UltraSoC analytics IP selected by Esperanto Technologies for RISC-V many-core parallel processing in AI and ML applications

May7, 2018 -- UltraSoC today announced the company’s embedded analytics IP has been selected by Esperanto Technologies for the development of massively parallel and many-core RISC-V SoCs. Esperanto is now integrating UltraSoC’s embedded analytics and debug technology into Esperanto’s high-performance and energy-efficient ‘AI Supercomputer on a Chip’ that employs thousands of 64-bit RISC-V cores, to serve advanced applications in artificial intelligence (AI) and machine learning (ML).

“Esperanto’s goal is to make RISC-V the architecture of choice for the most demanding AI and machine learning applications,” said Dave Ditzel, Founder and CEO of Esperanto. “UltraSoC’s IP will help our users see complex interactions between processors so they can understand what is going on and how to optimize performance better. Developing and debugging software utilizing thousands of RISC-V cores will be easier with the advanced analytics that UltraSoC’s IP will provide.”

Rupert Baines, CEO of UltraSoC added: “Esperanto is at the leading edge of technology putting thousands of RISC-V cores on a single chip. The track record of its founding group, and its backing from industry leaders like Western Digital, tells you that they are poised to make a big impact with their next chip design. We’re very proud to have been selected to help Esperanto achieve its vision for computing architectures of the future.”

Founded by processor industry veteran Dave Ditzel, Esperanto brings together a team of experienced processor and software engineers including some of the top computer scientists, architects and processor experts in the world. As a founding member of the RISC-V Foundation, Esperanto is committed to delivering the best in processing capabilities, and making full use of RISC-V’s power and performance improvements without the baggage of legacy architectures.

UltraSoC has taken a leading role within the RISC-V Foundation and provides the industry’s first – and still only – commercial RISC-V debug solution. Esperanto and its customers will benefit from this experience and expertise, with the ability to gain valuable insights and analysis of the operation of an entire SoC and system, not only during chip development, but also throughout the process of system integration and into the in-life phase of a product’s deployment.

Handling applications at the forefront of AI and ML technologies requires the application of thousands of 64-bit RISC-V cores in a single chip; Esperanto’s SoCs will deliver Teraflops of scalable parallel processing performance to developers using them. The company’s high-performance ET-Maxion™ cores are designed to deliver the best single thread RISC-V performance. The ET-Minion™ cores are designed for energy efficiency and high floating point throughput, including vector acceleration.

In support of systems based on these RISC-V CPUs, Esperanto will leverage advanced UltraSoC features such as transaction-level bus and status monitoring and static instrumentation to monitor program and execution flow. Such features are part of a broader UltraSoC product offering that supports all industry standard processor architectures, and is particularly powerful in heterogeneous multicore and many-core designs. More broadly, UltraSoC’s IP supports companies in improving design efficiency and giving designers system-wide analysis capabilities, and ultimately helps to improve and maintain power, performance and reliability of customers’ end products.

About Esperanto Technologies

Esperanto Technologies develops high-performance, energy-efficient computing solutions based on the open standard RISC-V ISA. Esperanto is headquartered in Mountain View, California with engineering sites in the United States, European Union, and Eastern Europe. Esperanto has brought together a seasoned team of experienced processor and software engineers with the goal of making RISC-V the architecture of choice for compute-intensive applications such as Machine Learning. For more information, please visit https://esperanto.ai.