Parallel Animated Image File Generation Nishad Patel Department of - - PowerPoint PPT Presentation

▶
parallel animated image file generation
SMART_READER_LITE
LIVE PREVIEW

Parallel Animated Image File Generation Nishad Patel Department of - - PowerPoint PPT Presentation

Mar 26, 2024 502 likes •778 views

Parallel Animated Image File Generation Nishad Patel Department of Computer Science Rochester Institute of Technology May 2012 Introduction Animated images Parallel generation Agenda Background Framework Hypotheses

slide-1
SLIDE 1

Parallel Animated Image File Generation

Nishad Patel Department of Computer Science Rochester Institute of Technology May 2012

slide-2
SLIDE 2

Introduction

  • Animated images
  • Parallel generation
slide-3
SLIDE 3

Agenda

  • Background
  • Framework
  • Hypotheses
  • Simulations
  • The File Format
  • Performance
  • Conclusions
slide-4
SLIDE 4

Background

Parallel Simulation and Visualization Modern-day simulations

  • Limited resources
  • Large data

– Study offline

  • Simulate over time – moving images
  • Difficulty - performance
slide-5
SLIDE 5

Background

File formats

  • Lossless compression
  • Few formats exist

– File size vs. Data Integrity

  • None support parallel generation
  • MNG – Multiple-image Network Graphics

– Lossless – High compression (Network-oriented) – No support for parallel generation

slide-6
SLIDE 6

Framework

  • PJGMovie

– int[][] matrix – PJGColorFrame

  • 24-bit Run-length encoding

– PJGHueFrame

  • 24-bit Huffman encoding

– PJGGrayFrame

  • 8-bit Huffman encoding
slide-7
SLIDE 7

Framework

  • write(frame)
  • writeColSlice(Range, frame)
  • writeRowSlice(Range, frame)
  • writePatch (Range, Range, frame)
  • getPixelDataSegmentParameters()
  • readSegment()
slide-8
SLIDE 8

Hypotheses

  • Serial PJM processing times vs. Parallel PJM processing times

– Parallel PJM should be quicker

  • Post-processing times vs. In-situ processing times

– In-situ processing should be quicker

  • PJM processing times vs. processing times for other formats

– PJM processing times will be smaller

  • PJM file sizes vs. file sizes for other formats

– PJM file sizes expected to be larger

slide-9
SLIDE 9

Simulations

Belushov-Zhabotinsky Reaction

  • Temporally oscillating chemical reaction
  • Reaction-diffusion system
  • Three substrate model
  • Solution in 2-D plane
  • Equations defined in (Turner, 2009)*

A + B -> 2A B + C -> 2B C + A -> 2C

*Turner, A. (March 2009). A Simple Model of the Belousov-Zhabotinsky Reaction from First Principles. Implemenation Note, University College London, Bartlett School of Graduate Studies, London

slide-10
SLIDE 10

Simulations

Belushov-Zhabotinsky Reaction

  • Reduced to

at+1 = at + at (bt − ct ) bt+1 = bt + bt (ct − at ) ct+1 = ct + ct (at − bt )

  • Diffusion effect

From Wikipedia en.wikipedia.org/wiki/File:The_Belousov-Zhabotinsky_Reaction.gif

slide-11
SLIDE 11

Simulations

Belushov-Zhabotinsky Reaction Parallelizing the program

  • Split frames into slices
slide-12
SLIDE 12

Simulations

Antiproton n-body simulation

  • All antiprotons have equal negative charges
  • Repulsive force between two antiprotons

– Directly proportional to product of charges – Inversely proportional to distance between them

  • Perpendicular magnetic field

– Forcing circular motion

  • Also in 2-D space
slide-13
SLIDE 13

Simulations

Antiproton n-body simulation Position calculation:

v' = v + δ a p' = p + δ v + 1/2 (δ*δ) a

slide-14
SLIDE 14

Simulations

Antiproton n-body simulation Parallelizing the program

  • Split the antiprotons over the processes
  • Each process stores

– Entire position array – Slice of velocity array – Slice of acceleration array

  • All-gather
slide-15
SLIDE 15

File Format

  • Header segment 0x00, 0x50, 0x4A, 0x4D, 0x00, 0x00, 0x00, 0x01
  • Image type segment
  • Height segment
  • Width segment
  • Number of Frames segment
  • Frame rate segment
  • Creation time segment
  • Comment segment
  • Run length encoded 24-bit color pixel data segment
  • Huffman delta encoded 8-bit grayscale pixel data segment
  • Huffman delta encoded 24-bit hue pixel data segment
slide-16
SLIDE 16

Performance

Serial PJM processing times vs. Parallel PJM processing times

NT T Speedup Serial 17753 1 19280 0.921 2 9225 1.924 3 6433 2.760 4 5160 3.441 8 3596 4.937 NT T Speedup Serial 78486 1 68876 1.140 2 35217 2.229 3 24214 3.241 4 18713 4.194 8 10541 7.446

1000 10000 1 10 Running Times (msecs) Number of Threads

Serial Post-processing vs Parallel Post- processing

200 x 200 10000 100000 1 10 Running Times (msecs) Number of Threads

Serial Post-processing vs Parallel Post- processing

400 x 400

B-Z Reaction 400 x 400 200 x 200

slide-17
SLIDE 17

Performance

Post-processing times vs. In-situ processing times

NT T TT Speedup 1 16129 43734 2.712 2 8634 22036 2.552 3 6206 15335 2.471 4 4982 12198 2.448 8 3299 7906 2.396

1000 10000 100000 1 10 Running Times (msecs) Number of Threads

Post-processing vs Insitu-processing

Post-processing Insitu-processing

NT T TT Speedup 1 74079 175886 2.374 2 36071 88119 2.442 3 24436 60197 2.463 4 18848 46599 2.472 8 11141 34181 3.068

10000 100000 1 10 Running Times (msecs) Number of Threads

Post-processing vs Insitu-processing

B-Z Reaction 200 x 200 400 x 400

slide-18
SLIDE 18

Performance

Post-processing times vs. In-situ processing times

NT T TT Speedup Serial 200182 202010 1.009 1 194127 192680 0.993 2 118412 120321 1.016 3 98029 100449 1.025 4 84279 85547 1.015 8 77303 78574 1.016

50000 1 10 Running Times (msecs) Number of Threads

Post-processing vs Insitu-processing

Post-processing Insitu-processing

Antiproton Simulation 400 Antiprotons

slide-19
SLIDE 19

Performance

PJM processing times vs. processing times for other formats

NT T Speedup MNG File 22527 1 16129 1.397 2 8634 2.609 3 6206 3.630 4 4982 4.521 8 3299 6.828

1000 10000 1 10 Running Times (msecs) Number of Threads

PJM vs MNG

PJM MNG

NT T Speedup MNG File 108769 1 74079 1.468 2 36071 3.015 3 24436 4.451 4 18848 5.771 8 11141 9.762

10000 100000 1 10 Number of Threads

PJM vs MNG

200 x 200 400 x 400 B-Z Reaction

slide-20
SLIDE 20

Performance

PJM processing times vs. processing times for other formats

NP T TM Speedup 1 54058 175979 3.255 2 41873 161936 3.867 3 39719 161859 4.076 4 41351 163864 3.963 8 53834 177052 3.289

10000 100000 1 10 Running Times (msecs) Number of Threads

PJM vs MNG

MNG PJM

Antiproton Simulation 200 Antiprotons

slide-21
SLIDE 21

Performance

PJM file sizes vs. File sizes for other formats

MNG bytes PJM bytes Larger by % 200 x 200 pixels 20901545 22578512 8.023 400 x 400 pixels 83593947 91148715 9.037

10000000 20000000 30000000 40000000 50000000 60000000 70000000 80000000 90000000 100000000 200 x 200 400 x 400 File size (bytes) Size of image

PJM vs MNG

MNG PJM

slide-22
SLIDE 22

MNG bytes MNG bytes PJM bytes Larger by % Larger by % Antiprotons Compression 1 Compression 9 Compression 1 Compression 9 200 420515 116927 241407

  • 42.593

106.457 400 451694 129942 277286

  • 38.612

113.921

50000 100000 150000 200000 250000 300000 350000 400000 450000 500000 200 400 File size (bytes) Number of Antiprotons (N)

PJM vs MNG

MNG 1 PJM MNG 9

Performance

PJM file sizes vs. File sizes for other formats

slide-23
SLIDE 23

Conclusions

  • Serial PJM processing times vs. Parallel PJM processing times

– Parallel PJM is quicker

  • Post-processing times vs. In-situ processing times

– In-situ processing is quicker

  • PJM processing times vs. processing times for other formats

– PJM processing times are smaller

  • PJM file sizes vs. file sizes for other formats

– PJM file sizes expected are larger - generally

slide-24
SLIDE 24

Future Work

  • Delta encoding
  • Player features
slide-25
SLIDE 25
slide-26
SLIDE 26

Thank you.

  • Questions?