RISC-V Summit 2022 has ended
December 13-14, 2022 | San Jose, CA + Virtual
Learn More & Register Now
Virtual Event Access
Back To Schedule
Wednesday, December 14 • 11:40am - 12:00pm
HW-SW Co-development for RISC-V Based Secure ML Systems with the Sparrow Project and Renode - Michael Gielda, Antmicro & Kai Yick, Google

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
Sparrow is a project aiming to build a low-power secure embedded platform for Ambient ML applications. The target platform leverages RISC-V and OpenTitan, and uses the open source Renode simulation framework for pre-silicon co-development of the complete hardware and software solution. Sparrow includes a complex multi-core hardware setup with RISC-V vector instruction support as well as an advanced Rust-based software stack built on top of seL4. Various Renode capabilities have been developed to support the pre-silicon development of Sparrow, including vector instructions support, seL4 OS-aware debugging, ability to use custom, alternative CPU core models as well as various tracing and performance tracking capabilities. The open source simulation-driven development flow used by Sparrow is used to explore tradeoffs and prototype the complete solution including ML software in a continuous integration environment, and can serve as a template for other ML-focused RISC-V silicon projects of this kind.

avatar for Michael Gielda

Michael Gielda

Vice President, Business Development & Co-Founder, Antmicro
Michael Gielda is Co-Founder at Antmicro, Chair of Outreach for CHIPS Alliance and Chair of Marketing for the Zephyr Project. He is involved in many open source software and hardware projects related to software-driven tools and methodologies, AI, FPGA & ASIC development.

Kai Yick

Hardware Manager, Google
Kai Yick is Hardware Manager at Google, leading the AmbiML project aiming to develop an open source hardware and software platform based on RISC-V for secure and private edge ML computing.

Wednesday December 14, 2022 11:40am - 12:00pm PST
Grand Ballroom 220 A
  Technical, Software