Parallelized

As conventional performance gains in CPUs slow down, it is more important than ever for us to truly utilize our hardware in our software. Thus, Parallelized is an event where teams compete to achieve the maximum possible speedup by developing and optimizing a program within 3 hours; whichever team has the highest speedup, wins! Learn more on the website. Please make sure to bring your own devices.

How It Works

• Each team is provided with guidelines for the task and has a 3-hour window to create a fully functional Python program that adheres to these guidelines and achieves maximum speedup compared to the OC’s base implementations, which are naive and not parallelized/optimized.
• While working on the task, teams are restricted in their web access, with access limited to specific websites. Attempting to circumvent the thinking process by searching for entire code blocks on the web is a disqualifiable offence.
• All code must be contained within a single Python file, which will serve as a library used for testing the implementations. - Additionally, a testing Python file is provided to assess the functionality of the implementations and measure the execution time on each team’s respective system. It’s important to note that the execution time on the benchmarking system may differ from that on individual systems, serving as a progress gauge and verifying the program’s functionality.

Note: You are not expected to memorize any algorithms - in fact, you are expected to thoroughly read the guidelines and recognize patterns in the inputs and desired outputs and thus create your own execution algorithm

Timeline:

2nd November: 5 pm onwards.