Due to the overwhelming technical advantages of having on-chip memories, embedded memories are ubiquitous in most chip designs, and can comprise significant portions of a chip (upwards of 50%, ...

The new Intel “Knights Landing” processor’s topology includes what it calls near memory, an up to 16 GB block of on-package memory accessible faster and with higher bandwidth than traditional main ...

Generic test and repair approaches to embedded memory have hit their limit. Smaller feature sizes, such as 130 nm and 90 nm, have made it possible to embed multiple megabits of memory into a single ...

Using dual-port memories (“dual ports”) as system interconnects has proven to be an effective interface strategy for bridging multiple processing elements in high-performance applications. Not only do ...

In the first part of this series, we discussed the need to perform power optimizations and exploration at higher levels of abstractions, where the potential to reduce the power consumption was highest ...

The dawn of GPU computing came about in large part due to the immense gap in compute performance between traditional CPUs and programmable GPUs. Whereas CPUs excel with serial workloads, modern GPUs ...

Leaks suggest that NVIDIA’s future Feynman GPU architecture, expected around 2028, could introduce stacked SRAM memory blocks as part of a broader push toward more specialized and efficient processing ...