Blog Review: Feb. 21

Siemens’ John McMillan digs into physical verification maturity for high-density advanced packaging (HDAP) designs and major differences in the LVS verification flow compared to the well-established process for SoCs.

Synopsys’ Varun Shah identifies why a cloud adoption framework is key to getting the most out of deploying EDA tools in the cloud, including by ensuring that different types of necessary compute are accessible for all stages of the design cycle.

Cadence’s Reela Samuel suggests that a chiplet-based approach will provide improved performance and reduced complexity for the automotive sector, enabling OEMs to construct a robust yet flexible electronic architecture.

Keysight’s Emily Yan finds that today’s chip design landscape is facing challenges reminiscent of those encountered by the Large Hadron Collider in managing data volume, version control, and global collaboration.

Ansys’ Raha Vafaei explains why the finite-difference time-domain (FDTD) method, an algorithmic approach to solving Maxwell’s equations, is key for modeling nanophotonic devices, processes, and materials.

Arm’s Ed Player explains the different components of the Common Microcontroller Software Interface Standard (CMSIS) to help identify which are useful for particular Arm-based microcontroller projects.

SEMI’s Mark da Silva, Nishita Rao and Karim Somani check out the state of digital twins in semiconductor manufacturing and challenges such as the need for standardization and communication between different digital twins.

Plus, check out the blogs featured in the latest Low Power-High Performance newsletter:

Rambus’ Lou Ternullo looks at why performance demands of generative AI and other advanced workloads will require new architectural solutions enabled by CXL.

Ansys’ Raha Vafaei shines a light on how the evolution of photonics engineering will encompass novel materials and cutting-edge techniques.

Siemens’ Keith Felton explains why embracing emerging approaches is essential for crafting IC packages that address the evolving demands of sustainability, technology, and consumer preferences.

Cadence’s Mark Seymour lays out how CFD simulation software can predict time-dependent aspects and various failure scenarios for data center managers.

Arm’s Adnan Al-Sinan and Gian Marco Iodice point out that LLMs already run well on small devices, and that will only improve as models become smaller and more sophisticated.

Keysight’s Roberto Piacentini Filho shows how a modular approach can improve yield, reduce cost, and improve PPA/C.

Quadric’s Steve Roddy finds that smart local memory in an AI/ML subsystem solves SoC bottlenecks.

Synopsys’ Ian Land, Kenneth Larsen, and Rob Aitken detail why the traditional approach using monolithic system-on-chips (SoCs) falls short when addressing the complex needs of modern systems.