News for hackers. Stuff that matters.

A comment on An In-Depth Look at Google's Tensor Processing Unit Architecture. Parent comment

Re:An In-Depth Look at Google's Tensor Processing Unit Archi... by tarlinian on Wednesday April 05, 2017 @04:09PM (#14045156)
The lack of real 8-bit comparison data makes the whole paper a little suspect IMO...it's sort of like the early GPU papers that claimed 100x improvement over the CPU while running x87 scalar CPU instructions...the benefits are definitely still there but handicapping one architecture when it has features that are specifically capable of doing this is a bit stupid. It's not like they didn't have to do a lot of work on TF to make it output TPU instructions. When you're down to a 4x improvement...the benefits of specialized accelerators start to become somewhat questionable.

I do like that they highlighted the importance of low latency output though...that's even more critical for future non "Web" applications which have to run in real time.

[ Parent | on HN ]
Re:An In-Depth Look at Google's Tensor Processing Unit Archi... by bobdole1234 on Wednesday April 05, 2017 @05:16PM (#14045697)
The difference is the power numbers.

3.5x faster than CPU doesn't sound special, but when you're building inference capacity by the megawatt, you get a lot more of that 3.5x faster TPU inside that hard power constraint.

[ Parent | on HN ]