Paving the Road to Datacenter-Scale RISC-V
By Martin Dixon, Engineering Director, Google
The computing industry has shifted to an era of specialization, and Google’s heterogeneous datacenters—powered by x86, Arm, GPUs, and TPUs—are at the forefront of this trend. This keynote delivers our playbook for this new reality.
We will share insights from our legacy single architecture to our multi-architecture journey, detailing how AI was used to automate the complex porting of our software stack from x86 to other ISAs like Arm and RISC-V. We will discuss how we can build on this "paved road" of tooling and expertise to further accelerate our joint RISC-V journey. We conclude with a call to action, to move faster, together, to build on our model with collaborators like the RISE Project to accelerate the pace of innovation and adoption for RISC-V.
Related Semiconductor IP
- 64-bit, RISC-V, ultra-high performance processors
- 64-bit, RISC-V, performance and data computation processors
- 32-bit, RISC-V, deeply embedded processors
- RISC-V Display Connectivity Subsystem (DCS)
- RISC-V IOPMP IP
Related Videos
- Hardware Innovation in the World's First RISC-V 50 TOPS AI Compute for Mass Production Development
- The Next Computing Megatrends are Enabled by RISC-V
- The Rise of RISC-V
- Bringing High-Performance RISC-V Platforms to Life
Latest Videos
- Software to silicon with RISC-V for Physical AI
- Breaking the Memory Wall: How New Memory Architectures are Reshaping AI Inference
- Functional Safety & Security Aspects of CAN XL
- Powering the AI Supercycle: Design for AI and AI for Design - Anirudh Devgan
- Scaling AI from Edge to Data Center with SiFive RISC-V Vectors