An Active Glitch Elimination Technique for FPGAs

Slides:



Advertisements
Similar presentations
1/48 ENERGY OPTIMIZATION TECHNIQUES: FPGA GLITCH REDUCTION Patrick Cooke and Elizabeth Graham.
Advertisements

June 6, Using Negative Edge Triggered FFs to Reduce Glitching Power in FPGA Circuits Tomasz S. Czajkowski and Stephen D. Brown Department of Electrical.
EELE 367 – Logic Design Module 2 – Modern Digital Design Flow Agenda 1.History of Digital Design Approach 2.HDLs 3.Design Abstraction 4.Modern Design Steps.
Reducing the Pressure on Routing Resources of FPGAs with Generic Logic Chains Hadi P. Afshar Joint work with: Grace Zgheib, Philip Brisk and Paolo Ienne.
Architecture Design Methodology. 2 The effects of architecture design on metrics:  Area (cost)  Performance  Power Target market:  A set of application.
Clustering of Large Designs for Channel-Width Constrained FPGAs Marvin TomGuy Lemieux University of British Columbia Department of Electrical and Computer.
CMOS Circuit Design for Minimum Dynamic Power and Highest Speed Tezaswi Raja, Dept. of ECE, Rutgers University Vishwani D. Agrawal, Dept. of ECE, Auburn.
Yan Lin, Fei Li and Lei He EE Department, UCLA
ENGIN112 L38: Programmable Logic December 5, 2003 ENGIN 112 Intro to Electrical and Computer Engineering Lecture 38 Programmable Logic.
Dec. 6, 2005ELEC Glitch Power1 Low power design: Insert delays to eliminate glitches Yijing Chen Dec.6, 2005 Auburn university.
Lecture 3: Field Programmable Gate Arrays II September 10, 2013 ECE 636 Reconfigurable Computing Lecture 3 Field Programmable Gate Arrays II.
Architecture and Synthesis for Power-Efficient FPGAs Jason Cong University of California, Los Angeles Partially supported by NSF Grants.
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.
9/27/05ELEC / Lecture 91 ELEC / (Fall 2005) Special Topics in Electrical Engineering Low-Power Design of Electronic Circuits.
CS 151 Digital Systems Design Lecture 38 Programmable Logic.
Yehdhih Ould Mohammed Moctar1 Nithin George2 Hadi Parandeh-Afshar2
156 / MAPLD 2005 Rollins 1 Reducing Energy in FPGA Multipliers Through Glitch Reduction Nathan Rollins and Michael J. Wirthlin Department of Electrical.
Lecture 2: Field Programmable Gate Arrays September 13, 2004 ECE 697F Reconfigurable Computing Lecture 2 Field Programmable Gate Arrays.
Lecture #3 Page 1 ECE 4110– Sequential Logic Design Lecture #3 Agenda 1.FPGA's 2.Lab Setup Announcements 1.No Class Monday, Labor Day Holiday 2.HW#2 assigned.
Power Reduction for FPGA using Multiple Vdd/Vth
FPGA Switch Block Design Dr. Philip Brisk Department of Computer Science and Engineering University of California, Riverside CS 223.
Coarse and Fine Grain Programmable Overlay Architectures for FPGAs
Un/DoPack: Re-Clustering of Large System-on-Chip Designs with Interconnect Variation for Low-Cost FPGAs Marvin Tom* Xilinx Inc.
Channel Width Reduction Techniques for System-on-Chip Circuits in Field-Programmable Gate Arrays Marvin Tom University of British Columbia Department of.
Julien Lamoureux and Steven J.E Wilton ICCAD
Design Space Exploration for Application Specific FPGAs in System-on-a-Chip Designs Mark Hammerquist, Roman Lysecky Department of Electrical and Computer.
Impact of Interconnect Architecture on VPSAs (Via-Programmed Structured ASICs) Usman Ahmed Guy Lemieux Steve Wilton System-on-Chip Lab University of British.
FPGA Global Routing Architecture Dr. Philip Brisk Department of Computer Science and Engineering University of California, Riverside CS 223.
Lecture #3 Page 1 ECE 4110–5110 Digital System Design Lecture #3 Agenda 1.FPGA's 2.Lab Setup Announcements 1.HW#2 assigned Due.
Configuration Bitstream Reduction for SRAM-based FPGAs by Enumerating LUT Input Permutations The University of British Columbia© 2011 Guy Lemieux Ameer.
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
EE5970 Computer Engineering Seminar Spring 2012 Michigan Technological University Based on: A Low-Power FPGA Based on Autonomous Fine-Grain Power Gating.
1 Carnegie Mellon University Center for Silicon System Implementation An Architectural Exploration of Via Patterned Gate Arrays Chetan Patel, Anthony Cozzie,
A Synthesizable Datapath-Oriented Programmable Logic Core Steven J.E. Wilton, Chun Hok Ho, Philip Leong, Wayne Luk, Brad Quinton University of British.
Directional and Single-Driver Wires in FPGA Interconnect Guy Lemieux Edmund LeeMarvin TomAnthony Yu Dept. of ECE, University of British Columbia Vancouver,
An Improved “Soft” eFPGA Design and Implementation Strategy
March 28, Glitch Reduction for Altera Stratix II devices Tomasz S. Czajkowski PhD Candidate University of Toronto Supervisor: Professor Stephen D.
1 Area-Efficient FPGA Logic Elements: Architecture and Synthesis Jason Anderson and Qiang Wang 1 IEEE/ACM ASP-DAC Yokohama, Japan January 26-28,
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.
1 Field-programmable Gate Array Architectures and Algorithms Optimized for Implementing Datapath Circuits Andy Gean Ye University of Toronto.
FPGA Logic Cluster Design Dr. Philip Brisk Department of Computer Science and Engineering University of California, Riverside CS 223.
Routing Wire Optimization through Generic Synthesis on FPGA Carry Hadi P. Afshar Joint work with: Grace Zgheib, Philip Brisk and Paolo Ienne.
Architecture and algorithm for synthesizable embedded programmable logic core Noha Kafafi, Kimberly Bozman, Steven J. E. Wilton 2003 Field programmable.
ECE 506 Reconfigurable Computing Lecture 5 Logic Block Architecture Ali Akoglu.
Interconnect Driver Design for Long Wires in FPGAs Edmund Lee, Guy Lemieux & Shahriar Mirabbasi University of British Columbia, Canada Electrical & Computer.
Chandrasekhar 1 MAPLD 2005/204 Reduced Triple Modular Redundancy for Tolerating SEUs in SRAM based FPGAs Vikram Chandrasekhar, Sk. Noor Mahammad, V. Muralidharan.
Reconfigurable Computing - Performance Issues John Morris Chung-Ang University The University of Auckland ‘Iolanthe’ at 13 knots on Cockburn Sound, Western.
Time-borrowing platform in the Xilinx UltraScale+ family of FPGAs and MPSoCs Ilya Ganusov, Benjamin Devlin.
Topics SRAM-based FPGA fabrics: Xilinx. Altera..
IPF: In-Place X-Filling to Mitigate Soft Errors in SRAM-based FPGAs
CS184a: Computer Architecture (Structure and Organization)
MAPLD 2005 Reduced Triple Modular Redundancy for Tolerating SEUs in SRAM based FPGAs Vikram Chandrasekhar, Sk. Noor Mahammad, V. Muralidharan Dr. V. Kamakoti.
ECE 4110–5110 Digital System Design
CS184a: Computer Architecture (Structure and Organization)
From Silicon to Microelectronics Yahya Lakys EE & CE 200 Fall 2014
Overview Part 1 – The Design Space
ELEN 468 Advanced Logic Design
A. Mishchenko S. Chatterjee1 R. Brayton UC Berkeley and Intel1
Multiple Drain Transistor-Based FPGA Architectures
COMBINATIONAL LOGIC.
FPGA Glitch Power Analysis and Reduction
ESE534: Computer Organization
Give qualifications of instructors: DAP
EE216A – Fall 2010 Design of VLSI Circuits and Systems
Measuring the Gap between FPGAs and ASICs
Arithmetic Building Blocks
Chapter 3b Leakage Efficient Chip-Level Dual-Vdd Assignment with Time Slack Allocation for FPGA Power Reduction Prof. Lei He Electrical Engineering Department.
Programmable logic and FPGA
Reconfigurable Computing (EN2911X, Fall07)
Presentation transcript:

An Active Glitch Elimination Technique for FPGAs by Julien Lamoureux, Guy Lemieux, and Steven J.E. Wilton University of British Columbia Funding provided by Altera and NSERC

Context of this Research FPGAs use a lot of power Dynamic power still dominates Source: Altera Stratix II Study (99 Circuits)

Overview Reducing power in FPGAs by minimizing glitching Using programmable delay circuits to align arrival times Results: 18% power savings 5% area overhead 1% critical-path delay overhead No changes to the existing CAD flow

What is glitching? Unnecessary transitions Generated by uneven arrival times Propagated by certain gates

How much glitching is there? Circuit Switching Activity Functional Glitching % Glitching C1355 0.319 0.231 0.088 27.5 C1908 0.255 0.167 34.6 C2670 0.267 0.208 0.059 22.2 C3540 0.419 0.230 0.189 45.2 C432 0.260 0.184 0.076 29.3 C499 0.341 0.232 0.109 31.9 C5315 0.400 0.253 0.147 36.7 C6288 1.562 0.295 1.267 81.1 C7552 0.392 0.228 0.165 42.0 C880 0.186 0.046 19.8 alu4 0.081 0.070 0.011 13.1 apex2 0.049 0.042 0.007 13.7 apex4 0.044 0.030 0.014 32.3 des 0.169 0.098 36.8 ex1010 0.032 0.015 0.017 52.9 ex5p 0.168 0.082 0.086 51.0 misex3 0.064 0.050 0.013 20.9 pdc 0.035 0.024 31.8 seq 0.048 0.040 0.008 16.0 spla 0.028 0.021 42.7 Average - 34.1

Context of this Research Circuit Switching Activity Functional Glitching % Glitching C1355 0.319 0.231 0.088 27.5 C1908 0.255 0.167 34.6 C2670 0.267 0.208 0.059 22.2 C3540 0.419 0.230 0.189 45.2 C432 0.260 0.184 0.076 29.3 C499 0.341 0.232 0.109 31.9 C5315 0.400 0.253 0.147 36.7 C6288 1.562 0.295 1.267 81.1 C7552 0.392 0.228 0.165 42.0 C880 0.186 0.046 19.8 alu4 0.081 0.070 0.011 13.1 apex2 0.049 0.042 0.007 13.7 apex4 0.044 0.030 0.014 32.3 des 0.169 0.098 36.8 ex1010 0.032 0.015 0.017 52.9 ex5p 0.168 0.082 0.086 51.0 misex3 0.064 0.050 0.013 20.9 pdc 0.035 0.024 31.8 seq 0.048 0.040 0.008 16.0 spla 0.028 0.021 42.7 Average - 34.1 1/3 of Dynamic Power!

Idea Use programmable delay circuits to lineup arrival times.

Idea Use programmable delay circuits to lineup arrival times.

ASIC vs. FPGA ASIC Circuit and delays are known before fabrication Fixed delay circuits can be used FPGA Circuit and delays are unknown Delay circuits needed to be programmable Location of delays must be carefully considered

Where should the delays go? Option 1: Global Routing Option 2: Logic Blocks (LABs)

Where in the LAB?

Where in the LAB?

Where in the LAB? Too Expensive?

4 Schemes

Programmable Delay Circuit

Programmable Delay Circuit Biased PMOS Biased NMOS

Programmable Delay Circuit Biased PMOS Biasing Circuit Biased NMOS

Calibrating Scheme 1

Calibrating Scheme 1 Three parameters: Number of delay circuits per LUT Maximum delay of the delay circuit Minimum delay of the delay circuit

Number of delay circuits per LUT

Number of delay circuits per LUT

Number of delay circuits per LUT

Number of delay circuits per LUT

Number of delay circuits per LUT K-1

Maximum Delay 4ns

Maximum Delay 6ns

Maximum Delay 8ns

Maximum Delay 8ns

Minimum Delay Increment 100ps

Minimum Delay Increment 200ps

Minimum Delay Increment 300ps

Minimum Delay Increment 250ps

Glitch Elimination Results Circuits % Glitch Elimination Scheme 1 C1355 92.5 C1908 91.2 C2670 96.3 C3540 84.7 C432 62.8 C499 94.5 C5315 83.9 C6288 62.7 C7552 89.6 C880 84.6 alu4 96.6 apex2 97.0 apex4 95.0 des 86.4 ex1010 ex5p 82.7 misex3 97.2 pdc 89.3 seq 96.9 spla 95.9 Average 88.5

Glitch Elimination Results Circuits % Glitch Elimination Scheme 1 Scheme 2 Scheme 3 Scheme 4 C1355 92.5 92.2 89.6 C1908 91.2 88.4 74.2 93.2 C2670 96.3 90.5 77.3 96.1 C3540 84.7 73.3 74.4 83.9 C432 62.8 65.8 63.4 62.0 C499 94.5 94.1 C5315 71.5 69.6 82.3 C6288 62.7 59.2 56.6 48.8 C7552 76.3 77.8 90.0 C880 84.6 80.5 81.1 75.7 alu4 96.6 85.6 96.2 apex2 97.0 86.5 85.9 apex4 95.0 91.6 88.2 des 86.4 70.3 75.6 87.2 ex1010 83.0 79.9 89.4 ex5p 82.7 74.5 71.4 82.2 misex3 97.2 88.8 87.9 pdc 89.3 87.1 seq 96.9 88.9 91.5 96.7 spla 95.9 90.6 91.7 Average 88.5 81.3 79.8

Overhead

Overhead Area count minimum width transistor areas 5.3 %

Overhead Area Tcrit count minimum width transistor areas 5.3 % VPR delay + HSPICE delay circuit 0.21 %

Overhead Area Tcrit Power count minimum width transistor areas 5.3 % VPR delay + HSPICE delay circuit 0.21 % Power VPR power + HSPICE delay circuits 0.45 %

Overall Results

Final Power Savings Circuits % Glitch Elimination Ideal Scheme 1 C1355 28.8 26.7 C1908 21.1 17.6 C2670 13.4 12.1 C3540 31.7 26.5 C432 17.1 11.2 C499 34.6 33.0 C5315 22.8 19.2 C6288 73.1 46.3 C7552 25.5 22.6 C880 9.6 7.8 alu4 3.6 3.3 apex2 4.3 4.1 apex4 10.1 9.5 des 17.9 15.4 ex1010 18.4 ex5p 28.1 25.4 misex3 8.1 pdc 13.3 11.8 seq 6.1 5.9 spla 21.4 20.8 Average 18.0

Summary Proposed an active glitch elimination technique for FPGAs Examined how to implement the technique Reduced power by 18% with only 5% area and 1% speed Proposed technique requires little or no modifications to the CAD flow or routing architecture