Experiences Building Edge TPU with Chisel
Channel:Chisel · Length:37:47 · Published:2018-11-17 · Views:10560
Summary
A Google talk from the Chisel community in which engineers recount designing the Edge TPU with the Chisel hardware construction language. It covers the productivity gains and the problems they hit.
Why HN recommends it
Pointed to as a rare first-hand account of a real, shipping ASIC built in an open-source HDL, useful in HDL and FPGA choice discussions.
Google had a really interesting talk on their experiences with Chisel in designing the first Edge TPU chip. — HN, 2023
Machine learning & deep learning · Computer architecture & CPUs · Semiconductors & chip manufacturing
Subtitles (excerpt)
awesome thank you hello everyone my name is Derrick Lockhart and I'm from Google me and my co-presenter Stephen twig are going to be talking about our experiences using chisel to build edge TPU to my knowledge this is the first kind of public acknowledgement that we have used chisel at Google and use chisel to build actual hardware so that's pretty exciting some caveats about this talk even though we're in the architecture section you're gonna be sorely disappointed about the architecture I talked about because there won't be any this is primarily about us using chisel and our experiences with that in that regard there's gonna be two main components one is going to be Steven talking about the actual extensions and infrastructure we built in chisel to make the HTTP you project successful and the other aspect will be me talking on behalf of the designers in the project that's great sorry about that me talking on behalf of the designers of the project worked well what worked well and what didn't work well in the project and another caveat about that so we use chisel 3 and it's a really very early stages a lot of the things I talked about may be addressed at this point but I just wanted to bring up highlight
On Hacker News
Google had a really interesting talk on their experiences with Chisel in designing the first Edge TPU chip. A lot of the issues are specific to industrial chip design, where verification and coverage are essential, but many would apply to small-scale personal designs as well. A good overview of the advantages as well as the problems. And many of the problems would apply to other front ends, which… — therealcamino, 2023-02
Google TPU engineers used open-source Chisel for ASIC design (2018), [VIDEO] "Google Edge TPU devices" , 100 comments (2019), [LINK] & [LINK] "Coral Edge TPU review" , 100 comments (2020), [LINK] "TPU transformation: 10 years of our AI-specialized chips" , 60 comments (2024), [LINK] — walterbell, 2025-05
It was in fact the EdgeTPU (which is different from the TPU used in data centers). The talk from a google engineer can be found on youtube: [VIDEO] Please note that they were using a version of Chisel from 5+ years ago and many things have changed since then. It is still true though that Chisel can be hard to learn for typical hardware engineers, which is why it may be best suited for small and h… — ekiwi, 2021-07
Google built one of their TPUs in Chisel [1]. TL;DW: Chisel is beautiful/fun to write in, with a definite productivity bonus, but has a pretty large learning curve and had a much greater verification cost, partly because it's an HLS (most have that problem) and also lack of any tooling. Both of those costs are gradually being reduced (though in my opinion, not enough to not make verification a PI… — ljhsiung, 2021-01
In Nov 2018, Google engineers who designed the TPU gave a presentation about the use of open-source Chisel for designing the ASIC, [VIDEO] From [LINK] > Google will have the cloud TPU ... to handle training models for various machine learning-driven tasks, and then run the inference from that model on a specialized chip that runs a lighter version of TensorFlow that doesn’t consume as much power … — walterbell, 2019-02
Related videos
- Indistinguishable From Magic: Manufacturing Modern Computer Chips · Andor Gafotas · 1:01:20
- The spelled-out intro to neural networks and backpropagation: building micrograd · Andrej Karpathy · 2:25:52
- But what is a neural network? | Deep learning chapter 1 · 3Blue1Brown · 18:40
- 8-bit computer update · Ben Eater · 6:53
- Light Years Ahead | The 1969 Apollo Guidance Computer · TNMoC · 1:21:22
- “Hello, world” from scratch on a 6502 — Part 1 · Ben Eater · 27:24
- AlphaGo - The Movie | Full award-winning documentary · Google DeepMind · 1:30:28
- Transformers, the tech behind LLMs | Deep Learning Chapter 5 · 3Blue1Brown · 27:14
- Finding the Missing Memristor - R. Stanley Williams · The Qualcomm Institute · 47:52
- The Ultimate Game Boy Talk (33c3) · media.ccc.de · 1:01:06
- USENIX Security '18-Q: Why Do Keynote Speakers Keep Suggesting That Improving Security Is Possible? · USENIX · 51:22
- The Thirty Million Line Problem · Molly Rocket · 1:48:55
- Lecture 1 | Machine Learning (Stanford) · Stanford · 1:08:40
- MarI/O - Machine Learning for Video Games · SethBling · 5:57
- Deep Dive into LLMs like ChatGPT · Andrej Karpathy · 3:31:23
- Computer program that learns to play classic NES games · suckerpinch · 16:17