ydau

Yes! It’s coming to all regions except Utah in a couple weeks. Utah will be a bit longer (couple months)

Hey! FYI, we are hoping to roll out a fix for the PyTorch issue tomorrow (I’m one of the founders of Lambda).

Also, that’s good feedback on the CPUs bottlenecking. I’ll let our HPC hardware team know about this.

We are also looking into GPU direct storage to help resolve this:

https://developer.nvidia.com/blog/gpudirect-storage/

You might also enjoy “Ted the Caver”

http://www.angelfire.com/trek/caver/index.html

GPT-3 was another buckets worth of evidence in favor of the scaling hypothesis. Performance kept improving (and cost to train kept increasing) as more parameters were added. Even with 175 billion parameters, the performance had not yet plateaued. One take-away is that throwing a lot of compute at the problem helps tremendously :).

You can read more about GPT-3 here: https://lambdalabs.com/blog/gpt-3/

Happy to add any demos / papers to this post that you think are worth linking.

TLDR:

Tesla V100s have I/O pins for at most 6x 25 GB/s NVLink traces. So, systems with more than 6x GPUs cannot fully connect GPUs over NVLink. This causes I/O bottlenecks that significantly diminish returns of scaling beyond six GPUs.

This article provides an overview of their architecture that bypasses this limitation using additional high bandwidth links. Looking at the benchmarks, multi-GPU performance scales almost perfectly linearly from 1x GPU 16x GPUs.

I'm one of the engineers who worked on this project. Happy to answer any questions!

A large chunk of this page is related to simplifying ML stack installation. See the section on tensorman.

Lambda Stack supports 16.04 and 18.04 at the moment.

It's value prop is to enable people to easily install TensorFlow / PyTorch and their dependencies in a container-less fashion. Though it doesn't provide the isolation of containers.

What I've learned from talking to customers is that many people don't care that much about handling multiple versions of the same framework. I wouldn't be surprised if you find that, like Lambda Stack, people are mainly using this product to easily get started with TensorFlow/Pytorch.

Now that TensorFlow 2.0 is out, we will see a much more stable API. People won't have to change their code if TensorFlow bumps up a dependency version. For many, this will reduce the impetus for moving to containers.

Lambda Stack is an alternative to Pop!_OS's tensorman, which doesn't use containers:

https://lambdalabs.com/lambda-stack-deep-learning-software

This is a one-line apt/aptitude installation for TensorFlow, PyTorch, CUDA, cuDNN, etc. When NVIDIA releases a new version of CUDA, you can simply apt-get upgrade to the latest version.

Disclosure: I work for Lambda Labs.

Lambda Labs: https://lambdalabs.com has 1080 Ti GPUs for rent. We also sell GPU workstations and servers for AI.