mandelbrot

Client/Server python scripts for mandelbrot set calculations.

Client script can generate an image in multiple ways:

1. Locally, by using mandelbrot functions
2. Locally, by extracting starting points of sub-images and calculating each sub-image color values
3. By sending starting points of sub-images to multiple servers (where each server calculates color values for assigned sub-images)
4. The same as 3. just by using multiple threads to distribute data to defined servers

Server script

• Created using standard python libraries and BaseHTTPRequestHandler and HTTPServer from BaseHTTPServer
• Starting a server with a port defined as an argument. Example: server.py 8080
• Function serve_forever() keeps the server running
• After running the server on the specified port the output is the following:

  HTTP server is starting on port 1111...
  HTTP server is running...
  

Client script

• Created using standard python libraries
• For the creation of the image, needed Image module from Python Image Library (PIL). Install with: pip install image
• The script is started without any arguments: client.py
• Output after opening the script looks like this:

-------------------- Menu --------------------

Data defined for the calculations:

MAX_N = 255      #Iterations
WIDTH = 1024, HEIGHT = 1024     #Image
MIN_C_RE = -2.0, MAX_C_RE = 2.0, MIN_C_IM = -2.0, MAX_C_IM = 2.0        #Plot
DIV = 4         #Width/height of sub-pictures
SERVER_LIST = 127.0.0.1:1111 127.0.0.1:2222 127.0.0.1:3333 #List of servers

[0] - Exit.

[1] - Change parameters used for calculations.
[2] - #Debug: Mandelbrot function iterations.
Generate image by:
[3] - #Debug: Use mandelbrot function (client only).
[4] - #Debug: Collect start points & calculate colors of sub-pictures (client only).
[5] - Send start points of sub-pictures to defined servers (single-thread).
[6] - Send start points of sub-pictures to defined servers (multi-threaded).

Enter menu number to continue:

Menu choices

1. Updating of the parameters used for calculations:
   • Iterations, width, height, div must be integers greater than 0
   • Div must be divisible with the width/height of the image, as it represents the size of the sub-images
   • Plot parameters must be in the range [-2, 2]
   • List of servers has to contain valid IP addresses and port numbers
2. Prints raw output from the mandelbrot function. #for debugging and better understanding
3. Generate the image on the client-side only. #for debugging and comparison
4. The job that should be done on the server-side is done on the client-side. #for debugging and comparison
5. Single thread sends the data to defined servers and collects results. The image is generated after all requests are completed.
6. Use multiple threads to distributes the data to defined servers and connect results. The image is generated after all requests are completed.

Parameters

On the image below, comparison when using different iteration parameter is shown: max_n = [10, 50, 255, 1024]

Execution time

Execution time is depending on the number of iterations and the size of the image. On smaller images, execution time on the client is faster than by using servers. Example below:

max_n = 1024, image size = 2000x2000
3. 229.246000051 seconds #client only
4. 247.770999908 seconds #client with sub-images
5. 423.139000177 seconds #servers single thread
6. 248.412000179 seconds #servers multiple threads

As the image size and iterations increases, the amount of work also grows. After each function execution time is calculated so that it can be used for comparison.

The image generated from (6.) is shown below. TIP: Open it and zoom it 🔍 👀

Running with multiple servers

On the image below, running on 4 different servers with multiple threads is shown:

License

© 2019 Dragana Vučić