Combining High Level Synthesis and Floorplan Together EDA Lab, Tsinghua University Jinian Bian.

Slides:



Advertisements
Similar presentations
TOPIC : SYNTHESIS DESIGN FLOW Module 4.3 Verilog Synthesis.
Advertisements

Dan Lander Haru Yamamoto Shane Erickson (EE 201A Spring 2004)
ECE Synthesis & Verification - Lecture 2 1 ECE 667 Spring 2011 ECE 667 Spring 2011 Synthesis and Verification of Digital Circuits High-Level (Architectural)
Courtesy RK Brayton (UCB) and A Kuehlmann (Cadence) 1 Logic Synthesis Sequential Synthesis.
Software Process Models
1 Post RTL structures/flows targeting low power Srinivas R Jammula Intel Corporation Bangalore, India Naveen M Kumar Intel Corporation Bangalore, India.
Xing Wei, Wai-Chung Tang, Yu-Liang Wu Department of Computer Science and Engineering The Chinese University of HongKong
Improving Placement under the Constant Delay Model Kolja Sulimma 1, Ingmar Neumann 1, Lukas Van Ginneken 2, Wolfgang Kunz 1 1 EE and IT Department University.
High-Level Constructors and Estimators Majid Sarrafzadeh and Jason Cong Computer Science Department
FPGA Latency Optimization Using System-level Transformations and DFG Restructuring Daniel Gomez-Prado, Maciej Ciesielski, and Russell Tessier Department.
- 1 -  P. Marwedel, Univ. Dortmund, Informatik 12, 05/06 Universität Dortmund Hardware/Software Codesign.
Fuzzy Simulated Evolution for Power and Performance of VLSI Placement Sadiq M. Sait Habib Youssef Junaid A. KhanAimane El-Maleh Department of Computer.
Behavioral Synthesis Outline –Synthesis Procedure –Example –Domain-Specific Synthesis –Silicon Compilers –Example Tools Goal –Understand behavioral synthesis.
1 HW/SW Partitioning Embedded Systems Design. 2 Hardware/Software Codesign “Exploration of the system design space formed by combinations of hardware.
CS244-Introduction to Embedded Systems and Ubiquitous Computing Instructor: Eli Bozorgzadeh Computer Science Department UC Irvine Winter 2010.
Behavioral Design Outline –Design Specification –Behavioral Design –Behavioral Specification –Hardware Description Languages –Behavioral Simulation –Behavioral.
ECE Synthesis & Verification1 ECE 667 Spring 2011 Synthesis and Verification of Digital Systems Verification Introduction.
A Timing-Driven Soft-Macro Resynthesis Method in Interaction with Chip Floorplanning Hsiao-Pin Su 1 2 Allen C.-H. Wu 1 Youn-Long Lin 1 1 Department of.
Simulated-Annealing-Based Solution By Gonzalo Zea s Shih-Fu Liu s
Logic Design Outline –Logic Design –Schematic Capture –Logic Simulation –Logic Synthesis –Technology Mapping –Logic Verification Goal –Understand logic.
A Cost-Driven Lithographic Correction Methodology Based on Off-the-Shelf Sizing Tools.
林永隆 (Youn-Long Lin) Department of Computer Science National Tsing Hua University High-Level Synthesis of VLSIs THEDA Tsing Hua Electronic Design Automation.
Merging Synthesis With Layout For Soc Design -- Research Status Jinian Bian and Hongxi Xue Dept. Of Computer Science and Technology, Tsinghua University,
Layout-based Logic Decomposition for Timing Optimization Yun-Yin Lien* Youn-Long Lin Department of Computer Science, National Tsing Hua University, Hsin-Chu,
Triple Patterning Aware Detailed Placement With Constrained Pattern Assignment Haitong Tian, Yuelin Du, Hongbo Zhang, Zigang Xiao, Martin D.F. Wong.
Integration of Retiming with Architectural Floorplanning: A New Design Methodology for DSM Abdallah and Bassam Tabbara Profs: R.K.Brayton, A.R.Newton,
Universität Dortmund  P. Marwedel, Univ. Dortmund, Informatik 12, 2003 Hardware/software partitioning  Functionality to be implemented in software.
Hierarchical Physical Design Methodology for Multi-Million Gate Chips Session 11 Wei-Jin Dai.
MGR: Multi-Level Global Router Yue Xu and Chris Chu Department of Electrical and Computer Engineering Iowa State University ICCAD
Design methodology.
TM Efficient IP Design flow for Low-Power High-Level Synthesis Quick & Accurate Power Analysis and Optimization Flow JAN Asher Berkovitz Yaniv.
ISE. Tatjana Petrovic 249/982/22 ISE software tools ISE is Xilinx software design tools that concentrate on delivering you the most productivity available.
CRISP: Congestion Reduction by Iterated Spreading during Placement Jarrod A. Roy†‡, Natarajan Viswanathan‡, Gi-Joon Nam‡, Charles J. Alpert‡ and Igor L.
COE4OI5 Engineering Design. Copyright S. Shirani 2 Course Outline Design process, design of digital hardware Programmable logic technology Altera’s UP2.
CAD Techniques for IP-Based and System-On-Chip Designs Allen C.-H. Wu Department of Computer Science Tsing Hua University Hsinchu, Taiwan, R.O.C {
CAD for Physical Design of VLSI Circuits
Logic Synthesis for Low Power(CHAPTER 6) 6.1 Introduction 6.2 Power Estimation Techniques 6.3 Power Minimization Techniques 6.4 Summary.
A New Method For Developing IBIS-AMI Models
Section 10: Advanced Topics 1 M. Balakrishnan Dept. of Comp. Sci. & Engg. I.I.T. Delhi.
1 Wire Length Prediction-based Technology Mapping and Fanout Optimization Qinghua Liu Malgorzata Marek-Sadowska VLSI Design Automation Lab UC-Santa Barbara.
Budget-based Control for Interactive Services with Partial Execution 1 Yuxiong He, Zihao Ye, Qiang Fu, Sameh Elnikety Microsoft Research.
HDL-Based Layout Synthesis Methodologies Allen C.-H. Wu Department of Computer Science Tsing Hua University Hsinchu, Taiwan, R.O.C {
FPGA-Based System Design: Chapter 6 Copyright  2004 Prentice Hall PTR Topics n Design methodologies.
Building Information Systems & Managing Projects.
Massachusetts Institute of Technology 1 L14 – Physical Design Spring 2007 Ajay Joshi.
Chonnam national university VLSI Lab 8.4 Block Integration for Hard Macros The process of integrating the subblocks into the macro.
A Graph Based Algorithm for Data Path Optimization in Custom Processors J. Trajkovic, M. Reshadi, B. Gorjiara, D. Gajski Center for Embedded Computer Systems.
L11: Lower Power High Level Synthesis(2) 성균관대학교 조 준 동 교수
ASIC, Customer-Owned Tooling, and Processor Design Nancy Nettleton Manager, VLSI ASIC Device Engineering April 2000 Design Style Myths That Lead EDA Astray.
- 1 - EE898_HW/SW Partitioning Hardware/software partitioning  Functionality to be implemented in software or in hardware? No need to consider special.
ILP-Based Pin-Count Aware Design Methodology for Microfluidic Biochips Chiung-Yu Lin and Yao-Wen Chang Department of EE, NTU DAC 2009.
RF network in SoC1 SoC Test Architecture with RF/Wireless Connectivity 1. D. Zhao, S. Upadhyaya, M. Margala, “A new SoC test architecture with RF/wireless.
CS244-Introduction to Embedded Systems and Ubiquitous Computing Instructor: Eli Bozorgzadeh Computer Science Department UC Irvine Winter 2010.
1 - CPRE 583 (Reconfigurable Computing): VHDL to FPGA: A Tool Flow Overview Iowa State University (Ames) CPRE 583 Reconfigurable Computing Lecture 5: 9/7/2011.
6. A PPLICATION MAPPING 6.3 HW/SW partitioning 6.4 Mapping to heterogeneous multi-processors 1 6. Application mapping (part 2)
System-level power analysis and estimation September 20, 2006 Chong-Min Kyung.
Parallelization Strategies Laxmikant Kale. Overview OpenMP Strategies Need for adaptive strategies –Object migration based dynamic load balancing –Minimal.
Meenakshi Kaul, Vinoo Srinivasan, Sriram Govindarajan, Iyad Ouaiss, and Ranga Vemuri University of Cincinnati
A Design Flow for Optimal Circuit Design Using Resource and Timing Estimation Farnaz Gharibian and Kenneth B. Kent {f.gharibian, unb.ca Faculty.
Static Timing Analysis
Courtesy RK Brayton (UCB) and A Kuehlmann (Cadence) 1 Logic Synthesis Timing Optimization.
03/30/031 ECE Digital System Design & Synthesis Lecture Design Partitioning for Synthesis Strategies  Partition for design reuse  Keep related.
CAD for VLSI Ramakrishna Lecture#2.
Proximity Optimization for Adaptive Circuit Design Ang Lu, Hao He, and Jiang Hu.
Prediction of Interconnect Net-Degree Distribution Based on Rent’s Rule Tao Wan and Malgorzata Chrzanowska- Jeske Department of Electrical and Computer.
Architecture and Synthesis for Multi-Cycle Communication
Timing Optimization Andreas Kuehlmann
Sheqin Dong, Song Chen, Xianlong Hong EDA Lab., Tsinghua Univ. Beijing
Michele Santoro: Further Improvements in Interconnect-Driven High-Level Synthesis of DFGs Using 2-Level Graph Isomorphism Michele.
Department of Computer Science and Technology
Presentation transcript:

Combining High Level Synthesis and Floorplan Together EDA Lab, Tsinghua University Jinian Bian

Outlines Background Combining High Level Synthesis and Floorplan together The evolution of Combing HLS and Floorplan The Basic Structure of Combining HLS and Floorplan Conclusion and our future work

Background The progress of manufactory technique for silicon chips has been slowed down under traditional design methodology High performance / Low cost / Low power should be achieved by a more refined design, which means we need more powerful EDA tools

Background The traditional EDA methodology has been challenged The interconnect takes up most of resources: including area resource, delay resource, and power resource, etc. The traditional design flow may cause a long design time, low design quality, even design failure under nanometer design environment High performance / Low cost / Low power should be achieved by a more refined design, which means we need more powerful EDA tools

Traditional Design Flow High Level Synthesis Floorplan HLS Optimize OK ? FP Optimize OK ? Chip Optimize OK? N Y N Y N

Problems High Level Synthesis and Floorplan are based on different Optimization Model No Interconnect Information in High Level Synthesis No Behavior Information in Floorplan May cause a long design time, low design quality, even design failure

Forgoing Optimized Flow High Level Synthesis Meet Constraints ? Fast Floorplan HLS & Fast FP Optimized OK ? Final Floorplan Chip Optimized OK ? N Y Y N Y Y N Y

Forgoing optimized Flow High level synthesis using floorplan to get some physical information; Re-synthesis after floorplanning to optimize the design. But, Long loop time ; The estimation of Interconnect Information may far from final Chip ; Floorplan is passive.

Optimized Design Flow Interconnect Information can be got through a fast floorplan tool The estimation of Interconnect Information may far from final Chip

Proposed New Design Flow High Level SynthesisFloorplan Q/A The Chip Controller

Proposed Design Flow High Level Synthesis and Floorplan can communicate with each other The functionality of High Level Synthesis and Floorplan are both enhanced The result of allocation and scheduling can be restructured by Floorplan The Information of Interconnect can be retrieved easily The optimization of HLS and FP are based on a same estimation model Evolutional synthesis and floorplanning, to guarantee quicker astringency

Target Design Flow The main loop from High Level Synthesis and Floorplan can be avoid The optimizations of High Level Synthesis and Floorplan are consistent Can achieve a shorter running time of tools and a better quality of circuits

To Achieve the “ Target Design Flow ” 1. High Level Synthesis Oriented Physical Information Estimation 2. Floorplan Aware High Level Synthesis 3. Behavior Aware Floorplan 4. High Level Synthesis and Floorplan Oriented Parameterized Functional Unit Library

1. High Level Synthesis Oriented Physical Information Estimation HLS oriented high level physical information model and estimation technology Physical Information Estimation oriented fast floorplan and placement

1.1 HLS oriented physical information model and estimation technology of a module To create HLS oriented physical information model , through studying the existing information of physical design and IP core To study the relationship of logic struction vs. physical information (interconnect, power, conjestive, etc.) before phsical design

1.2 quick floorplanning technique to estimate physical information To get physical information with different accurate Techniques: Partitioning and clustering Quick floorplanning algorithm Placement information in a module

2. Floorplan Aware High Level Synthesis Floorplan aware High Level Synthesis Pre-partition before floorplan Interconnect aware HLS Analysis and Information retrieval of the result of Floorplan Constraints generation for Floorplan Incremental High Level Synthesis after Floorplan

Floorplan Aware High Level Synthesis Constraints from HLS to floorplan boundary constraints, adjacent constraints, separation constraints, delay constraints of each net, alternative modules for each functional unit, area constraints, etc. Constraints from floorplan to HLS delay constraints for each functional unit, area constraints for each functional unit, etc.

High Level Physical Information Estimation Scheduling and Allocation Floorplan Floorplan Constraints Generation FeedBack: HLS Constraints Generation Delay Constraint In HLS Adjacent Constraint In Floorplan Allocation Constraint In HLS

2.1. HLS techniques combined with floorplanning HLS algorithm considering physical info. of every module and the interconnect information between modules Get information from a module library with functionality and performance parameters and justify the design result Technology mapping and functional unit assignment

2.2 floorplanning constraint generation after HLS To transfer the information, requests and constraints of HLS to floorplanning E.g. group information, adjacent relationship, critical paths, etc.

2.3 Re-synthesis after floorplanning Adjust functional unit assignment and binding without florplan result to enhance the performance. Adjust schedule result to change the timing constraint To enhance the layout result.

3. Behavior Aware Floorplan Behavior aware Floorplan Partition and Clustor based on behavior information Floorplan under uncertain data Floorplan can change the result of allocation from HLS Incremental Floorplan after High Level Resynthesis

3.1. Constraints and behavior driven floorplanning To satisfy the circuit functionality and timing constaints Behavior constraints are as a guidance for floorplanning

3.2. floorplanning technique with incomplete information including : Undetermined module shapes or areas Undetermined pins , Undetermined module numbers Possible solutions Soft-module floorplanning techniques Shape-alterable polygons Unit and module mixed Etc.

3.3. Incremental floorplanning after synthesis To keep the basic structure and to keep the parameter unchanged.

4. HLS and Floorplan Oriented Parameterized FU Library Functional Unit Interface for both High Level Synthesis and Floorplan Provide prototype and instance of each functional unit Fast Estimation of physical information for un-stored functional unit

Conclusion and future work The main frame of Combining High Level Synthesis and Floorplan The structure of the parameterized functional unit library

Conclusion and future work Floorplan aware High Level Synthesis Floorplan under uncertain conditions Constraints and feedback generation and transfer between High Level Synthesis and Floorplan

Thank You!