Download presentation
Presentation is loading. Please wait.
Published byGuilherme Deluca Minho Modified over 6 years ago
1
Closing Remarks Cyrus M. Vahid, Principal Solutions Architect,
Principal Solutions AWS Deep Learning June 2017
2
1014 parameters Brain runs at 100Hz
3
Brain-sized DNN Today Today: g2.16xlarge has 192 GB of GPU RAM. 32 bits/param. SGD needs 3x copies. => 12B parameters So you’d need 8,000x p2.16xl instances Cost: $115,000/hr
4
Amdahl’s Law Evolution according to Amdahl's law of the theoretical speedup in latency of the execution of a program in function of the number of processors executing it, for different values of p. The speedup is limited by the serial part of the program. For example, if 95% of the program can be parallelized, the theoretical maximum speedup using parallel computing would be 20 times.
5
Moore’s Law Price/performance doubles every 12-18 months
Already on the fast side of that. TPUs from Google / IBM / Nervana
6
Yann Le Cun “The best neural networks have always taken 3 weeks to train.”
7
Brain-sized DNN in 2026 256x = 28, Moore’s law: 8 doublings = 8-12 years 400x p7.256xl* instances. $1,000/hr* 3 weeks to train: $500k *Pure speculation, obviously
8
Cyrus M. Vahid
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.