中文 English

Blog Review: July 6

Shifting data center paradigms; optimal designs with MDAO; DDR design rules; battery management.

popularity

Synopsys’ Mike Gianfagna looks at how the data center paradigm has shifted in the last ten years with an exponential increase in the amount of data demanding new approaches to storage that rely on distributed networks.

Cadence’s Frank Schirrmeister explains multidisciplinary design analysis and optimization, or MDAO, and how it is being combined with machine learning models to enhance classic heuristics and guide a design implementation towards the optimal design.

Siemens EDA’s Tim Rauscher points to the importance of correctly using design rules in DDR and other high-speed circuits.

Ansys’ Mazen El Hout finds out how simulation can help design a battery management system that meets efficiency, safety, and reliability requirements.

In a blog for Arm, Vivotek’s Joseph Chen suggests retailers can take advantage of surveillance video already being captured to analyze data and improve product placement, and other ways to connect and make more use of security camera footage.

NXP’s Arjan Leeuwenburgh checks out how plant microbial fuel cells could be used to power small electronics, demonstrated by a collaboration that created a network of sensors that could run off of bacteria in soil to monitor environmental qualities such as water level.

In a blog for SEMI, GE Research’s Radislav Potyrailo, Bosch’s Ryotaro Sakauchi, TDK’s Sreeni Rao, and Renesas’ Christian Meye explore the growing market for pervasive gas monitoring, what users of gas sensors are looking for, and the common pain points that sensor manufacturers need to address.

For a change from reading, watch some of the latest videos:

Co-Packaged Optics In The Data Center can help minimize energy and improve performance as Ethernet speeds increase.

Deep Learning In Industrial Inspection is helpful for finding defects.

Considerations and fixes for hardware attacks are needed for Protecting ICs Against Specific Threats.

Transforming AI Models For Accelerator Chips shows why floating point data needs to be converted into integer point data.

Reducing data bubbles in domain-specific designs for Zero Dark Silicon.

Improve accuracy of what gets printed on a photomask, while accelerating the process, by Using GPUs In Semiconductor Manufacturing.

Speeding Up AI Algorithms by addressing inferencing challenges at the edge.

How to improve video performance with minimal impact on resolution with Better Video Compression.

No hardware is bulletproof, but understanding potential weaknesses can help, which is what Common Weakness Enumeration sets out to do.

Moving increasing amounts of data is inefficient, which is why designers are Moving Intelligence To The Edge.



Leave a Reply


(Note: This name will be displayed publicly)