Universal Chiplet Interconnect Express: An Open Industry Standard for Memory and Storage Applications
By Debendra Das Sharma (CXL Board Technical Task Force, Intel), Thomas Coughlin (Coughlin Associates)
Abstract
Continuing advances in computing will require chiplets connected by the Universal Chiplet Interconnect Express (UCIe) to create cost-effective high-performance customized applications and new memory hierarchies. The optical UCIe can expand this functionality to the pod and rack level.
There are many drivers for on-package integration of chiplets. Overcoming reticle limitations to deliver performance, functionality, and yield tradeoffs with larger dies is a primary reason why today volume CPUs and GPUs use chiplet-based solutions with proprietary scale-up interconnects. Chiplet approaches reduce the compromises required with monolithic designs. Lowering the overall portfolio cost with a time-to-market advantage is another motivation for deploying chiplets.
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
- On-Package Memory with Universal Chiplet Interconnect Express (UCIe): A Low Power, High Bandwidth, Low Latency and Low Cost Approach
- High-Bandwidth Chiplet Interconnects for Advanced Packaging Technologies in AI/ML Applications: Challenges and Solutions
- FoldedHexaTorus: An Inter-Chiplet Interconnect Topology for Chiplet-based Systems using Organic and Glass Substrates
- Fast and Accurate Jitter Modeling for Statistical BER Analysis for Chiplet Interconnect and Beyond
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