StaticRoute: A novel router for the dynamic partial reconfiguration of FPGAs Brahim Al Farisi, Karel Bruneel, Dirk Stroobandt 2/9/2013.

Slides:



Advertisements
Similar presentations
Hybrid BDD and All-SAT Method for Model Checking Orna Grumberg Joint work with Assaf Schuster and Avi Yadgar Technion – Israel Institute of Technology.
Advertisements

Digital Design Copyright © 2006 Frank Vahid 1 FPGA Internals: Lookup Tables (LUTs) Basic idea: Memory can implement combinational logic –e.g., 2-address.
A Routing Technique for Structured Designs which Exploits Regularity Sabyasachi Das Intel Corporation Sunil P. Khatri Univ. of Colorado, Boulder.
Implementation Approaches with FPGAs Compile-time reconfiguration (CTR) CTR is a static implementation strategy where each application consists of one.
Floating-Point FPGA (FPFPGA) Architecture and Modeling (A paper review) Jason Luu ECE University of Toronto Oct 27, 2009.
38 th Design Automation Conference, Las Vegas, June 19, 2001 Creating and Exploiting Flexibility in Steiner Trees Elaheh Bozorgzadeh, Ryan Kastner, Majid.
Chapter 2 – Netlist and System Partitioning
Fast and Area-Efficient Phase Conflict Detection and Correction in Standard-Cell Layouts Charles Chiang, Synopsys Andrew B. Kahng, UC San Diego Subarna.
An Efficient Chiplevel Time Slack Allocation Algorithm for Dual-Vdd FPGA Power Reduction Yan Lin 1, Yu Hu 1, Lei He 1 and Vijay Raghunathan 2 1 EE Department,
EDA (CS286.5b) Day 14 Routing (Pathfind, netflow).
CS294-6 Reconfigurable Computing Day 14 October 7/8, 1998 Computing with Lookup Tables.
Leakage Efficient Chip-Level Dual-Vdd Assignment with Time Slack Allocation for FPGA Power Reduction Yan Lin and Lei He EE Department, UCLA Partially supported.
HARP: Hard-Wired Routing Pattern FPGAs Cristinel Ababei , Satish Sivaswamy ,Gang Wang , Kia Bazargan , Ryan Kastner , Eli Bozorgzadeh   ECE Dept.
EECC694 - Shaaban #1 lec #7 Spring The OSI Reference Model Network Layer.
CDCTree: Novel Obstacle-Avoiding Routing Tree Construction based on Current Driven Circuit Model Speaker: Lei He.
1 25\10\2010 Unit-V Connecting LANs Unit – 5 Connecting DevicesConnecting Devices Backbone NetworksBackbone Networks Virtual LANsVirtual LANs.
Field Programmable Gate Array (FPGA) Layout An FPGA consists of a large array of Configurable Logic Blocks (CLBs) - typically 1,000 to 8,000 CLBs per chip.
Performance and Power Efficient On-Chip Communication Using Adaptive Virtual Point-to-Point Connections M. Modarressi, H. Sarbazi-Azad, and A. Tavakkol.
1 A survey on Reconfigurable Computing for Signal Processing Applications Anne Pratoomtong Spring2002.
Connecting LANs, Backbone Networks, and Virtual LANs
RUN-TIME RECONFIGURATION FOR AUTOMATIC HARDWARE/SOFTWARE PARTITIONING Tom Davidson, Karel Bruneel, Dirk Stroobandt Ghent University, Belgium Presenting:
Juanjo Noguera Xilinx Research Labs Dublin, Ireland Ahmed Al-Wattar Irwin O. Irwin O. Kennedy Alcatel-Lucent Dublin, Ireland.
Register-Transfer (RT) Synthesis Greg Stitt ECE Department University of Florida.
High Performance Embedded Computing © 2007 Elsevier Lecture 16: Interconnection Networks Embedded Computing Systems Mikko Lipasti, adapted from M. Schulte.
Yehdhih Ould Mohammed Moctar1 Nithin George2 Hadi Parandeh-Afshar2
MGR: Multi-Level Global Router Yue Xu and Chris Chu Department of Electrical and Computer Engineering Iowa State University ICCAD
Automating Shift-Register-LUT Based Run-Time Reconfiguration Karel Heyse, Brahim Al Farisi, Karel Bruneel, Dirk Stroobandt
Networking Virtualization Using FPGAs Russell Tessier, Deepak Unnikrishnan, Dong Yin, and Lixin Gao Reconfigurable Computing Group Department of Electrical.
Overview Dynamic reconfiguration of FPGAs:
An automatic tool flow for the combined implementation of multi-mode circuits Brahim Al Farisi, Karel Bruneel, João Cardoso, Dirk Stroobandt.
Power Reduction for FPGA using Multiple Vdd/Vth
LOPASS: A Low Power Architectural Synthesis for FPGAs with Interconnect Estimation and Optimization Harikrishnan K.C. University of Massachusetts Amherst.
Paper Review: XiSystem - A Reconfigurable Processor and System
Un/DoPack: Re-Clustering of Large System-on-Chip Designs with Interconnect Variation for Low-Cost FPGAs Marvin Tom* Xilinx Inc.
Abhishek Pandey Reconfigurable Computing ECE 506.
Wen-Hao Liu 1, Yih-Lang Li 1, and Kai-Yuan Chao 2 1 Department of Computer Science, National Chiao-Tung University, Hsin-Chu, Taiwan 2 Intel Architecture.
VLSI Physical Design: From Graph Partitioning to Timing Closure Chapter 5: Global Routing © KLMH Lienig 1 EECS 527 Paper Presentation High-Performance.
THE TESTING APPROACH FOR FPGA LOGIC CELLS E. Bareiša, V. Jusas, K. Motiejūnas, R. Šeinauskas Kaunas University of Technology LITHUANIA EWDTW'04.
A Routing Approach to Reduce Glitches in Low Power FPGAs Quang Dinh, Deming Chen, Martin D. F. Wong Department of Electrical and Computer Engineering University.
Ping-Hung Yuh, Chia-Lin Yang, and Yao-Wen Chang
Design Space Exploration for Application Specific FPGAs in System-on-a-Chip Designs Mark Hammerquist, Roman Lysecky Department of Electrical and Computer.
Task Graph Scheduling for RTR Paper Review By Gregor Scott.
Rinoy Pazhekattu. Introduction  Most IPs today are designed using component-based design  Each component is its own IP that can be switched out for.
1 A Min-Cost Flow Based Detailed Router for FPGAs Seokjin Lee *, Yongseok Cheon *, D. F. Wong + * The University of Texas at Austin + University of Illinois.
Timing-Driven Routing for FPGAs Based on Lagrangian Relaxation
McGraw-Hill©The McGraw-Hill Companies, Inc., 2004 Connecting Devices CORPORATE INSTITUTE OF SCIENCE & TECHNOLOGY, BHOPAL Department of Electronics and.
Parallel Routing for FPGAs based on the operator formulation
Maze Routing Algorithms with Exact Matching Constraints for Analog and Mixed Signal Designs M. M. Ozdal and R. F. Hentschke Intel Corporation ICCAD 2012.
FPGA CAD 10-MAR-2003.
In-Place Decomposition for Robustness in FPGA Ju-Yueh Lee, Zhe Feng, and Lei He Electrical Engineering Dept., UCLA Presented by Ju-Yueh Lee Address comments.
High-Performance Global Routing with Fast Overflow Reduction Huang-Yu Chen, Chin-Hsiung Hsu, and Yao-Wen Chang National Taiwan University Taiwan.
Specialized Virtual Configurable Arrays Dominique Lavenier - Frederic Raimbault IRISA Rennes, France UBS Vannes, France
1 Field-programmable Gate Array Architectures and Algorithms Optimized for Implementing Datapath Circuits Andy Gean Ye University of Toronto.
Routing Wire Optimization through Generic Synthesis on FPGA Carry Hadi P. Afshar Joint work with: Grace Zgheib, Philip Brisk and Paolo Ienne.
FPGA Routing Pathfinder [Ebeling, et al., 1995] Introduced negotiated congestion During each routing iteration, route nets using shortest.
SEMI-SYNTHETIC CIRCUIT GENERATION FOR TESTING INCREMENTAL PLACE AND ROUTE TOOLS David GrantGuy Lemieux University of British Columbia Vancouver, BC.
Register-Transfer (RT) Synthesis Greg Stitt ECE Department University of Florida.
Congestion-Driven Re-Clustering for Low-cost FPGAs MASc Examination Darius Chiu Supervisor: Dr. Guy Lemieux University of British Columbia Department of.
Placing Relay Nodes for Intra-Domain Path Diversity Meeyoung Cha Sue Moon Chong-Dae Park Aman Shaikh Proc. of IEEE INFOCOM 2006 Speaker 游鎮鴻.
FPGA-Based System Design: Chapter 3 Copyright  2004 Prentice Hall PTR Topics n FPGA fabric architecture concepts.
A Study of the Scalability of On-Chip Routing for Just-in-Time FPGA Compilation Roman Lysecky a, Frank Vahid a*, Sheldon X.-D. Tan b a Department of Computer.
1 Architecture of Datapath- oriented Coarse-grain Logic and Routing for FPGAs Andy Ye, Jonathan Rose, David Lewis Department of Electrical and Computer.
Hybrid BDD and All-SAT Method for Model Checking
Interconnection topologies
2 University of California, Los Angeles
Chapter 2 – Netlist and System Partitioning
SAT-Based Area Recovery in Technology Mapping
ECE453 – Introduction to Computer Networks
Off-path Leakage Power Aware Routing for SRAM-based FPGAs
Chapter 3b Leakage Efficient Chip-Level Dual-Vdd Assignment with Time Slack Allocation for FPGA Power Reduction Prof. Lei He Electrical Engineering Department.
Presentation transcript:

StaticRoute: A novel router for the dynamic partial reconfiguration of FPGAs Brahim Al Farisi, Karel Bruneel, Dirk Stroobandt 2/9/2013

Dynamic partial reconfiguration (DPR) 1 Advantages: Smaller area Lower power usage M1M1 M2M2 M3M3 Goal: area reduction with reduced reconfiguration time M1M1 M2M2 M3M3 Disadvantage: Reconfiguration time

Conventional DPR tool flow Different circuits are implemented independently Complete area is rewritten  Problem: long reconfiguration times 2

Static vs dynamic bits After implementation: – Each memory cell corresponds to a collection of bit values – This collection of bit values is called a static bit, when the values are the same for all circuits a dynamic bit, otherwise 3

Clustering of dynamic bits Only memory cells that contain a dynamic bit need to be rewritten during run-time Configuration memory of an FPGA is frame-based Dynamic bits are scattered over the frames Approach in this work: – Divide configuration frames into dynamic and static ones – Cluster dynamic bits into the dynamic frames 4

CLBs vs routing In our experiments: – 10% of the configuration memory consists of CLB bits – 90% of the configuration memory consists of routing bits  Most of the time spent in reconfiguring the routing infrastructure Focus on reducing reconfiguration time of routing – All CLB frames are dynamic – Routing frames are divided in static and dynamic ones – Novel router, called StaticRoute, that clusters dynamic routing bits in the dynamic routing frames 5

Proposed DPR tool flow 6

PathFinder Makes use of a routing resource graph (RRG): directed graph where nodes represent wires and edges represent routing switches For each net PathFinder finds a minimum cost tree in the RRG In a first iteration nets are allowed to share resources, i.e. wire congestion is allowed Negotiated congestion Cost function of a node in the RRG: 7

StaticRoute Extended Pathfinder algorithm that also clusters dynamic routing bits into dynamic routing frames Makes use of an extended routing resource graph (eRRG): – switches are also represented by nodes – mark switches as static/dynamic – keep information about the switches during routing Detecting dynamic bits – After routing: easy – During routing: not obvious 8

Detecting dynamic bits 9 In general, in the extended RRG, a switch node S connects two wire nodes W in and W out. Let us assume that S is used by a set of circuits C S. W in and W out are used by Cin and C out respectively. We state that S is controlled by a dynamic bit if: ((C S ̸=C in ) ∨ (C S ̸=C out )) ∧ C S ̸=φ.

Novel cost function StaticRoute Switch congestion: when a static switch is controlled by a dynamic bit StaticRoute iterates until both wire and switch congestion are resolved Novel cost function: 10

Switch congestion penalty 11 p

Experiments 12 Regular expression matching, adaptive filtering, general MCNC and MCNC20 benchmarks LBs Considered only 2 circuits at a time Comparison of conventional DPR and new flow that uses StaticRoute Metrics: Reconfiguration time Wire length (of each circuit separately)

Results – Reconfiguration time 13

Results – Reconfiguration time 14

Results – Wire length 15

Conclusions 16 Possible to detect dynamic bits during routing Introduced notion of switch congestion Novel router, called StaticRoute, that resolves both wire and switch congestion Using novel DPR tool flow that uses StaticRoute: Total reconfiguration speed-up of approx. 2X Increase in wire length is limited

StaticRoute: A novel router for the dynamic partial reconfiguration of Brahim Al Farisi, Karel Bruneel, Dirk Stroobandt 2/9/2013