The end of Moore's law has been predicted for many years, in combinations with statements about exaggerations of its end. However, we are currently observing a phase of substantial slow-down in key metrics associated with silicon device manufacturing, and, as a result, huge issues with sustained performance scaling. In this talk, we will review the most important recent and current trends in computer architecture, in particular in the light of diminishing returns from feature size scaling and power constraints of CMOS technology. Ultimately, we will see that energy efficiency is and will be key for a continuous performance scaling, however in substantially different flavors. In this context, particular attention will be put on predictive models for performance and power, as well as a short review of our efforts in the context of the DeepChip project, and how both can help from a system's perspective for continued performance scaling.
Holger Fröning is a full professor at the Faculty of Mathematics and Computer Science at Heidelberg University (Germany), and leads the Computing Systems Group at the Institute of Computer Engineering. His research interests focus on machine learning, data analytics and high performance computing, and include hardware and software architectures, programmability, co-design, data movement optimizations, and associated power and energy aspects. Previously, he was associate professor at the same university. In 2016, he was with NVIDIA Research (Santa Clara, CA, US) as visiting scientist, sponsored by Bill Dally. Early 2015 he was visiting professor at the Graz University of Technology (Austria), sponsored by Gernot Kubin. From 2008 to 2011 he reported to Jose Duato from the Technical University of Valencia (Spain). He has received his PhD and MSc degrees 2007 respectively 2001 from the University of Mannheim, Germany. He was awarded a Google Faculty Research Award in 2014. Four of his publications have received a best paper award, and parts of his research results have been commercialized. He currently co-organizes the Workshop on Embedded Machine Learning (WEML) and the Workshop on IoT, Edge, and Mobile for Embedded Machine Learning (ITEM). He chaired tracks for EuroPar 2015 and International Supercomputer Conference 2017, and recently served as program committee member for IPDPS2021/19/18, CCGRID2020/19/18, SC2017, ICPP2020/19/18/17/16, CLUSTER2018/16, FPL2020/19, and Euro-Par2019. He is frequently providing reviews for established journals, such as IEEE Micro, TPDS, and JPDC.
His recent sponsors include BMBF, DFG, FWF, NVIDIA, SAP, and XILINX. For more information, visit his website: http://www.ziti.uni-heidelberg.de/compsys
Further information about the virtual format
- Meeting ID: 851 6207 0256; Passcode: 031439