Toward Open-Source Chiplets for HPC and AI: Occamy and Beyond
By Paul Scheffler 1, Thomas Benz 1, Tim Fischer 1, Lorenzo Leone 1, Sina Arjmandpour 1, Luca Benini 1,2
1 Integrated Systems Laboratory, ETH Zurich, Switzerland
2 Department of Electrical, Electronic, and Information Engineering, University of Bologna, Italy
Abstract
We present a roadmap for open-source chiplet-based RISC-V systems targeting high-performance computing and artificial intelligence, aiming to close the performance gap to proprietary designs. Starting with Occamy, the first open, silicon-proven dual-chiplet RISC-V manycore in 12nm FinFET, we scale to Ramora, a mesh-NoC-based dual-chiplet system, and to Ogopogo, a 7nm quad-chiplet concept architecture achieving state-of-the-art compute density. Finally, we explore possible avenues to extend openness beyond logic-core RTL into simulation, EDA, PDKs, and off-die PHYs.
Index Terms—Chiplets, RISC-V, HPC, NoC, AI, Machine Learning
To read the full article, click here
Related Chiplet
- DPIQ Tx PICs
- IMDD Tx PICs
- Near-Packaged Optics (NPO) Chiplet Solution
- High Performance Droplet
- Interconnect Chiplet
Related Technical Papers
- Occamy: A 432-Core 28.1 DP-GFLOP/s/W 83% FPU Utilization Dual-Chiplet, Dual-HBM2E RISC-V-based Accelerator for Stencil and Sparse Linear Algebra Computations with 8-to-64-bit Floating-Point Support in 12nm FinFET
- Advanced Packaging and Chiplets Can Be for Everyone
- Fast and Accurate Jitter Modeling for Statistical BER Analysis for Chiplet Interconnect and Beyond
- PICNIC: Silicon Photonic Interconnected Chiplets with Computational Network and In-memory Computing for LLM Inference Acceleration
Latest Technical Papers
- Link Quality Aware Pathfinding for Chiplet Interconnects
- Effects of Poor Workload Partitioning on System Performance for Chiplet-Based Systems
- Mozart: Modularized and Efficient MoE Training on 3.5D Wafer-Scale Chiplet Architectures
- Network Design for Wafer-Scale Systems with Wafer-on-Wafer Hybrid Bonding
- CarbonPATH: Carbon-aware pathfinding and architecture optimization for chiplet-based AI systems