Christoph Hagleitner

IBM, Switzerland

Heterogeneous Computing Systems and Disaggregation in Cloud Datacenter

Abstract

The long success story of Moore’s law and technology scaling is slowly approaching its limits and the whole industry is looking for hardware accelerators, which can continue to advance performance and energy efficiency. At the same time, cloud computing is evolving into the dominant computing platform. The industry is moving away from homogeneous computing platforms. Today, GPUs are the dominant accelerators, but an increasing number of companies is considering a broader use of heterogeneous computing systems. One option is reconfigurable computing based on FPGAs. Reconfigurable computing offers the architectural flexibility to accelerate a wide range of workloads in combination with improved power efficiency. We are going to discuss hetergeneous computing systems and disaggregation in the datacenter using cloudFPGAs an example. We developed CloudFPGA as a reconfigurable computing platorm, which based on network-attached FPGAs. CloudFPGA achieves scalability through a novel packaging concept and disaggregation. In this presentation we are going to describe the basic components of the platform and how applications can be developed, deployed and scaled.

Biography

Christoph Hagleitner leads the “Heterogeneous Cognitive Computing Systems” group at the IBM Research –Zurich Lab (ZRL) in Ruschlikon, Switzerland. The group focuses on heterogeneous computing systems for cloud datacenters and HPC. Applications include security, big-data analytics and cognitive computing. He obtained a diploma degree in Electrical Engineering from ETH, Zurich, Switzerland in 1997 and a Ph.D. degree for a thesis on CMOS-integrated Microsensors from ETH, Zurich, Switzerland in 2002. In 2003 he joined IBM Research to work on the system architecture of a novel probe-storage device (“millipede”-project). In 2008, he started to build up a new research group in the area of accelerator technologies. The team initially focused on on-chip accelerator cores and gradually expanded its research to heterogeneous systems and their applications.