How do UCIe and BoW interconnects support generative AI on chiplets?
By Jeff Shepard, Microcontroller Tips (February 28, 2024)
The bunch-of-wires (BoW) and Universal Chiplet Interconnect Express (UCIe) standards provide designers with tradeoffs in terms of throughput, interconnect density, delay, and bump pitch. This FAQ compares the performance of BoW and UCIe and looks at how optical interconnects may provide a path to even higher performance interconnects in chiplets.
To realize optimal performance for generative artificial intelligence (AI), machine learning (ML), and other high-performance computing (HPC) applications, designers are turning to chipsets that can combine AI accelerators, GPUs, CPU, memory, and networking in a single package. Interconnecting heterogeneous devices in a chiplet can be especially challenging.
To read the full article, click here
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 News
- How does UCIe on chiplets enable optical interconnects in data centers?
- How can in-package optical interconnects enhance chiplet generative AI performance?
- Alphawave Semi to Reveal Ecosystem and Key Architectures Unlocking Generative AI Potential at EE Times' "Chiplets: Building the Future of SoCs" Seminar
- Chiplets Make Progress Using Interconnects As Glue
Latest News
- ACM Research Announces Shipment of Its First 310 × 310 mm Ultra ECP ap-p Panel-Level Horizontal Electroplating Production Tool
- Entegris Enters Silicon Carbide and Diamond-Enabled Thermal Management Materials Market for AI Advanced Packaging
- Fraunhofer IPMS Showcases New Technologies for the Future of Semiconductor Industry
- Marvell and Lumentum to Combine, Creating a Category Leader in AI Connectivity
- Silicon Austria Labs Adopts PulseForge Photonic Debonding Technology to Accelerate Advanced Packaging and Heterogeneous Integration Research