1 / 7

Enhancing Performance of Conway’s Game of Life through Parallelization Techniques

Explore the intricacies of Conway's Game of Life in a simple 2D universe, where basic rules lead to complex and fascinating results. This project requires improving performance via parallelization using pthreads, ideally employing at least 4 threads to maximize efficiency on multi-core systems. Implement various optimizations, including hand-coded enhancements, cache optimizations, synchronization mechanisms, and improved cell representation, to eliminate redundancy and unnecessary work. Experience the evolution of interesting patterns such as Gliders, Pulsars, and Gosper’s Glider Gun, while ensuring the board wraps around itself.

byron
Download Presentation

Enhancing Performance of Conway’s Game of Life through Parallelization Techniques

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. HW5: Parallelization

  2. Conway’s Game of Life • Simple 2D universe with simple rules, complex results! • 1. Any live cell with fewer than two live neighbors dies • 2. Any live cell with more than three live neighbors dies • 3. Any live cell with two or three live neighbors lives on • 4. Any dead cell with three live neighbors becomes alive

  3. Interesting Stable Patterns Glider LightWeight Space Ship Pulsar

  4. “Gosper’s glider gun”

  5. Typical Random Initial Universe DEMO NOTE: board wraps around (top/bottom, L/R)

  6. Key Part of the Code: • for (curgen = 0; curgen < gens_max; curgen++) • for (i = 0; i < nrows; i++){ • for (j = 0; j < ncols; j++){ • intinorth = mod (i-1, nrows); intisouth = mod (i+1, nrows); • intjwest = mod (j-1, ncols); intjeast = mod (j+1, ncols); • const char neighbor_count = • BOARD (inboard, inorth, jwest) + BOARD (inboard, inorth, j) + • BOARD (inboard, inorth, jeast) + BOARD (inboard, i, jwest) + • BOARD (inboard, i, jeast) + BOARD (inboard, isouth, jwest) + • BOARD (inboard, isouth, j) + BOARD (inboard, isouth, jeast); • BOARD(outboard, i, j) = alivep (neighbor_count, BOARD (inboard, i, j)); • } • } • SWAP_BOARDS( outboard, inboard ); • }

  7. HW5: Your Goal • Improve performance via parallelization • use pthreads • use at least 2 threads (probably 4 since 4 CPUs on UG*) • Further improve performance via other optimizations by putting what we learnt all together • hand code optimizations • cache optimizations • locks/synchronization • better cell representation • eliminating redundancy and/or unnecessary work • compiler flags

More Related