Thermal Guardbanding


Stephen Crosher, CEO of Moortec, looks at the causes of thermal runaway in racks of servers and explains why accurate temperature measurement in AI and advanced-node chips is more critical, and what impact this has on performance when temperatures begin approaching acceptable limits. » read more

Analog: Avoid Or Embrace?


We live in an analog world, but digital processing has proven quicker, cheaper and easier. Moving digital data around is only possible while the physics of wires can be safely abstracted away enough to provide reliable communications. As soon as a signal passes off-chip, the analog domain reasserts control for modern systems. Each of those transitions requires a data converter. The usage ... » read more

Priorities Shift In IC Design


The rush to the edge and new applications around AI are causing a shift in design strategies toward the highest performance per watt, rather than the highest performance or lowest power. This may sound like hair-splitting, but it has set a scramble in motion around how to process more data more quickly without just relying on faster processors and accelerators. Several factors are driving th... » read more

Interdependencies Complicate IC Power Grid Design


Creating the right power grid is a growing problem in leading-edge chips. IP and SoC providers are spending a considerable amount of time defining the architecture of logic libraries in order to enable different power grids to satisfy the needs of different market segments. The end of Dennard scaling is one of the reasons for the increased focus. With the move to smaller nodes, the amount of... » read more

Defining And Improving AI Performance


Many companies are developing AI chips, both for training and for inference. Although getting the required functionality is important, many solutions will be judged by their performance characteristics. Performance can be measured in different ways, such as number of inferences per second or per watt. These figures are dependent on a lot of factors, not just the hardware architecture. The optim... » read more

Week In Review: Design, Low Power


Cadence signed a deal to buy National Instruments’ AWR business unit for about $160 million in cash, a move that Cadence describes as a way to broaden its market into intelligent system design. AWR’s strength is high-frequency RF design automation tools, particularly in the millimeter wave and microwave spectrums, which are critical for radar and 5G. It also has technology for III-V materia... » read more

Week In Review: IoT, Security, Auto


Internet of Things SiFive is bringing RISC-V to IoT makers and university developers through the RISC-V-based SiFive Learn Initiative, an open-source learning package that can be used to create a low-cost RISC-V hardware compatible with AWS IoT Core. The development platform SiFive Learn Inventor has a software package and education enablement course. It includes: The programmable SiFive Lear... » read more

Verification Pilgrims Show A Historical Case For DFT


The Mayflower Steps, where the Pilgrims are believed to have embarked on their journey to America, are located in the beautiful Barbican area of Plymouth, a small town in the southwest of England. As the lone American working for Moortec, a British company based in Plymouth, I stood and stared at them this past September. Separated by a few yards of distance but 399 years of history I found my... » read more

The Challenge Of Defining Worst Case


Worst case conditions within a chip are impossible to define. But what happens if you missed a corner case that causes chip failure? As the semiconductor market becomes increasingly competitive — startups and systems companies are now competing with established chipmakers — no one can afford to consider theoretical worst cases. Instead, they must intelligently prune the space to make sur... » read more

Finding Hotspots In AI Chips


Things are getting far more complicated as we move down to 7nm & 5nm but the tolerances of some of the physical effects that we have been measuring in the past are much tighter than they were at the older nodes. How do we track all that? What we see is that as we descend through the advanced nodes, say from 16nm down to 12nm, 7nm and more recently 5nm, we see that gate density starts to ... » read more

← Older posts