Evolving EDA Beyond its E-Roots: An Overview (invited paper) Andrew B. Kahng†‡ and Farinaz Koushanfar†∗ †ECE and ‡CSE Depts., UC San Diego *ECE Dept., Rice University {abk,fkoushanfar}@ucsd.edu Outline • Motivation, challenges, and our story • Study groups overview – Group I: Systemization of prior efforts – Group II: Metrics of DA research impact – Group III: Vision for the EDA field • Research gap analysis • The 2015 DA perspective challenge • Looking forward Motivation • Electronic design automation (EDA) – – – – – is an engineering success story has mainly focused on supporting cost scaling, i.e., “More Moore” one of the first truly inter-disciplinary fields has reached a mature state as semiconductors also matured is witnessing a diminished interest among ECE and CS students • Urgent to revisit how the EDA field will evolve & grow • Healthy growth requires exciting new directions as well as a steady supply of new experts trained at the graduate level Challenges • How can the EDA paradigms and methodologies be leveraged for DA in other, emerging domains to solve standing real-world problems? • EDA researchers actively contribute to other fields’ DA • But… evolution and growth as a community requires a more systematic, coherent effort – as well as vision • How can EDA systematically “move beyond its E-roots”? • To address the challenges, we have formed a recent initiative of the IEEE Council on Electronic Design Automation (CEDA) Our story • Two seeds of this initiative: – Series of CCC workshops on extreme-scale design automation – CANDE (ComputerAided Network DEsign) committee (since 1972) • Redirecting CANDE under CEDA president-elect, Dr. Shishpal Rawat, focusing on new initiatives/ activities • Our team (organized into three study groups): – Group I: Farinaz Koushanfar, Gang Qu, Zhiru Zhang – Group II: Andrew B. Kahng, Gi-Joon Nam, David Pan – Group III: Deming Chen, Priyank Kalla, Subhasish Mitra, Steven Levitan, Miodrag Potkonjak Study Groups Overview • Group I: Systemization of prior efforts – Summarized & analyzed: (i) Recent roadmaps; (ii) NSF EDA Expedition grants; and (iii) SRC focused research centers • Group II: Metrics of DA research impact – Analysis of research outputs, leveraged funding, industry and publication data with modern text mining • Group III: Vision for the EDA field – Focusing on new opportunities inspired by technologies and applications, and real-world problems • More details in this session talks and papers[1][2][3] [1] Koushanfar, Mirhoseini, Qu, Zhang, "DA Systemization of Knowledge: A Catalog of Prior Forward-Looking Initiatives (invited paper)", ICCAD 2015. [2] Kahng, Luo, Nam, Nath, Pan, Robins, " Toward Metrics of Design Automation Research Impact (invited paper)", ICCAD 2015. [3] Potkonjak, Chen, Kalla, Levitan, “DA Vision 2015: From Here to Eternity (invited paper)", ICCAD 2015. RESEARCH GAP ANALYSIS Research gap analysis (1/2) • SRC studies to determine the magnitude of research and world-wide “research gap” – Research needs derived from ITRS technology requirements in the 2001 and 2003 studies – Massive funding gaps identified in studies Estimated Worldwide annual research investment to support 2008-14 needs Research gap analysis (2/2) • 2001 Recommendations for addressing research gap – The US semiconductor industry must consider increasing $ for long-term horizon ITRS research – SRC and MARCO must explore options to attract foreign semiconductor companies as full members – SRC and MARCO explore collaborations w foreign consortia – SRC needs better coordination with US government semiconductor related funding programs • Many of the recommendations and conclusions of the analysis seem to be timely even today! DA PERSPECTIVE CHALLENGE 2015 DA perspective challenge 2015 (1/2) • Submitters of the challenge asked to suggest longterm problems and their challenges: – – – – What is the long-term problem Why is the problem important and challenging? What is the state-of-the-art? What is the problem’s relevance to existing DA tools and methods? How can DA help in addressing the challenges? – What knowledge, skills, and/or tools needed for addressing? – Is the problem interdisciplinary, requiring expertise other than DA? – What are the broader impacts? DA perspective challenge 2015 (2/2) • 30 submissions, 13 invited to the workshop @DAC’15 – – – – – – 3-minutes talks to a panel of 6 academic/industrial judges 1st prize: Huang&Cheng (UCSB) - DA of flexible electronics 2nd prize: Chen&Li (Penn State) - DA for neuromorphic 3rd prize: Chang (KAIST) - Energy optimization for EVs Audience favorite: (U of Calgary) DA of energy systems Honorable mentions: (IBM Research) DA for networks of autonomous vehicles; (MIT) DA for trusted hardware • The prize winners will be participating in our panel LOOKING FORWARD Looking forward • EDA has been successful as a field, but has reached a mature state • Growing beyond our E-roots requires systematic community effort and vision • Report on our recent CEDA initiative and efforts of 3 study groups • Research gap analysis • DA perspective challenge 2015: – Input from the community about their XEDA vision • Need your help in moving forward! Thank you! • Questions? • {abk,fkoushanfar}@ucsd.edu