pipelining part 1
play

Pipelining (part 1) 1 Human pipeline: laundry whites sheets - PowerPoint PPT Presentation

Oct 17, 2022 •450 likes •1.11k views

Pipelining (part 1) 1 Human pipeline: laundry whites sheets sheets sheets colors colors colors whites whites whites colors colors colors whites whites 14:00 Washer 13:00 12:00 11:00 Table Folding Dryer Washer 14:00 13:00

  1. Pipelining (part 1) 1

  2. Human pipeline: laundry whites sheets sheets sheets colors colors colors whites whites whites colors colors colors whites whites 14:00 Washer 13:00 12:00 11:00 Table Folding Dryer Washer 14:00 13:00 12:00 11:00 Table Folding Dryer 2

  3. Human pipeline: laundry whites sheets sheets sheets colors colors colors whites whites whites colors colors colors whites whites 14:00 Washer 13:00 12:00 11:00 Table Folding Dryer Washer 14:00 13:00 12:00 11:00 Table Folding Dryer 2

  4. Waste (1) whites wasted time! wasted time! sheets sheets sheets colors colors colors whites Washer whites 14:00 13:00 12:00 11:00 Table Folding Dryer 3

  5. Waste (1) whites wasted time! wasted time! sheets sheets sheets colors colors colors whites Washer whites 14:00 13:00 12:00 11:00 Table Folding Dryer 3

  6. Waste (2) whites sheets sheets sheets colors colors colors whites whites Washer 14:00 13:00 12:00 11:00 Table Folding Dryer 4

  7. Latency — Time for One colors normal latency (1.8 h) colors colors colors pipelined latency (2.1 h) sheets sheets sheets colors colors Washer whites whites whites 14:00 13:00 12:00 11:00 Table Folding Dryer 5

  8. Latency — Time for One colors normal latency (1.8 h) colors colors colors pipelined latency (2.1 h) sheets sheets sheets colors colors Washer whites whites whites 14:00 13:00 12:00 11:00 Table Folding Dryer 5

  9. Latency — Time for One colors normal latency (1.8 h) colors colors colors pipelined latency (2.1 h) sheets sheets sheets colors colors Washer whites whites whites 14:00 13:00 12:00 11:00 Table Folding Dryer 5

  10. Throughput — Rate of Many colors time between starts (0.83 h) loads/h h load time between fjnishes (0.83 h) sheets sheets sheets colors colors Washer whites whites whites 14:00 13:00 12:00 11:00 Table Folding Dryer 6

  11. Throughput — Rate of Many Washer time between starts (0.83 h) time between fjnishes (0.83 h) sheets sheets sheets colors colors colors whites whites whites 14:00 13:00 12:00 11:00 Table Folding Dryer 6 1 load 0 . 83 h = 1 . 2 loads/h

  12. Throughput — Rate of Many Washer time between starts (0.83 h) time between fjnishes (0.83 h) sheets sheets sheets colors colors colors whites whites whites 14:00 13:00 12:00 11:00 Table Folding Dryer 6 1 load 0 . 83 h = 1 . 2 loads/h

  13. times three circuit 7 10 results/ns throughput 100 ps latency 100 ps 50 ps 0 ps 21 14 add add ADD ADD ADD ADD 7 A 2 × A 3 × A

  14. times three circuit 7 10 results/ns throughput 100 ps latency 100 ps 50 ps 0 ps 21 14 7 ADD ADD ADD ADD A 2 × A 3 × A A add A + A 2 × A add 2 A + A 3 × A

  15. times three circuit 7 100 ps 50 ps 0 ps 21 14 7 ADD ADD ADD ADD A 2 × A 3 × A 100 ps latency = ⇒ 10 results/ns throughput A add A + A 2 × A add 2 A + A 3 × A

  16. times three and repeat 2 21 17 34 51 4 8 12 1 3 7 23 46 69 0 ps 100 ps 200 ps 300 ps 400 ps 500 ps 14 add 8 2 7 14 17 34 4 8 add 1 23 46 0 ps 100 ps 200 ps 300 ps 400 ps 500 ps A add A + A 2 × A add 2 A + A 3 × A

  17. times three and repeat 2 21 17 34 51 4 8 12 1 3 7 23 46 69 0 ps 100 ps 200 ps 300 ps 400 ps 500 ps 14 8 2 23 7 14 17 34 4 8 1 46 0 ps 100 ps 200 ps 300 ps 400 ps 500 ps A add A + A 2 × A add 2 A + A 3 × A A add A + A 2 × A add 2 A + A 3 × A

  18. pipelined times three ( 34 17 17 21 14 7 7 ) ( ) ) ( ) ( ADD ADD ADD ADD 9 A ( t + 2 ) 2 × A ( t + 1 ) 3 × A ( t + 0 ) A ( t + 1 )

  19. pipelined times three 7 34 17 17 21 14 7 9 ADD ADD ADD ADD A ( t + 2 ) 2 × A ( t + 1 ) 3 × A ( t + 0 ) A ( t + 1 ) A ( t + 2 ) A ( t + 1 ) 2 × A ( t + 1 ) 3 × A ( t + 0 )

  20. register tolerances register output register input output changes input must not change register delay 10

  21. register tolerances register output register input output changes input must not change register delay 10

  22. register tolerances register output register input output changes input must not change register delay 10

  23. times three pipeline timing throughput: G operations/sec ps 11 ADD ADD ADD ADD A ( t + 2 ) 2 × A ( t + 1 ) 3 × A ( t + 0 ) A ( t + 1 ) 10 ps 50 ps 10 ps 50 ps 10 ps

  24. times three pipeline timing ADD throughput: 11 ADD ADD ADD A ( t + 2 ) 2 × A ( t + 1 ) 3 × A ( t + 0 ) A ( t + 1 ) 10 ps 50 ps 10 ps 50 ps 10 ps 1 60 ps ≈ 16 G operations/sec

  25. deeper pipeline ps Problem: Can we even do this? Problem: How much faster can we get? partial results partial results G ops/sec ps throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps G ops/sec ps throughput: exercise: throughput now? ps ps ADD ADD ADD ADD ps 12 ps ps ps ps ps A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 )

  26. deeper pipeline throughput: Problem: Can we even do this? Problem: How much faster can we get? partial results partial results G ops/sec ps throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps G ops/sec ps exercise: throughput now? 12 ADD ADD ADD ADD A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 ) 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps

  27. deeper pipeline throughput: Problem: Can we even do this? Problem: How much faster can we get? partial results partial results G ops/sec ps throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps G ops/sec ps exercise: throughput now? 12 ADD ADD ADD ADD A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 ) 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps

  28. deeper pipeline throughput: Problem: Can we even do this? Problem: How much faster can we get? partial results partial results G ops/sec ps throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps exercise: throughput now? 12 ADD ADD ADD ADD A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 ) 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 1 35 ps ≈ 28 G ops/sec

  29. deeper pipeline throughput: Problem: Can we even do this? Problem: How much faster can we get? partial results partial results G ops/sec ps throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps G ops/sec ps exercise: throughput now? 13 ADD ADD ADD ADD A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 ) 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps

  30. diminishing returns: register delays . 10 ps . . . . . . . logic (3/3) . . . . 1 ps 11 ps per cycle … 33 ps 10 ps logic (all) 33 ps 100 ps 110 ps per cycle 10 ps logic (1/2) 50 ps 60 ps per cycle 10 ps logic (2/2) 50 ps 10 ps logic (1/3) 33 ps 43 ps per cycle 10 ps logic (2/3) 14 10 ps 1 ps 10 ps 1 ps 10 ps 1 ps 10 ps

  31. diminishing returns: register delays number of stages time per completion (ps) 15 120 100 80 60 40 20 0 2 4 6 8 10 12 14

  32. diminishing returns: register delays register delay time per completion (ps) number of stages 15 120 100 80 60 40 20 0 2 4 6 8 10 12 14

  33. diminishing returns: register delays register delay time per completion (ps) number of stages 1.02x speedup 1.83x speedup 15 120 100 80 60 40 20 0 2 4 6 8 10 12 14

  34. diminishing returns: register delays 1.83x throughput throughput (ops/ns) number of stages 1.02x throughput 16 100 80 60 40 20 0 2 4 6 8 10 12 14

  35. diminishing returns: register delays 1.83x throughput throughput (ops/ns) number of stages max. rate of register updates 1.02x throughput 16 100 80 60 40 20 0 2 4 6 8 10 12 14

  36. deeper pipeline throughput: Problem: Can we even do this? Problem: How much faster can we get? partial results partial results G ops/sec ps throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps G ops/sec ps exercise: throughput now? 17 ADD ADD ADD ADD A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 ) 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps

  37. deeper pipeline throughput: Problem: Can we even do this? Problem: How much faster can we get? partial results partial results G ops/sec ps throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps G ops/sec ps exercise: throughput now? 18 ADD ADD ADD ADD A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 ) 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps

  38. deeper pipeline throughput: Problem: Can we even do this? Problem: How much faster can we get? partial results partial results throughput: exercise: throughput now? (didn’t split second add evenly) 20 ps 30 ps G ops/sec ps exercise: throughput now? 18 ADD ADD ADD ADD A ( t + 4 ) 2 × A 2 × A ( t + 2 ) 3 × A 3 × A ( t + 0 ) A ( t + 2 ) A ( t + 3 ) 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 25 ps 10 ps 1 40 ps ≈ 25 G ops/sec

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Pipelining Instruction Pipelining is the use of pipelining to allow more than one instruction to

Pipelining Instruction Pipelining is the use of pipelining to allow more than one instruction to

Pipelining Instruction Pipelining is the use of pipelining to allow more than one instruction to be in some stage of execution at the same time. Ferranti ATLAS (1963): Pipelining reduced the average time per instruction by

1.85k views • 147 slides
CS3350B Computer Organization Chapter 4: Instruction-Level Parallelism Part 1: Pipelining Alex

CS3350B Computer Organization Chapter 4: Instruction-Level Parallelism Part 1: Pipelining Alex

CS3350B Computer Organization Chapter 4: Instruction-Level Parallelism Part 1: Pipelining Alex Brandt Department of Computer Science University of Western Ontario, Canada Thursday March 7, 2019 Alex Brandt Chapter 4: ILP , Part 1: Pipelining

1.29k views • 116 slides
Lecture 2 (I ): Lecture 2 (I ): Pipelining & Retiming Pipelining & Retiming

Lecture 2 (I ): Lecture 2 (I ): Pipelining & Retiming Pipelining & Retiming

Lecture 2 (I ): Lecture 2 (I ): Pipelining & Retiming Pipelining & Retiming Hsie-Chia Chang E-mail : hcchang@mail.nctu.edu.tw Fall 2006 Outline Outline Pipelining of FI R Digital filters Data-Broadcast Structures

453 views • 28 slides
Computer Systems Lecture 15 Pipelining and Hazards CS 230 - Spring 2020 3-1 Pipelining CS

Computer Systems Lecture 15 Pipelining and Hazards CS 230 - Spring 2020 3-1 Pipelining CS

CS 230 Introduction to Computers and Computer Systems Lecture 15 Pipelining and Hazards CS 230 - Spring 2020 3-1 Pipelining CS 230 - Spring 2020 3-2 ISA Requirements for Pipelining Constant length instructions (fetch, decode)

628 views • 38 slides
Pipelining 1 Today Quiz Introduction to pipelining 2 Pipelining L L a a Logic

Pipelining 1 Today Quiz Introduction to pipelining 2 Pipelining L L a a Logic

Pipelining 1 Today Quiz Introduction to pipelining 2 Pipelining L L a a Logic (10ns) t t c c 10ns L L a a Logic (10ns) t t c c 20ns L L a a Logic (10ns) t t c c 30ns Whats the latency for one unit of

706 views • 27 slides
EE 457 Unit 6a Basic Pipelining Techniques 2 Pipelining Introduction Consider a drink

EE 457 Unit 6a Basic Pipelining Techniques 2 Pipelining Introduction Consider a drink

1 EE 457 Unit 6a Basic Pipelining Techniques 2 Pipelining Introduction Consider a drink bottling plant Filling the bottle = 3 sec. Placing the cap = 3 sec. Labeling = 3 sec. Would you want Machine 1 = Does all three (9

776 views • 45 slides
Overview General Principles of Pipelining Goal Computer Architecture: Pipelining

Overview General Principles of Pipelining Goal Computer Architecture: Pipelining

Overview General Principles of Pipelining Goal Computer Architecture: Pipelining Difficulties Creating a Pipelined Y86-64 Processor CSci 2021: Machine Architecture and Organization Rearranging SEQ March 25th-27th, 2020 Inserting

558 views • 6 slides
Pipelining is Hazardous! Hazards are situations where pipelining does not work as elegantly as

Pipelining is Hazardous! Hazards are situations where pipelining does not work as elegantly as

Pipelining is Hazardous! Hazards are situations where pipelining does not work as elegantly as we would like Three kinds Structural hazards -- we have run out of a hardware resource. Data hazards -- an input is not available on

356 views • 22 slides
Pipelining Raul Queiroz Feitosa Parts of these slides are from the support material provided by

Pipelining Raul Queiroz Feitosa Parts of these slides are from the support material provided by

Pipelining Raul Queiroz Feitosa Parts of these slides are from the support material provided by W. Stallings Objective To present the Pipelining concept, its limitations and the techniques for performance optimization 2 Pipelining Outline

736 views • 30 slides
Overview Basics of Pipelining Pipeline Hazards Appendix A Pipeline Implementation

Overview Basics of Pipelining Pipeline Hazards Appendix A Pipeline Implementation

Overview Basics of Pipelining Pipeline Hazards Appendix A Pipeline Implementation p p Pipelining + Exceptions Pipelining: Basic and Intermediate Pipeline to handle Multicycle Operations Concepts 1 2 Unpipelined

507 views • 14 slides
Pipelining PIPELINING what Seymour Cray taught the laundry industry How to correctly pipeline

Pipelining PIPELINING what Seymour Cray taught the laundry industry How to correctly pipeline

Pipelining PIPELINING what Seymour Cray taught the laundry industry How to correctly pipeline circuits Ive got 3 months Worth of laundry Funny, considering that hes only got To do tonight one outfit Acknowledgement: The

566 views • 7 slides
Appendix A Pipelining: Basic and Intermediate C Concepts t 1 Overview Basics of

Appendix A Pipelining: Basic and Intermediate C Concepts t 1 Overview Basics of

Appendix A Pipelining: Basic and Intermediate C Concepts t 1 Overview Basics of Pipelining Pipeline Hazards Pipeline Implementation Pipelining + Exceptions Pipeline to handle Multicycle Operations p y p 2 1

509 views • 40 slides
1 What Limits Performance? Stalls (Data Hazards) Data hazards Code Instruction depends on

1 What Limits Performance? Stalls (Data Hazards) Data hazards Code Instruction depends on

Instruction Scheduling Background: Pipelining Basics Last time Idea Register allocation Begin executing an instruction before completing the previous one Today Without Pipelining With Pipelining Instruction scheduling The

389 views • 7 slides
Unit 5: Pipelining Load-use stalling Pipelined multi-cycle operations Control hazards

Unit 5: Pipelining Load-use stalling Pipelined multi-cycle operations Control hazards

This Unit: Pipelining App App App Single-cycle & multi-cycle datapaths System software Latency vs throughput & performance Basic pipelining CIS 501: Computer Architecture Mem CPU I/O Data hazards Bypassing Unit

522 views • 24 slides
Computer Architecture Summer 2020 Pipelining Tyler Bletsch Duke University Includes material

Computer Architecture Summer 2020 Pipelining Tyler Bletsch Duke University Includes material

ECE/CS 250 Computer Architecture Summer 2020 Pipelining Tyler Bletsch Duke University Includes material adapted from Dan Sorin (Duke) and Amir Roth (Penn). This Unit: Pipelining Basic Pipelining Application Pipeline control OS

821 views • 68 slides
Basic Pipelining Wrap-up Exploiting ILP (from Slide Set 20) Chapter 6 and beyond 1 2

Basic Pipelining Wrap-up Exploiting ILP (from Slide Set 20) Chapter 6 and beyond 1 2

Slide Set #21: Basic Pipelining Wrap-up Exploiting ILP (from Slide Set 20) Chapter 6 and beyond 1 2 Pipelining Big Picture Remember the single-cycle implementation Improve performance by increasing instruction throughput

392 views • 5 slides
Thou Shalt Abstract Higher-Order List Functions Theory of Programming Languages Computer Science

Thou Shalt Abstract Higher-Order List Functions Theory of Programming Languages Computer Science

Introduction Mapping Filtering Folding Thou Shalt Abstract Higher-Order List Functions Theory of Programming Languages Computer Science Department Wellesley College Introduction Mapping Filtering Folding Table of contents Introduction

269 views • 11 slides
What is the largest topless trapezoid that can be made by folding up the ends of a 30 in

What is the largest topless trapezoid that can be made by folding up the ends of a 30 in

Read the problem Formul Cut down independent variables Domain Calculus (interior) Calculus (boundary) Extrema What is the largest topless trapezoid that can be made by folding up the ends of a 30 in length of wire? Read the problem

522 views • 41 slides
Welcome to... Meal Plan #046 @stemettes | #stemillions SNACK CACKLE POP @stemettes |

Welcome to... Meal Plan #046 @stemettes | #stemillions SNACK CACKLE POP @stemettes |

Welcome to... Meal Plan #046 @stemettes | #stemillions SNACK CACKLE POP @stemettes | #stemillions Meet Zaha Hadid... Meet Her: Dame Zaha Mohammad Hadid RA was an Iraqi-British architect. She was the first woman to receive the Pritzker

717 views • 18 slides
On Simulated Annealing in EDA A tribute to Prof. C. L. Liu at ISPD 2012 Martin D.F. Wong

On Simulated Annealing in EDA A tribute to Prof. C. L. Liu at ISPD 2012 Martin D.F. Wong

On Simulated Annealing in EDA A tribute to Prof. C. L. Liu at ISPD 2012 Martin D.F. Wong Department of Electrical and Computer Engineering University of Illinois at Urbana Champaign 1 ICCAD Panel on Simulated Annealing 1987 ICCAD Panel:

574 views • 30 slides
Fundamentals of Sigma-Delta ADCs Jinseok Koh, Ph.D. Texas Instruments Inc. Dallas, TX IEEE SSCS

Fundamentals of Sigma-Delta ADCs Jinseok Koh, Ph.D. Texas Instruments Inc. Dallas, TX IEEE SSCS

Fundamentals of Sigma-Delta ADCs Jinseok Koh, Ph.D. Texas Instruments Inc. Dallas, TX IEEE SSCS Dallas Chapter, June 2007 Jinseok Koh (jinseok@ti.com) Why Analog to Digital Conversion? Naturally occurring signals are analog signal

792 views • 45 slides
Collaborating within and across the organizational boundary of a Synchrotron facility: Innovation

Collaborating within and across the organizational boundary of a Synchrotron facility: Innovation

Collaborating within and across the organizational boundary of a Synchrotron facility: Innovation at structural folds? Florian Stache, Management Department Freie Universitt Berlin Andrey Indukaev, IDHES cole Normale Superieure (France)

600 views • 20 slides
ENERGY LANDSCAPES AND FOLDING KINETICS OF PAIRWISE INTERACTING RNAS Stefan Badelt 1 , Christoph

ENERGY LANDSCAPES AND FOLDING KINETICS OF PAIRWISE INTERACTING RNAS Stefan Badelt 1 , Christoph

ENERGY LANDSCAPES AND FOLDING KINETICS OF PAIRWISE INTERACTING RNAS Stefan Badelt 1 , Christoph Flamm , Ivo L. Hofacker 2 2 (1) DNA and Natural Algorithms Group, California Institute of Technology (2) Theoretical Biochemistry Group (tbi),

302 views • 27 slides
Protein molecular dynamics on OSG using CHARMM Ana Damjanovic, Tim Miller, Bernard Brooks (NIH)

Protein molecular dynamics on OSG using CHARMM Ana Damjanovic, Tim Miller, Bernard Brooks (NIH)

Protein molecular dynamics on OSG using CHARMM Ana Damjanovic, Tim Miller, Bernard Brooks (NIH) Petar Maksimovic (JHU) Wensheng Deng, Torre Wenaus (BNL), Frank Wuerthwein (UCSD) Petar Maksimovic, CHARMM @ OSG Why Molecular Dynamics (MD)

669 views • 29 slides

More recommend