Chiplet Architecture for AI Will Create New Demands for Assembly
May 28, 2024 | Nolan Johnson, SMT007 MagazineEstimated reading time: 2 minutes
![](https://iconnect007.com/application/files/5517/1644/2377/Arvind_Kumar_300.jpg)
As we examine the entire AI ecosystem more closely, it becomes clear that AI algorithms are intensely hungry for compute power. This demand is accelerating beyond the customary rate predicted by Moore’s Law, just as traditional semiconductor fabrication methods are failing to maintain Moore’s Law. It’s a real dilemma.
Those watching AI say that advanced packaging techniques, which have been in R&D development for some time, see AI as their killer app. AI is needed to propel these cutting-edge packages into the mainstream.
At a 2022 symposium on advanced packaging in Washington, D.C., I met Dale McHerron, a researcher on AI compute hardware. As we discussed IBM’s work in this area, Dale introduced me to Arvind Kumar, a principal research scientist and manager in AI hardware and chiplet architectures.
I reached out to Arvind to discuss his keynote presentation at the recent IMAPS conference where he discussed the AI hardware ecosystem and role of advanced packaging. Those in the assembly services industry know that any new packages will require accurate and reliable placement on the EMS manufacturing floor. Arvind shares his perspective and some predictions based on his research. It is also clear there is still much coordination and communication needed to make this work.
Nolan Johnson: What is chiplet architecture and why does it matter? How is advanced packaging moving forward?
Arvind Kumar: Chiplet architectures, which allow the partitioning of complex designs into tightly co-packaged sub-elements, are influencing the way we think about packaging. We would like to put more chips into a single package and have them talk to each other with high bandwidth, low latency, and low-energy interconnects. That goal is driving emerging packaging technologies to higher interconnect densities, more routing layers, and larger body sizes.
Johnson: Ever since semiconductors were developed, we’ve used the thought pattern of making bigger monolithic chips. Why the change?
Kumar: For a long time, the fundamental idea was that we could get more performance out of larger dies at the most advanced technology node. Fabricating all parts of a chip at the most advanced node is getting very expensive and has major yield challenges, so this drives us toward smaller die sizes. Additionally, we can partition the chip such that some parts that don’t scale well can remain in an older technology node. That's a very natural fit for chiplet architecture.
To continue reading this interview that originally appeared in the May 2024 issue of SMT007 Magazine, click here.
Suggested Items
IPC WorksAsia – Quality & High Reliability Conference Thailand
06/26/2024 | IPCQuality & High Reliability are critical when electronics assemblies are working in harsh environments where there is high temperature difference, vibration; and components get smaller and demand for high reliability products increases.
Space Engine Systems Announces Partnership With ASEI (USA) to Develop Hypersonic Solutions for Missiles
06/26/2024 | BUSINESS WIRESpace Engine Systems (SES), following on its success in securing a place in UK MoD’s Hypersonic Technology and Capability Development Framework (HTCDF), is pleased to announce a partnership with Applied Systems Engineering Inc (ASEI) of the US.
Siemens, Intel Foundry Collaborate to Deliver New Tools Certifications and EMIB/3D-IC Innovation
06/25/2024 | Siemens Digital IndustriesSiemens Digital Industries Software announced that its ongoing collaboration with Intel Foundry has established a new Electronic Design Automation (EDA) product certification, as well as an embedded multi-die interconnect bridge (EMIB) enablement breakthrough that can help mutual customers leverage the performance, space and power efficiency advantages of 3D-IC (3D integrated circuit) designs much more quickly.
Cadence, Intel Foundry Collaborate on Systems Foundry Enablement for the AI Era
06/25/2024 | Cadence Design Systems, Inc.Cadence Design Systems, Inc. announced the achievement of several key milestones in its ongoing strategic partnership with Intel Foundry.
Yamaha Robotics Expands Sales Team in Germany
06/25/2024 | Yamaha Robotics SMT SectionYamaha Robotics SMT Section has appointed Johannes Vorwerk as Sales Manager for Germany, responsible for the complete 1 STOP SMART SOLUTION for high-speed surface-mount assembly.