JUNE 18–22, 2017

Presentation Details

Name: Memory-Centric Server Architecture
Time: Tuesday, June 20, 2017
09:15 am - 09:30 am
Room:   Panorama 1
Messe Frankfurt
Breaks:07:30 am - 10:00 am Welcome Coffee
Speaker:   Babak Falsafi, EPFL
Abstract:   Large-scale IT services are now increasingly run in memory due to tight application latency demands. With services centered mostly around data, datacenter owners often integrate as much DRAM into a single blade as technology allows, and use low-latency high-bandwidth network fabrics to aggregate near-neighbor DRAM into large memory pools. DRAM accounts for a substantial fraction of overall server capital and operation costs, and as such designers are increasingly customizing server hardware, software and infrastructure for online services around memory. In this talk, I will first motivate specialized server design for in-memory computing and then present promising avenues to explore specialization to improve server design.