Why the GPGPU is Less Efficient than the TPU for DNNs
The GPGPU’s massive multithreading is unnecessary for DNNs, and imposes performance, area, and energy overheads. By avoiding such multithreading, the TPU is more efficient.
The Case for Quantum Computing
In a recent IEEE Spectrum article, Mikhail Dyakonov makes The Case Against Quantum Computing, focusing on the idea that building a quantum computer would require precise control over 2300 continuous variables. This view is absolutely correct if we were building an...
The Year in Review
Beyond Vertex Programs: EDA 3.0
In the past ten years, there has been a lot of progress in the design of programming models for parallel graph algorithms and their implementation on a variety of hardware platforms including multicore CPUs and GPUs. Vertex programs for graph analytics For the most...
Expanding mentoring opportunities for Young Architects
“A lot of people have gone further than they thought they could because someone else thought they could”. — Zig Ziglar, renowned motivational speaker Good mentoring can have an outsized impact on the careers of students, especially during early stages of...
The Future in Visual Computing: Research Challenges
The tremendous growth in visual computing is fueled by the rapid increase in deployment of visual sensing (e.g. cameras) in many usages ranging from digital security/surveillance and automated retail (e.g. smart cameras & analytics) to interactive/immersive...Subscribe
