First Generative AI tool for power integrity sign off

First Generative AI tool for power integrity sign off

Technology News |
By Nick Flaherty

Cadence Design Systems has developed the industry’s first generative AI-driven technology to predict, root cause and resolve IR drop issues early in the design process.

The Voltus InsightAI technology uses machine learning-generated power grid models to enable fast IR analysis of power integrity. This eliminates the need to overdesign the power grid in system on chip and chiplet designs due to early IR drop prediction and prevention capabilities.

The Voltus InsightAI tool integrates with the Cadence digital full flow and system design and analysis platform for efficient EM-IR-driven design closure, fixing up to 95% of violations prior to signoff. The tool has been used by ARM on a 3nm design as well as by MediaTek and the ASIC group at Cisco.

Power integrity is a major design challenge at advanced nodes, with designers regularly facing a significant number of EM-IR violations at signoff, making it imperative to address this challenge early in the design phase.

One of the major bottlenecks of in-design EM-IR analysis is that it is computationally very expensive due to the size and coupled nature of the power network. Using generative AI helps to overcome this bottleneck with fast incremental IR analysis.

The tool uses proprietary neural networks to build models of the power grid and can perform extremely fast incremental IR analysis to provide instant feedback on the impact of design changes. This can determine the root cause of IR drop problems and can quickly identify aggressors, victims and resistance bottlenecks. It uses electrical, spatial and timing factors for predicting IR drop issues during design.

Decision-tree methods are used for timing and design rule check (DRC)-aware fixes of IR drop, using multiple methods like placement, grid reinforcement, routing and engineering change orders (ECOs). The tool selects precise fixing methods based on the root cause of the problem, driving better utilization and improved PPA.

“As we move to more advanced nodes, EM-IR is quickly becoming one of the most pressing challenges, requiring novel and innovative approaches to address customer needs. With Voltus InsightAI, Cadence has pioneered applying generative AI technology to EM-IR, focusing not just on signoff, but early detection and prevention of EM-IR violations as well,” said Ben Gu, Corporate Vice President of R&D, Multiphysics Systems Analysis Business Unit at Cadence.

“With this capability, designers don’t need to over-design the power grid, thereby enabling far better PPA. Customers are seeing impressive results as they can leverage this breakthrough technology to fix up to 95% of violations prior to signoff and achieve more than 2X productivity improvement in EM-IR closure.”

The generative AI-driven Voltus InsightAI is part of the Cadence Intelligent System Design strategy.

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News


Linked Articles