Breaking Down Die Size Barriers with Chiplets
Why are hyperscalers moving towards chiplet ecosystems?
Anu Ramamurthy, Associate Research Fellow from Microchip Technology, and Open Chiplet Economy Sub-Project Co-Lead for Open Compute Project, explains:
- Large die sizes are reaching their physical limits, requiring division into smaller chiplets
- Chiplet architecture helps maintain high yields and lower costs while achieving needed computational power
- OCP enables multiple vendors to participate in the chiplet marketplace, expanding beyond proprietary solutions
Related Chiplet
- Optical Chiplet
- Reverse Gearbox Chiplet
- FlexGen Multi-Die Smart Network-on-Chip (NoC) IP
- Ncore Multi-Die Interconnect IP
- Integrated voltage regulator (IVR) chiplet
Related Videos
- Unleashing Chiplet potential with BoW - "The Knot That Ties Chiplets Together"
- Getting Moore with Less: How Chiplets and Open Interconnect Accelerate Cloud-Optimized AI Silicon
- Chiplets tackle semiconductor barriers
- Rapid Innovation with Chiplets and FPGAs
Latest Videos
- The European Chip Ecosystem
- Scaling up AI Compute, Memory and Connectivity
- CURIE: A Cycle-Accurate and Synthesizable UCIe Interface for the Agile Design and Co-Simulation of Chiplets
- Federated Chiplet Design and Simulation
- Custom HBM as a System: Co-designing the Base Die, Die to Die Power, and Package Thermals for AI Accelerators