3 ms·
>It seems likely that Microsoft and Google will introduce ARM-based instances. Actually google was the first. Its called TPU - Tensor Processing Units and has
by sandGorgon 5y ago
>It seems likely that Microsoft and Google will introduce ARM-based instances.
Actually google was the first. Its called TPU - Tensor Processing Units and has been available for their ML platform for a long time. Interestingly, TPU replaces nvidia in this case.
There are already largescale Kaggle benchmarks on CPU vs GPU vs TPU - https://towardsdatascience.com/when-to-use-cpus-vs-gpus-vs-tpus-in-a-kaggle-competition-9af708a8c3eb https://towardsdatascience.com/when-to-use-cpus-vs-gpus-vs-t...
- ma2rten 5y agoI'm pretty sure TPUs are not based on ARM.
- sandGorgon 5y agoI'm not claiming they are on ARM - and that in itself is a moot point. I'm mentioning that Google Cloud primarily hosts and bills your ML code on TPUs. Today TPU v4 run at exascale. so the equivalence "scared" behavior exists for both Intel and nvidia on Google already.
- bushbaba 5y agoGCP cloud TPUs are comprised of multiple ASICS underneath. To say 1 TPU is faster than 1 GPU is moot as the Cloud TPU is a logical construct. Instead $/ML training job or time to complete training given X budget is likely a better measure.