Achieving Best QoR and Fastest TAT with Synopsys Fusion Technology

Slides:



Advertisements
Similar presentations
Protect Your Business and Simplify IT with Symantec and VMware Presenter, Title, Company Date.
Advertisements

How to Accelerate the Analog Design Verification Flow Itai Yarom Senior Verification Expert Synopsys.
Accelerating Productization. Functional Metrology TM Challenges of Semiconductor Productization Leading IDM’s Solution Novel Solution -> In-product Functional.
© 2004 Visible Systems Corporation. All rights reserved. 1 (800) 6VISIBLE Holistic View of the Enterprise Business Development Operations.
EECE579: Digital Design Flows
The Design Process Outline Goal Reading Design Domain Design Flow
ENEE 644 Dr. Ankur Srivastava Office: 1349 A.V. Williams URL: Computer-Aided Design of.
Hierarchical Physical Design Methodology for Multi-Million Gate Chips Session 11 Wei-Jin Dai.
ASIC Design Introduction - 1 The history of Integrated Circuit (IC) The base for such a significant progress –Well understanding of semiconductor physics.
Introduction to RUP Spring Sharif Univ. of Tech.2 Outlines What is RUP? RUP Phases –Inception –Elaboration –Construction –Transition.
Fast & Furious: Taming the Challenges of Advanced-Node Design Anirudh Devgan, Senior Vice President, Digital & Signoff Group.
1 Designing for 65nm and Beyond Where’s The Revolution ?!? Greg Spirakis Absolutely, positively not working for Intel (or anyone else) EDP 2005.
SOC Design Challenge Rajeev Madhavan Chairman and CEO.
ECO Methodology for Very High Frequency Microprocessor Sumit Goswami, Srivatsa Srinath, Anoop V, Ravi Sekhar Intel Technology, Bangalore, India Introduction.
ITRS Factory Integration Difficult Challenges Last Updated: 30 May 2003.
COMPUTER SCIENCE &ENGINEERING Compiled code acceleration on FPGAs W. Najjar, B.Buyukkurt, Z.Guo, J. Villareal, J. Cortes, A. Mitra Computer Science & Engineering.
Design Verification An Overview. Powerful HDL Verification Solutions for the Industry’s Highest Density Devices  What is driving the FPGA Verification.
A New Method For Developing IBIS-AMI Models
Xilinx Programmable Logic Design Solutions Version 2.1i Designing the Industry’s First 2 Million Gate FPGA Drop-In 64 Bit / 66 MHz PCI Design.
System Design with CoWare N2C - Overview. 2 Agenda q Overview –CoWare background and focus –Understanding current design flows –CoWare technology overview.
CSE 494: Electronic Design Automation Lecture 2 VLSI Design, Physical Design Automation, Design Styles.
Marty Sanders, VP of Americas Economic Benefits of Deploying Disruptive Hyperconvergence Technology September 2015.
© 2012 xtUML.org Bill Chown – Mentor Graphics Model Driven Engineering.
ASIC, Customer-Owned Tooling, and Processor Design Nancy Nettleton Manager, VLSI ASIC Device Engineering April 2000 Design Style Myths That Lead EDA Astray.
Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display. PowerPoint to accompany Krar Gill Smid Technology of Machine.
1 Optimizing compiler tools and building blocks project Alexander Drozdov, PhD Sergey Novikov, PhD.
1. CAD Challenges for Leading-Edge Multimedia Designs Ira Chayut, Verification Architect (opinions are my own and do not necessarily represent the opinion.
SSV Summit November 2013 Cadence Tempus™ Timing Signoff Solution.
EE 466/586 VLSI Design Partha Pande School of EECS Washington State University
Ready to Use Programmable Logic Design Solutions.
Cisco Consulting Services for Application-Centric Cloud Your Company Needs Fast IT Cisco Application-Centric Cloud Can Help.
Microprocessor Design Process
Azure Stack Foundation
Scales Technology FZCO
Introduction to ASICs ASIC - Application Specific Integrated Circuit
Gopakumar.G Hardware Design Group
Overview Modern chip designs have multiple IP components with different process, voltage, temperature sensitivities Optimizing mix to different customer.
Makes Insurance Smarter.
Time-borrowing platform in the Xilinx UltraScale+ family of FPGAs and MPSoCs Ilya Ganusov, Benjamin Devlin.
CIM Modeling for E&U - (Short Version)
Ph.D. in Computer Science
Digital Workplace.
Enabling machine learning in embedded systems
Navision Business Analytics
课程名 编译原理 Compiling Techniques
DEFECT PREDICTION : USING MACHINE LEARNING
Intelligent Design of Electronic Assets (IDEA) Workshop
Chapter 4: Switched Networks
How to Increase HCM Cloud User Adoption with Guided Learning
The Value of SAP’s Research and Development Solutions Transforming research and development in the digital economy Enterprise Portfolio and Project Management.
Oil and Gas Industry Business Management Software Extends Across the Office Ecosystem “Building seamless integration between the systems that our customers.
A Must to Know - Testing IoT
11/14/2018 Changing Paradigms - Fast-Turn RF and Mixed-Signal IP IEEE Annual EDP-2001 Workshop session on Analog/MS Design Flows James Spoto Enablix.
Digital Transformation Asia 2018 – CALL FOR SPEAKERS
Matlab as a Development Environment for FPGA Design
ITRS Roadmap Design Process Open Discussion EDP 2001
Chapter 10: IC Technology
The performance requirements for DSP applications continue to grow and the traditional solutions do not adequately address this new challenge Paradigm.
Chapter 10: IC Technology
HIGH LEVEL SYNTHESIS.
GRC - A Strategic Approach
Key Note Address Uncertain Economic Order Challenges for CIO
Chapter 10: IC Technology
Xilinx Alliance Series
Opening comments… KT Moore
Electromagnetic Crosstalk Analysis and Sign-off For Advanced Node SoCs
I4.0 in Action The importance of people and culture in the Industry 4.0 transformation journey Industry 4.0 Industry 3.0 Industry 2.0 Industry 1.0 Cyber.
The Intelligent Enterprise and SAP Business One
Presentation transcript:

Achieving Best QoR and Fastest TAT with Synopsys Fusion Technology Dr. Thomas Andersen, Senior Director, Engineering, Synopsys Design Group May 1, 2018 May 1, 2018

Synopsys Fusion Technology Best QoR and Fastest TTR Fusion Data Model PT, StarRC, PTPX, ICV, RedHawk Synthesis IC Compiler II Test Design Fusion ECO Fusion Signoff Fusion Test Fusion Technology Leadership #1 Anchors: Synthesis, P&R, STA, Extraction Massively Parallel New Fusion Architecture Best QoR: Tech Moves Across Products Fastest TTR: Common Platform Engines 4 Fusion Types: Design, ECO, Signoff & Test New Digital Intelligence Machine Learning Throughout Additional QoR, TTR May 1, 2018

Design Fusion Common Technology Between Synthesis and P&R Test Fusion Signoff Fusion ECO Fusion Enables Convergence, Fastest TTR e.g. Congestion . Synthesis P&R Synthesis Technology moves inside P&R for Best QoR Synthesis QoR inside P&R P&R QoR inside Synthesis Unified Optimization Strategy In addition, we are working on ‘Flexible Handoff’ from synt to P&R. This allows the front-end designers to go deeper into physical design – from netlist level all the way to legalized placement. Handoff can be of netlist, pre-legalized placement or legalized placement. May 1, 2018

Design Fusion Results Synthesis based Logic Optimization inside P&R Improved QoR, flexible handoff and faster convergence Customer Results Up to 5% Area reduction Up to 10% Leakage reduction Significant Congestion Reduction          May 1, 2018

ECO Fusion 30% Reduction in ECO TTR & Better QoR Exiting P&R Test Fusion Signoff Fusion Design Fusion route_opt with PT & StarRC ECO Fusion IC Compiler II PrimeTime & StarRC PBA (Path Based Analysis) Optimization in P&R for Best QoR 30% Reduction in ECO TTR Better QoR from P&R Flow May 1, 2018

4 Manual ECOs Iterations ECO Fusion ECO Fusion Convergence with Implementation Scenarios Accelerates Sign-off Closure ICC2 PT&ICC2 PT route_opt 4 Manual ECOs Iterations ICC2 PT 30% Faster ECO Fusion Reduced route_opt May 1, 2018

Signoff Fusion PrimeTime, StarRC, PT-PX & RedHawk for Optimization AND Signoff Test Fusion Design Fusion ECO Fusion Golden Sign-off Analysis in Optimization Zero Margin Platform Eliminates Pessimism With Signoff fusion, we are using PT and StarRC for signoff and for optimization across the flow. Industry golden signoff tech will be fused with implementation. The timing and analysis engines will be brought to synthesis and P&R. This enables perfect correlation. As a result, the platform becomes zero margin enabled. We are going to extend the backbone to IR and Thermal as well. Exclusive RedHawk Fusion Extends to IR, Thermal May 1, 2018

Signoff Fusion Exclusive Partnership with Ansys CALL TO ACTION Traditional Signoff Fusion IC Compiler II IC Compiler II RedHawk Fusion 237 mV worst IR drop StarRC StarRC PT / PT PX PT / PT PX 196 mV worst IR drop RedHawk Manual Iterative Automated Clean Block Handoff Excellent Correlation and Results May 1, 2018

Synopsys Fusion Technology Best QoR and Fastest TTR Fusion Data Model PT, StarRC, PTPX, ICV, RedHawk Synthesis IC Compiler II Test Design Fusion ECO Fusion Signoff Fusion Test Fusion Technology Leadership #1 Anchors: Synthesis, P&R, STA, Extraction Massively Parallel New Fusion Architecture 10% Better QoR: Tech Moves Across Products 2X Faster TTR: Common Platform Engines 4 Fusion Types: Design, ECO, Signoff & Test New Digital Intelligence Machine Learning Throughout Additional QoR, TTR, Productivity May 1, 2018

The Rise of Machine-Learning Enabling new applications across almost all aspects of the global economy May 1, 2018

Machine-Learning for EDA: Why? A quick look at current semiconductor and design economics Cost Complexity Design INTEL, TECHNOLOGY & MANUFACTURING DAY, 2017 ARXIV:1402.2459V1 [CS.AR] 11 FEB 2014 NVIDIA, SNUG SILICON VALLEY 2018 Moore’s Law has slowed down Designing ICs is expensive (7nm: >$250M [Gartner]) New node ROI: No less than perfect required Most problems are intractable Competing heuristics, iterations, chaos Agony of choice: what to do next? EDA solutions deterministic in nature High-automation, high-configurability Paradox Many, many moving parts in EDA flows Promise of Machine-Learning in EDA: Make our Design World SIMPLER May 1, 2018

Synopsys’ Approach to Machine-Learning Initiatives Bringing “Digital Intelligence” to the Design Platform BETTER QOR, FASTER TTR ML-Enhanced Engines “NO-HUMAN-IN-THE-LOOP” PRODUCTIVITY New, ML-Driven Applications Power Optimization Flow ML Apps Tool/Flow Parameters Route Prediction Debug ML Apps Failure Prediction High-Σ Simulation Design ML Apps Exploration DRC Scheduling S/W Quality ML Apps Release, regression, triage May 1, 2018

Machine Learning: PT ECO Power Recovery Machine Learning for ECO Optimization Training High power circuits Low power circuits Equivalent 2018 Target 5X Training DB User generated/owned Design style specific Prediction Optimize power? 5-10X Faster Target 3-5X speedup by 2018 production May 1, 2018

New, ML-Driven Applications Addressing moving parts and manual steps in the broader design environment Tool and Flow Settings (max_util, placement bounds, ICG) Design Collateral (PVT selection, layer stack, track height) Foundry / Process (DRC, stack variability, DFM, patterning) Hierarchy (Floorplan, macro placement, IP integration) 2017 Study: High-Performance ARM Core Implementation Question: Which is the best setting for channel blockage density? May 1, 2018

Automatically Optimized multiple parameters concurrently Arm Next Generation Core Example Better QoR and Productivity with Machine Learning Full-flow Within the Tools 7nm Arm NextGen Core 3.5 GHz Compute Envelope Syn params Baseline, ARM NextGen Reference Flow, 4 engineers over 7 months Synthesis Calendar Envelope P&R params 1 engineer 3 days Placement 100 MHz higher FMAX 60% TNS reduction 3% congestion reduction Automatically Optimized multiple parameters concurrently May 1, 2018

The Road to Digital Intelligence Goal: Make Design SIMPLER More Predictive Models More Predictive Flows Design Assistance Design Process Automation Power optimization Routability & SI Layer assignment Memory & runtime Tool/flow parameters “Doomed” runs MCMM impact QoR evolution Risk assessment Anomaly detection Debug assistance Natural language Floorplan assessment Constraints, blockages Resource optimization Enterprise-level planning Predict tool outcomes Accelerate closure Predict flow outcomes Simplify decisions Influence outcomes Minimize risk Prescribe outcomes “No-human-in-the-loop” May 1, 2018

Summary May 1, 2018 Market and Design Leadership #1 market and segment leadership: Computing, Mobile, AI, Automotive and Memory Enabling 92% of 10/8/7nm designs and 6X more tapeouts exclusively with Synopsys Bold vision redefines Chip Design Systematic execution of our Fusion Technology Strategy redefines design Our exclusive Ansys partnership is benefiting all customers Innovation across DG products is leading the industry and enabling our customers Strong Investment in Machine Learning Machine-learning has emerged as a powerful technology for addressing high-complexity, high-cost challenges Synopsys is investing in ML-based solutions across the digital platform ML enhancements within the platform with impressive early results Entire new applications around the platform May 1, 2018

May 1, 2018