Wednesday, August 17th 2022
Tachyum Submits Bid for 20-Exaflop Supercomputer to U.S. Department of Energy Advanced Computing Ecosystems
Tachyum today announced that it has responded to a U.S. Department of Energy Request for Information soliciting Advanced Computing Ecosystems for DOE national laboratories engaged in scientific and national security research. Tachyum has submitted a proposal to create a 20-exaflop supercomputer based on Tachyum's Prodigy, the world's first universal processor.
The DOE's request calls for computing systems that are five to 10 times faster than those currently available and/or that can perform more complex applications in "data science, artificial intelligence, edge deployments at facilities, and science ecosystem problems, in addition to the traditional modeling and simulation applications."
The DOE request prefers systems that "operate within a power envelope of 20-60 MW." Tachyum's submission would meet these challenging requirements.
The DOE is interested in deploying this system as soon as 2025.
In its bid, Tachyum introduced its advanced universal processor technology and long-term roadmap that includes Prodigy, and Prodigy 2. The first-generation Prodigy supercomputer processor chip offers 4x the performance of the fastest Xeon, 3x more raw performance than NVIDIA's H100 on HPC, 6x more raw performance on AI training and inference workloads, and up to 10x performance at the same power.
"Tachyum's revolutionary design could deliver strategic superiority to the United States and its Department of Energy's NNSA. In addition, it will provide high performance, low TCO, low energy consumption, ease of deployment and low maintenance," said Dr. Radoslav Danilak, founder and CEO of Tachyum. "The DOE should be particularly interested in the carbon footprint reduction, as both a matter of environmental stewardship and national security."
Tachyum's Prodigy integrates 128 high-performance custom designed 64-bit compute cores with the functionality of a CPU, a GPU, and a TPU in a single device within a homogeneous architecture. This allows Prodigy to deliver performance up to 4x that of the highest performing x86 processors (for cloud workloads) and up to 3x that of the highest performing GPU for HPC and 6x for AI applications.
Prodigy's unique architecture delivers industry-leading performance in both data center and AI workloads. Therefore, during off peak hours, Prodigy-powered data center servers can be seamlessly and dynamically switched to AI workloads, eliminating the need for expensive dedicated AI hardware, and dramatically increasing server utilization. Prodigy delivers unprecedented data center performance, power, and economics, reducing CAPEX and OPEX significantly.
The DOE's request calls for computing systems that are five to 10 times faster than those currently available and/or that can perform more complex applications in "data science, artificial intelligence, edge deployments at facilities, and science ecosystem problems, in addition to the traditional modeling and simulation applications."
The DOE is interested in deploying this system as soon as 2025.
In its bid, Tachyum introduced its advanced universal processor technology and long-term roadmap that includes Prodigy, and Prodigy 2. The first-generation Prodigy supercomputer processor chip offers 4x the performance of the fastest Xeon, 3x more raw performance than NVIDIA's H100 on HPC, 6x more raw performance on AI training and inference workloads, and up to 10x performance at the same power.
"Tachyum's revolutionary design could deliver strategic superiority to the United States and its Department of Energy's NNSA. In addition, it will provide high performance, low TCO, low energy consumption, ease of deployment and low maintenance," said Dr. Radoslav Danilak, founder and CEO of Tachyum. "The DOE should be particularly interested in the carbon footprint reduction, as both a matter of environmental stewardship and national security."
Tachyum's Prodigy integrates 128 high-performance custom designed 64-bit compute cores with the functionality of a CPU, a GPU, and a TPU in a single device within a homogeneous architecture. This allows Prodigy to deliver performance up to 4x that of the highest performing x86 processors (for cloud workloads) and up to 3x that of the highest performing GPU for HPC and 6x for AI applications.
Prodigy's unique architecture delivers industry-leading performance in both data center and AI workloads. Therefore, during off peak hours, Prodigy-powered data center servers can be seamlessly and dynamically switched to AI workloads, eliminating the need for expensive dedicated AI hardware, and dramatically increasing server utilization. Prodigy delivers unprecedented data center performance, power, and economics, reducing CAPEX and OPEX significantly.
6 Comments on Tachyum Submits Bid for 20-Exaflop Supercomputer to U.S. Department of Energy Advanced Computing Ecosystems
"This rock here has 100x more performance than any x86 CPU, but only on paper"
Tachyum's claims are relevant when compared to previous-generation and current-generation CPUs and GPUs (7nm or larger). In comparison to next-gen products (such as: 5nm x86 CPUs), Tachyum's claims would need to be readjusted and recalculated.
However, the claim that Tachyum Prodigy is at least 2 times more power efficient than any server with x86 CPUs and Nvidia GPUs is a sound claim, and simply follows from the fact that the x86 CPU cannot efficiently run a GPU-like AI workload and the GPU cannot efficiently run an x86-like general-purpose workload. If you have studied the PDF articles/papers/specs released by Tachyum, it is fairly obvious that the power-efficiency claim will very likely turn out to be true.
That said, Tachyum Prodigy isn't a gaming processor - if you are expecting Tachyum Prodigy to run PC games efficiently, it is obvious (again: based on the available Prodigy CPU documentation) that it won't be able to outperform AMD/Nvidia/Intel GPUs when running PC games.