Chiplet Interfaces Embrace Failures
Why lane swapping is essential to meet assembly yield.
By Anne Meixner, Semi Engineering | August 12th, 2025
Redundancy in chiplet interfaces is now a prerequisite for achieving sufficient yield in high-performance computing devices, which today are packed with tens of thousands of interconnects. And as the number and density of those interconnects increases, the prospects for yield only worsen.
For more than two decades, high-speed I/O interfaces have included reliability strategies to manage in-field system board failures. For example, the PCI Express 2.0 standard introduced in 2007 included 16 lanes for transactions. But if there was a lane failure then only 8 of those lanes were utilized, cutting the transaction rate in half. Commonly referred to as graceful degradation, this remains a strategy for computing chiplet interfaces such as UCIe.
To read the full article, click here
Related Chiplet
- High Performance Droplet
- Interconnect Chiplet
- 12nm EURYTION RFK1 - UCIe SP based Ka-Ku Band Chiplet Transceiver
- Bridglets
- Automotive AI Accelerator
Related News
- Renesas Unveils Industry’s First Automotive Multi-Domain SoC Built with 3-nm Process Technology with Chiplet Extensions
- Breaking Boundaries: Chiplet Interconnects, SDVs & electronica 2024
- Cadence Unveils Arm-Based System Chiplet
- Eliyan Ports Industry’s Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
Latest News
- Cadence Launches Partner Ecosystem to Accelerate Chiplet Time to Market
- Ambiq and Bravechip Cut Smart Ring Costs by 85% with New Edge AI Chiplet
- TI accelerates the shift toward autonomous vehicles with expanded automotive portfolio
- Where co-packaged optics (CPO) technology stands in 2026
- Qualcomm Completes Acquisition of Alphawave Semi