 125 HPC racks, can deliver an ExaFLOPS (a billion, billion double precision floating point operations per second) of capacity. Prodigy’s 3X lower cost per MIPS compared to other CPU competition, coupled with its 10X processor power savings, translates to a 4X reduction in Data Center TCO (Annual Total Cost of Ownership: CAPEX + OPEX). Even at 50 percent Prodigy attach rates, this translates to billions of dollars per year in real savings for hyperscalers such as Google, Facebook, and Amazon.
Since Prodigy can seamlessly and dynamically switch from data center workloads to AI or HPC workloads, unused servers can be powered up, on demand, as ad hoc AI or HPC networks – CAPEX free, since the servers themselves are already purchased. Every Prodigy-provisioned data center, by definition, becomes a low-cost AI center of excellence, and a low-cost HPC system.
“Our purpose in joining the OCP was to advance common-ground technological innovations and ensure their interoperability in order to advance beyond the limitations holding back the industry at large,” said Dr. Radoslav Danilak, Tachyum founder and CEO. “We continue that pursuit in moving to support the open-source SONiC community in reference designs for Tachyum HPC/AI customers. Our contributions will be part of the collective strength of this growing ecosystem and community as we develop the network solutions to overcome the limitations facing data centers today.”
SONiC is an open source network operating system based on Linux that runs on switches from multiple vendors and ASICs. SONiC offers a full-suite of network functionality, like BGP and RDMA, that has been production-hardened in the data centers of some of the largest cloud-service providers. It offers teams the flexibility to create the network solutions they need while leveraging the collective strength of a large ecosystem and community. Additional information about the project can be found at opencompute.org/projects/sonic-sub-project.
Tachyum is leading the way in delivering transformative semiconductor solutions that provide industry-leading speed, power efficiency and cost. Whether enabling supercomputers more powerful than the human brain or turning everyday data centers into low-cost HPC and AI centers of excellence, Tachyum’s unique approach disrupts markets and industries worldwide. Co-founded by seasoned architect and Silicon Valley entrepreneur Dr. Radoslav Danilak, Prodigy, the world’s first and only universal processor, begins volume production in 2021 and targets a $50B+ market that is increasing by more than 20 percent a year. With data centers consuming 3 percent of the planet’s electricity – and doubling every 5 years – Tachyum’s ultra-low power Prodigy is critical for continued data center growth. Tachyum has US offices in Santa Clara, California, and EU offices in Bratislava, Slovakia.