Si2 Annual Technology Forum
Si2 is an R&D Joint Venture Delivering EDA Interoperability Solutions and Standards
This is an online event.
The semiconductor industry is entering a new era of creativity and productivity, fueled largely by cloud computing, which enables manufacturers to streamline every process, including design and verification. Join us for discussions on semiconductor design-in-the-cloud strategy, and progress reports from the technical staff of Si2, a leading R&D joint venture whose members collaborate to create EDA interoperability solutions and standards.
_______________________________________________________________________________
Welcome (8:00 - 8:05)
John Ellis, President and CEO
_______________________________________________________________________________
Keynote Presentation (8:05 - 8:35)
Silicon-to-System: Addressing Design, Verification, and Implementation Challenges With Cloud-Based Engineering
David Pellerin
Head of Worldwide Business Development
Hitech/Semiconductor
Amazon Web Services
Rapid growth in smart, connected devices, coupled with AI-driven methods of advanced analytics, have led to a corresponding increase in demand for scalable, cloud-based computing and data management platforms as well as increasing the need for more productive design, verification, and implementation solutions. AI is at the heart of many of the newest, most advanced analytics and IoT applications, ranging from robotics and autonomous vehicles, to cloud-connected products such as Amazon Alexa, to smart factories and consumer-facing services in the financial and healthcare sectors.
This talk presents examples of cloud-based design and verification methods used within Amazon in the creation of cloud-connected products, as well as examples of how AWS semiconductor customers increasingly rely on cloud-based, multi-vendor design and verification flows to innovate faster, and at cloud scales.
About the Speaker
David Pellerin is head of Worldwide Business Development for Hitech/Semiconductor at Amazon Web Services. Prior to joining AWS, Mr. Pellerin had a career in electronic design automation and hardware-accelerated reconfigurable computing. He has experience with digital logic simulation and optimization, high-level synthesis, grid and cluster computing, and embedded systems for image, video, and network processing. He has published five Prentice Hall technical books.
_______________________________________________________________________________
Special Presentation (8:35 - 8:50)
AI and Cloud: Enabling Production AI/ML in Silicon Flows
Sashi Obilisetty
Chief Architect, Silicon Solutions
Google Cloud
The unprecedented demand for chips from the automotive, AI, datacenter and other related segments has created substantial growth in semiconductors. New use cases and the proliferation of deep learning applications into every aspect of our lives further cements this trend and the continued growth expected from this sector.
On the other hand, the industry has also been facing deep challenges to combat a slowing Moore’s law, lack of skilled engineers, and the exponential rise in the cost of designing chips. Market pressures are enormous in this industry where time-to-market is especially important. AI/ML technologies have been embraced by EDA vendors, foundries and chip design teams over the past five plus years, and interesting learnings are emerging.
This talk addresses the challenges of building production AI from various perspectives. It details the solutions for roadblocks to production AI, such as continuous training, model evaluation, low-latency model serving and model explainability.
_______________________________________________________________________________
_______________________________________________________________________________
Si2 Board of Directors and Staff Introductions (9:25 - 9:30)
John Ellis, President and CEO_______________________________________________________________________________
Si2 Technical Coalition Updates (9:30 - 9:40)
OpenAccess, oaScript Extensions
Marshall Tiner
Director of Production Standards
Standard Compact Models
Peter Lee
Chair, Compact Model Coalition
AI/ML, Unified Power Models
Dr. Leigh Anne Clevenger
Director of OpenStandards
_______________________________________________________________________________
Si2 Power of Partnerships Awards (9:40 - 9:45)
Authors of A Collaborative Model for AI/ML in EDA
Authors of the respected Si2 white paper, A Collaborative Data Model for AI/ML in EDA, are winners of the 2021 Power of Partnerships awards, presented annually to the Si2 team that has made the most significant contributions to the field of electronic design automation.
The first white paper from the AI/ML in EDA Special Interest Group, it forms the foundation for an Si2 API Working Group proposal and aligns with the executive three-year vision of the Si2 board and member companies. The ideas described extend beyond AI/ML to any secure data exchange between EDA tools, including analytics and other cloud-ready methodologies.
This year's winners are:
Kerim Kalafala, IBM
Veeravanallur Parthasarathy, AMD
Norman Chang, Ansys
Akhilesh Kumar, Ansys
Elias Fallon, Cadence Design Systems
Sriram Madhavan, GLOBALFOUNDRIES
Prateek Bhansali, Intel
Srinivas Bodapati, Intel
Chandramouli Kashyap, Intel
James Masters, Intel
Ramy Iskander, Intento Design
Larg Weiland, PDF Solutons
Karthik Aadithya, Sandia National Laboratory
Boon-Siang Cheah, Synopsys
Mengdi He, Synopsys
Runners Up
Compact Model Coalition
Committee/Working Group: Open Model Interface (OMI)
Chair: Colin Shaw, Silvaco
OpenAccess Coalition
Committee/Working Group: Polygon Operators Working Group (oaxPop)
Chair: James Masters, Intel
_______________________________________________________________________________
Special Announcement (9:45 - 9:55)
Learn About a New Si2 Industry Initiative Focusing on
Silicon-to-System Design Interoperability
Vic Kulkarni
Chief Strategy Officer
_______________________________________________________________________________