RISC-V Startup Esperanto Technologies Samples First AI Silicon
In Forbes, Enterprise Tech, by Karl Freund: "RISC-V Startup Esperanto Technologies Samples First AI Silicon."
Karl wrote:
- "I had the chance to watch a demo of the platform and came away quite impressed with the performance and power efficiency of the RISC-V based platform.
- I was also pleased to see that the Esperanto device is not a one-trick pony, as the team demonstrated Resnet50, DLRM, and the Transformer network underlying BERT."
Esperanto Technologies’ Massively Parallel RISC-V AI Inferencing Solution Now in Initial Evaluations
Delivering Industry-Leading Energy Efficiency, Esperanto’s ML Inference Accelerator
Is Designed to be the Highest Performance Commercial RISC-V AI Chip
MOUNTAIN VIEW, Calif., April 20, 2022 – Esperanto Technologies™, the leading developer of high performance, energy-efficient artificial intelligence (AI) inference accelerators based on the RISC-V instruction set, today announced that initial evaluations for its ET-SoC-1 AI inference accelerator are underway with lead customers. Additional slots are available to qualified customers who have an interest in AI inferencing accelerators for datacenter applications.
To inquire about the evaluation program, please visit esperanto.ai/technology/#eap.
“Our data science team was very impressed with the initial evaluation of Esperanto’s AI acceleration solution,” said Dr. Patrick Bangert, vice president of Artificial Intelligence at Samsung SDS. “It was fast, performant and overall easy to use. In addition, the SoC demonstrated near-linear performance scaling across different configurations of AI compute clusters. This is a capability that is quite unique, and one we have yet to see consistently delivered by established companies offering alternative solutions to Esperanto.”
Esperanto’s evaluation program enables users to obtain performance data from running a variety of off-the-shelf AI models including recommendation, transformer and visual networks on the ET-SoC-1 AI Inference Accelerator. Users can set options including model and dataset selection, data type, batch size and compute configuration of up to 32 clusters containing over 1,000 RISC-V cores with ML-optimized tensor units. Customers can run many inference jobs, with the results provided in detailed histogram reports as well as fine-grain visibility into silicon performance.
“Esperanto has made very impressive progress and is now providing customers evaluation access to their RISC-V hardware and software running off-the-shelf AI models with strong performance and efficiency. This really shows the company’s confidence in their first multi-core solution,” said Karl Freund, founder and principal analyst at Cambrian-AI Research. “In addition, because Esperanto’s chip is RISC-V-based, it has the programming tools and software stack to more easily adapt to new AI workloads, alongside non-AI workloads, all running on the same silicon. This step forward is another very strong indicator of the bright future of RISC-V.”
“Esperanto has achieved an industry first by demonstrating its massively parallel RISC-V silicon running a variety of real-world AI workloads,” said Richard Wawrzyniak, principal analyst at Semico Research. “It was exciting for me to see the company put the chip through its paces across a variety of scenarios including different models, data types, batch sizes and compute cluster combinations – all showing competitive results. This is another positive step forward for the RISC-V industry in the AI space as this new market continues to grow even faster than we had previously forecasted.”
“Harnessing the power of over 1,000 RISC-V processors is a major accomplishment, and we are very pleased with the results which validate our initial projections of performance and efficiency,” said Art Swift, president and CEO of Esperanto Technologies. “We look forward to extending access to a broader range of qualified companies, as we accelerate our RISC-V roadmap efforts with a growing number of strategic partners for applications spanning from Cloud to Edge.”
Esperanto Technologies is the AI RISC-V leader, offering massively parallel 64-bit RISC-V-based tensor compute cores currently delivered in the form of a single chip with 1088 ET-Minion compute cores and a shared high performance memory architecture. Designed to meet the performance, power and total cost of ownership (TCO) requirements of large-scale datacenter customers, Esperanto’s inference chip is a general purpose, parallel processing solution that can accelerate many parallelizable workloads. It is designed to run any machine learning (ML) workload well, and to excel at ML recommendation models, one of the most important types of AI workloads in many large datacenters.
About Esperanto Technologies:
Esperanto Technologies develops massively parallel, high-performance, energy-efficient computing solutions for Artificial Intelligence / Machine Learning based on the open standard RISC-V instruction set architecture. Esperanto is headquartered in Mountain View, California with additional engineering sites in Portland, Oregon; Austin, Texas; Barcelona, Spain; and Belgrade, Serbia. For more information, please visit https://www.esperanto.ai/
RISC-V AI Chips Are Joining GPU Race for AI Processing: CDO Trends
In CDO Trends, see Paul Mah's article on how "RISC-V AI Chips Are Joining GPU Race for AI Processing."
He writes:
"The future has RISC-V in it
One such startup, Esperanto Technologies, utilized a modified RISC-V design with 1,092 cores into a system-on-a-chip (SoC) half the size of the popular A100 GPU from Nvidia.
As reported on IEEE Spectrum, the team created their own vector instructions to complement RISC-V’s efficient 47 instructions (A typical Intel desktop processor has close to a thousand instructions) to support machine learning math such as matrix multiplication.
The ET-SoC-1 from Esperanto is envisioned to accelerate AI in power-constrained data centers through expansion boards that fit into a standard peripheral component interconnect express (PCIe) slot. According to the report, each board can deliver 800 trillion operations per second.
What sets Esperanto’s solution apart is how each board uses multiple low-power SoC chips instead of a giant SoC. According to the AI chip maker, each ET-SoC-1 chip consumes 20 watts when performing a recommender-system benchmark neural network, or less than one-tenth of what the A100 GPU draws.
This allowed the team to place six chips for over 6,000 cores on a single AI accelerator card and still stay at around 120 watts.
And according to a report on All About Circuits last year, Esperanto claims an ET-SoC-1 outperforms the Nvidia A100 in both relative performance and energy efficiency running the MLPerf Deep Learning Recommendation Model benchmark."
IEEE Spectrum: RISC-V AI Chips Will Be Everywhere
"Esperanto Techology’s chip heralds new era in open-source architecture; Intel set to cash in"
Article, by Samuel Moore in IEEE Spectrum: “RISC-V AI Chips Will Be Everywhere.”
- Dave Ditzel comments on the sea change in AI, along with Esperanto's approach and details of the ET-SoC-1 SoC.
Esperanto’s Dave Ditzel and Intel’s Bob Brennan discuss the RISC-V ecosystem
Esperanto’s Dave Ditzel and Intel’s Bob Brennan discuss the RISC-V ecosystem
by Karl Freund | Feb 9, 2022 | Video
Intel has announced it is joining RISC-V, and has created a $1B Innovation Fund to help jump-start the RISC-V ecosystem. Now, Intel competitors will be motivated to consider Intel Foundry Services as a Fab partner. Esperanto’s Dave Ditzel and Intel’s Bob Brennan join Karl Freund to discuss the implications.
Intel Creates $1B Innovation Fund to Grow RISC-V Market (And Attract New Foundry Customers)
Article, by Karl Freund in Forbes: “Intel Creates $1B Innovation Fund to Grow RISC-V Market (And Attract New Foundry Customers).”
- This announcement will have a long-lasting impact on the entire semiconductor industry.
- Intel must fabricate chips for its competitors to grow its foundry business, and is investing in the emerging RISC-V market as the catalyst.
- Esperanto Technologies is now an early Intel Foundry Services partner.
- Intel’s $1B IFS Innovation Fund will support early-stage startups and established companies building disruptive technologies for the foundry ecosystem. A large part of this will be to accelerate RISC-V.
- Intel also announced the IFS Accelerator.
Esperanto Technologies Partners with Intel to Advance Massively Parallel RISC-V-based AI Acceleration Solutions
MOUNTAIN VIEW, Calif., February 7, 2022 – Esperanto Technologies™, the leading developer of high-performance, energy-efficient artificial intelligence (AI) inference accelerators based on the RISC-V instruction set, today announced that the company is entering a strategic partnership with Intel to advance its massively parallel RISC-V-based AI acceleration solutions. As part of the partnership, Esperanto Technologies plans to use Intel Foundry Services (“IFS”) silicon and chiplet packaging technologies to advance its RISC-V-based technology and deliver its massively parallel AI acceleration silicon solutions spanning from cloud to edge.
“Intel Foundry Services is excited to add Esperanto’s massively-parallel AI accelerators to the IFS ecosystem,” said Bob Brennan, Vice President and General Manager of IFS Customer Solutions Engineering. “The combination of Esperanto’s low-power design techniques and IFS advanced silicon and packaging technologies should enable thousands of Esperanto’s RISC-V based processors to operate in a single package with remarkably low power.”
“Esperanto builds on top of the RISC-V ecosystem to deliver our leading AI acceleration solutions. Intel’s announcements of support and investment in the RISC-V Community are invaluable in helping to advance and proliferate RISC-V solutions from Esperanto and others, ultimately benefiting all end users,” said Art Swift, president and CEO at Esperanto Technologies.
“Esperanto plans to use Intel Foundry Services leading edge silicon and advanced packaging technologies with our next generation massively-parallel RISC-V based AI accelerators,” said David Ditzel, founder and executive chairman at Esperanto Technologies. “Intel Foundry Services’ unique packaging technologies will make chiplet solutions both practical and cost effective, enabling a range of acceleration products to span from cloud to edge.”
Esperanto Technologies is the AI RISC-V leader, offering massively-parallel 64-bit RISC-V-based Tensor compute cores currently delivered in the form of a single chip with 1088 ET-Minion compute cores and a high-performance memory system. Designed to meet the high-performance and low-power requirements of large-scale datacenter customers, Esperanto’s existing RISC-V-based inference chip is a general purpose, parallel processing solution that can accelerate many parallelizable workloads. It is designed to run any machine learning (ML) workload well, and to excel at ML recommendation models, one of the most important types of AI workloads in many large datacenters.
About Esperanto Technologies:
Esperanto Technologies develops massively parallel, high-performance, energy-efficient computing solutions for Artificial Intelligence / Machine Learning based on the open standard RISC-V instruction set architecture. Esperanto is headquartered in Mountain View, California with engineering sites in Portland, Oregon, Barcelona, Spain, and Belgrade, Serbia. For more information, please visit https://www.esperanto.ai/
