A recent technical review published by the Semiconductor Engineering platform addressed a range of prominent engineering topics and trends in the semiconductor and advanced systems sector, shedding light on technical transformations in modern design and architectural layouts.
Neeraj Paliwal from Synopsys pointed out that artificial intelligence workloads are driving radical changes in design and optimization considerations, moving beyond traditional server levels to target the rack level. Designs now start from the target workload and work backward to achieve higher efficiency in AI task performance per watt and dollar.
In the same context, Ravi Phoura from Cadence reviewed the mechanism of evolving the AMBA CXS interface to transform wide-band transport links into a shared, scalable transport medium supporting multi-protocol and multi-node flows, while maintaining flow ordering, routing, and packet reconstruction.
What stands out in the design and advanced systems sector?
- Digital Twins: Siemens experts discussed the adoption of digital twins in the semiconductor industry, ranging from chip design and verification to fab construction and automation.
- Edge AI Development: Christopher Sidel from Arm reviewed how to build an edge AI development loop based on real data feeds to fuel subsequent iterations.
- Space Power Systems: Nancy Friedrich from Keysight explored power generation and distribution requirements for satellites and laboratory simulation techniques to evaluate performance.
Global Workforce Challenges in the Chip Sector
On the business and human resources front, Jan Bart Smits from Stanton Chase warned in an article for SEMI of the disparity in specialized semiconductor talent shortages across North America, Europe, the Middle East, Africa, and Southeast Asia, emphasizing that government support and funding alone are not enough to buy the expertise needed for desired expansion operations.