2 ms·
You can take the example of nVidia Titan Xp (GPU that targets ML, scientific computing market) and Google TPUv2. - Titan Xp has single precision power of 12 Te
by fizixer 8y ago
You can take the example of nVidia Titan Xp (GPU that targets ML, scientific computing market) and Google TPUv2.
- Titan Xp has single precision power of 12 TeraFlops in boost mode.
- A Google TPUv2 device claims a power of 180 TF (consisting of four TPUv2 chips, of 45 TF each).
So I guess by dwarfed, I mean 6.67%
- Voloskaya 8y agoThis comparison makes 0 sense: You chose to compare TPUs to a Titan Xp, which is far from being the best ML GPU from NVIDIA and you are comparing it agains 4 TPUs. I invite you to check out the different benchmarks that were made comparing NVIDIA V100s againt TPUv2/v3: they all show that the performance are very similar, even slightly better for the V100 overall iirc. The only reason why TPUs are interesting is if you plan to rent an accelerator (that's also why Google doesn't sell them), you get around 50% savings for the same results using TPUs.