FRfreemack·Dec 10, 2024
In Exotic New Superconductors Delight and Confoundtensorflow
tpu
Jupyter Notebook2018–2024
tensorflow / tpu
tpu
Reference models and tools for Cloud TPUs.
7Distinct people
8Mentions
0Explicit recommenders
5.3KGitHub stars
What to know before exploring
What it is
Reference models and tools for Cloud TPUs.
GitHub
Where it is used
No documented product use has been verified yet.
What people are saying
No recommendation phrase was found in the indexed comments.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 12:54:37 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by year8 dated mentions
Observed mentions: 2018-02-12 18:02:55 UTC to 2024-12-10 11:33:53 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| 2018 | 2018-01-01 00:00:00 UTC | 2019-01-01 00:00:00 UTC | 2 | 1 | 1 |
| 2019 | 2019-01-01 00:00:00 UTC | 2020-01-01 00:00:00 UTC | 5 | 0 | 5 |
| 2020 | 2020-01-01 00:00:00 UTC | 2021-01-01 00:00:00 UTC | 0 | 0 | 0 |
| 2021 | 2021-01-01 00:00:00 UTC | 2022-01-01 00:00:00 UTC | 0 | 0 | 0 |
| 2022 | 2022-01-01 00:00:00 UTC | 2023-01-01 00:00:00 UTC | 0 | 0 | 0 |
| 2023 | 2023-01-01 00:00:00 UTC | 2024-01-01 00:00:00 UTC | 0 | 0 | 0 |
| 2024 | 2024-01-01 00:00:00 UTC | 2025-01-01 00:00:00 UTC | 1 | 0 | 1 |
What Hacker News says
8 sourcesAnecdotally, I have seen 50-100x (seriously) increases when moving from a V100 to TPUv3. To be fair, some of that is batch size increase. This was using the models in https://github.com/tensorflow/tpu/tree/master/models/officia.... On the other hand, a lot of those models are broken in some way and need fixing before running, caveat emptor.
In Filesystem devs should aim to make “badly written” app code “just work” (2009)Sure. The $2.48/hour per V100 GPU on GCP does not include the price of the CPU host; that is purely the price to rent a single accelerator. By contrast, a network-attached Cloud TPU v3 device includes both a CPU host and four connected TPU v3 chips that collectively deliver up to 420 teraflops. Furthermore, each individual V100 GPU on GCP has 16 GB of memory, whereas the Cloud TPU v3 device has 128 GB of HBM.
The best apples-to-apples performance-per-dollar comparison we have publicly available was published last fall, and it compared the performance and cost of using various Cloud TPU v2 Pod slice sizes with the performance and cost of using various numbers of V100 GPUs attached to a single GCP host:
https://cloud.google.com/blog/products/ai-machine-learning/n...
We went to great lengths to ensure that we trained exactly the same version
Read the rest
of ResNet-50 to the same accuracy in the same way across all hardware configurations. The methodology predated MLPerf and is documented in full here:
https://github.com/tensorflow/tpu/blob/master/benchmarks/Res...
If you were going to do a similar performance-per-dollar comparison today, the simplest approach might be to try to get the code from NVIDIA's MLPerf 0.6 submissions running at scale on one or more major public clouds using the fastest-available networking technology that each cloud provides:
https://github.com/mlperf/training_results_v0.6/tree/master/...
It would be very interesting to see how distributed training performance using large-scale GPU clusters in public clouds compares with the published on-premise MLPerf performance numbers using exactly the same MLPerf code and methodology. With these measurements in hand, it would then be straightforward to make performance-per-dollar comparisons with Cloud TPU v3 Pod slices of various sizes.
Associated google AI blog: https://ai.googleblog.com/2019/05/efficientnet-improving-acc...
and the source code: https://github.com/tensorflow/tpu/tree/master/models/officia...
In EfficientNet: Rethinking Model Scaling for Convolutional Neural NetworksYes I used Keras before PyTorch came out. IMO, Google needs to start over rather than try extensive wart removal on something that consists almost entirely of warts. These days Keras has subpar multi-GPU support (two GPUs are often slower than one), less flexibility, and more un-debuggable magic than PyTorch. And it doesn't look like things will be improving much (if at all) with TF 2.0.
And the TPU is a story of its own when it comes to usability. Take a look at their docs. TPU actually has to run on a separate machine, which you need to provision manually using their `ctpu` tool, and training of a trivial network on CIFAR looks like this: https://github.com/tensorflow/tpu/blob/master/models/experim...
In Google’s Cloud TPU Pods are now publicly available in betaHere is a larger-scale comparison of Cloud TPU and Google Cloud GPU performance and cost (focused on Cloud TPU Pods):
https://cloud.google.com/blog/products/ai-machine-learning/n...
All the code used in that comparison is open source, and there is a detailed methodology page with instructions that you can follow if you want to reproduce the results:
https://github.com/tensorflow/tpu/blob/master/benchmarks/Res...
Also, Cloud TPUs are available to everyone for free via Colab. Here is a sample Colab that shows how to train a Keras model on the Fashion MNIST dataset using the Adam optimizer:
https://colab.research.google.com/github/tensorflow/tpu/blob...
(I work on Cloud TPUs)
In Cost Comparison of Deep Learning Hardware: Google TPUv2 vs. Nvidia Tesla V100TPU port of select TensorFlow models
1 points · Open discussionThis is really Tensorflow as a service. You get an IP address and a port you send gRPC requests to:
https://github.com/tensorflow/tpu/blob/master/tools/diagnost...
Presumably, there's a whole server behind that address that has all the right drivers and libraries: details you don't need to care about.
The only partial lock-in is that not all ops are supported and you need to figure if there are any parts of the graph in the critical part that will run on the CPU instead. There's a tool for that:
https://cloud.google.com/tpu/docs/cloud-tpu-tools#tpu_compat...
Competitors could launch something similar that uses GPUs tomorrow. Now, if you don't already use TF and don't want to switch, that's another story.
In Cloud TPUs in Beta