Impact of Chiplets, Heterogeneous Integration and Modularity on AI and HPC systems
This panel explores the transformative impact of chiplets and heterogeneous integration on AI and HPC, addressing the escalating demands for memory and computing power. Chiplets, discrete components with specialized functions, are reshaping system architectures by enabling large-scale integration and modular designs. This approach optimizes performance, energy efficiency, and scalability in AI applications, from neural networks to data-intensive HPC workloads. Discussions will delve into the benefits of modular architecture, such as enhanced flexibility, easier upgrades, and accelerated innovation cycles. Key topics include design strategies for integrating diverse chiplet functionalities, interoperability standards, and ecosystem development. Attendees will gain insights into how these advancements are redefining the landscape of AI and HPC, fostering collaborative efforts to meet evolving computational challenges effectively.
Moderator:
John Shalf, Berkeley Laboratory
Panelists:
- Ramin Farjadrad (Ceo) - Eliyan
- William Chen (Senior Technical Advisor) - Ase Group
- Prith Banerjee (Chief Technology Officer) - Ansys, Inc.
- Dave Ditzel (Founder And CTO) - Esperanto Technologies
- Francesc Guim - Openchip & Software Technologies
Related Videos
- Revolutionizing System Design: Impact of Chiplets and Heterogeneous Integration on AI Server
- Podcast: AI, Chiplets, and the Future of Semiconductors
- Embedded World, Nuremberg: Arm’s Suraj Gajendra on AI, Chiplets, and the Future of Automotive Compute
- TSMC 3Dblox™: Unleashing 3D IC Innovations for the Next Generation of AI, HPC, Mobile and IoT
Latest Videos
- Revolutionizing SoC Design: The Shift to Chiplet-Based Architectures
- Embedded World, Nuremberg: Arm’s Suraj Gajendra on AI, Chiplets, and the Future of Automotive Compute
- Keysight Expands Chiplet Interconnect Support with UCIe 2.0 & BoW
- Revolutionizing AI & Chiplets: Baya Systems CEO on $36M Series B, UALink, Future of Data Movement
- Machine Learning Applications in EDA for Chiplet Reliability