Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
1–10 of 94 posts
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#2Well, putting out a tl;dr and then a graph that does not mention FP16/FP32 performance differences or anything related to TensorRT cannot be taken seriously if we talk about performance per watt. We need to see the a comparison that includes multiple scenarios so we can determine something like a break-even point between Nvidia GPUs and Apple M1 GPU, possibly even for several SotA models.
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#3No, but it's pretty good at retraining the final layer of low memory networks like MobileNet - weirdly a workload that the V100 is very poorly suited for...
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#4No, but it's pretty good at retraining the final layer of low memory networks like MobileNet - weirdly a workload that the V100 is very poorly suited for...
Not surprising since this is a training use case that Apple very much focuses on with CreateML.
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#5"trainable_params 12,810"
laughs
(for comparison, GPT3: 175,000,000,000 parameters)
Can Apple's M1 help you train tiny toy examples with no real-world relevance? You bet it can!
Plus it looks like they are comparing Apples to Oranges ;) This seems to be 16 bit precision on the M1 and 32 bit on the V100. So the M1-trained model will most likely yield worse or unusable results, due to lack of precision.
And lastly, they are plainly testing against the wrong target. The V100 is great, but it is far from NVIDIA's flagship for training small low-precision models. At the FP16 that the M1 is using, the correct target would have been an RTX 3090 or the like, which has 35 TFLOPS. The V100 only gets 14 TFLOPS because it lacks the dedicated TensorRT accelerator hardware.
So they compare the M1 against an NVIDIA model from 2017 that lacks the relevant hardware acceleration and, thus, is a whopping 60% slower than what people actually use for such training workloads.
I'm sure my bicycle will also compare very favorably against a car that is lacking two wheels :p
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#6> I chose MobileNetV2 to make iteration faster. When I tried ResNet50 or other larger models the gap between the M1 and Nvidia grew wider.
(and that's on CIFAR-10). But why not report these results and also test on a more realistic datasets? The internet is full of M1 TF brenchmarks on CIFAR or MNIST, has anyone seen something different?
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#7"trainable_params 12,810" laughs (for comparison, GPT3: 175,000,000,000 parameters) Can Apple's M1 help you train tiny toy examples with no real-world relevance? You bet it can! Plus it looks like they are comparing Apples to Oranges ;) This seems to be 16 bit precision on the M1 and 32 bit on the V100. So the M1-trained model will most likely yield worse or unusable results, due to lack of precision. And lastly, the…
thanks for the thorough comment. the article is, unfortunately, just clickbait.
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#8Can someone with more knowledge of Nvidia GPU's please say how much the V100 costs ($5-10K?) compared with the $900 mac mini.
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#9"trainable_params 12,810" laughs (for comparison, GPT3: 175,000,000,000 parameters) Can Apple's M1 help you train tiny toy examples with no real-world relevance? You bet it can! Plus it looks like they are comparing Apples to Oranges ;) This seems to be 16 bit precision on the M1 and 32 bit on the V100. So the M1-trained model will most likely yield worse or unusable results, due to lack of precision. And lastly, the…
This, not to mention one could get the GPU usage on the V100 way higher by training with larger batch sizes, which would also make training much faster.
Re: Analyzing the performance of Tensorflow training on M1 Mac Mini and Nvidia V100
#10"trainable_params 12,810" laughs (for comparison, GPT3: 175,000,000,000 parameters) Can Apple's M1 help you train tiny toy examples with no real-world relevance? You bet it can! Plus it looks like they are comparing Apples to Oranges ;) This seems to be 16 bit precision on the M1 and 32 bit on the V100. So the M1-trained model will most likely yield worse or unusable results, due to lack of precision. And lastly, the…
GPT3 is so big it would take 355 years to train on a nvidia V100, so your example is also not really useful for comparison. It would be interesting to see some mid-sized nn benchmarks though.