Power Challenges In ML Processors


The design of artificial intelligence (AI) chips or machine learning (ML) systems requires that designers and architects use every trick in the book and then learn some new ones if they are to be successful. Call it style, call it architecture, there are some designs that are just better than others. When it comes to power, there are plenty of ways that small changes can make large differences.... » read more

High-Performance Memory For AI And HPC


Frank Ferro, senior director of product management at Rambus, examines the current performance bottlenecks in high-performance computing, drilling down into power and performance for different memory options, and explains what are the best solutions for different applications and why. » read more

Enterprise-Class DRAM Reliability


Brett Murdock, product manager for memory interfaces at Synopsys, examines demand for DDR5 and DDR4 in both on-premise and cloud implementations, what features are available for which versions, how they affect performance and power, how ECC is implemented, and how the data moves throughout these systems. » read more

Thinking About AI Power In Parallel


Most AI chips being developed today run highly parallel series of multiply/accumulate (MAC) operations. More processors and accelerators equate to better performance. This is why it's not uncommon to see chipmakers stitching together multiple die that are larger than a single reticle. It's also one of the reasons so much attention is being paid to moving to the next process node. It's not ne... » read more

Reducing Power At RTL


Power management and reduction at the register transfer level is becoming more problematic as more heterogeneous elements are added into advanced designs and more components are dependent on interactions with other components. This has been a growing problem in leading-edge designs for the past couple of process nodes, but similar issues have begun creeping into less-sophisticated designs as... » read more

Changes In AI SoCs


Kurt Shuler, vice president of marketing at ArterisIP, talks about the tradeoffs in AI SoCs, which range from power and performance to flexibility, depending on whether processing elements are highly specific or more general, and the need for more modeling of both hardware and software together. » read more

Tradeoffs In Embedded Vision SoCs


Gordon Cooper, product marketing manager for embedded vision processors at Synopsys, talks with Semiconductor Engineering about the need for more performance in these devices, how that impacts power, and what can be done to optimize both prior to manufacturing. » 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

What Engineers Are Reading And Watching


By Brian Bailey And Ed Sperling An important indicator of where the chip industry is heading is what engineers are reading and what videos they are watching. While some subjects remain on top, such as the level of interest in the latest manufacturing technologies, other areas come and go. The stories with the biggest traffic numbers are almost identical to last year. Readers want to know wh... » 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

← Older posts Newer posts →