# How AI and DA combined with HPC shape innovation in Exascale architectures

- Institution: CERN
- Department: other events or meetings
- Speaker: Marie-Christine Sawley
- Year: 2019
- Language: English
- Source: [CERN](https://videos.cern.ch/record/3002151)
- Licence: [CERN](http://copyright.web.cern.ch)
- Who can watch: Teens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

Major efforts to build an exascale class computer are under way in four regions of the world. This time, the game changer compared with the previous decade—sustained PF—is the impressive growth of Deep Learning/Ai/Data analytics in all industry segments. This sustained growth and inflexion towards data-centric workflows is having a significant impact on the architectural features, from HW and SW perspectives, of an Exascale class computer. At very coarse grain, the main questions raised by this change are: how to move data faster; where to store data in the most efficient and cost effective way possible; how to process complex workflows and how to compute at unprecedented large scale and high performance. For each of these questions we will share how selected technology bricks help us building a powerful and sustainable solution.

This talk will also share the importance of collaborations with the community to helped tracing the path up to this new chapter of HPC and Data analytics..

0:00:00 Slide 1

0:02:01 Slide 2

0:02:23 Slide 3

0:03:44 Slide 4

0:04:21 Slide 5

0:06:31 Slide 6

0:08:35 Slide 7

0:10:06 Slide 8

0:10:11 Slide 9

0:11:51 Slide 10

0:13:09 Slide 11

0:14:19 Slide 12

0:15:03 Slide 13

0:16:08 Slide 14

0:16:44 Slide 15

0:17:33 Slide 16

0:18:27 Slide 17

0:19:55 Slide 18

0:20:42 Slide 19

0:21:18 Slide 20

0:23:01 Slide 21

## Lectures

1. [How AI and DA combined with HPC shape innovation in Exascale architectures](https://invisiblenetwork.ai/learn/cern/3002151/d2f2cd0201aa), 29 min (CERN's player)