Project 3 — Load balancer

Overview

In this project, we will learn how to model a simulation of a large company's load-balancing web requests using a load balancer. We will simulate requests that contain IP addresses from the requester, IP addresses for the results, and a time (in clock cycles).

This project will require a real-world implementation of a queue. We also need a struct, a lower-level data structure that allows data to be contained in an object.

Load balancers are available commercially, but in this project, you will build one that has:

• Requests - generates random IP addresses, both in and out, and random times for each request to be processed.
  • IP in
  • IP out
  • Time (integer)
  • Job type (Character P (processing), S (streaming))
• Webserver(s)
  • Takes requests from the Load Balancer
  • Process the requests
  • Asks for another
• Load Balancer(s)
  • Queue of requests
  • Keeps track of time.

Project Objectives

You will be able to:

• Show your understanding of building and using a load balancer.
• Develop a guide for your load balancer that does not get overloaded or underutilized.

Honors Project Objectives

You will be able to:

• Show your understanding of building and using a load balancer.
• Develop a guide for your load balancer that does not get overloaded or underutilized.
• Utilize threading for your load balancer and the servers.

Instructions:

1. This is an individual project. You must turn in your own work.
2. Collaboration for ideas between classmates is acceptable, but duplication of work is not.
3. Use C++ to write this program.
   1. A) you must design the workflow and write the driver or main program on your own.
   2. B) you must design the functions of the methods and classes you need, including the inputs and output of each function.
   3. C) you can then (and only after the design is complete) use ChatGPT to write the individual functions. For example, using ChatGPT, say "write a function in C++ that will take a real number value as US currency and convert it to Singapore currency". Your driver program or other functions can call that AI-written function.
4. Your program will contain multiple .cpp and .h files. The main program should set up the number of servers (user input), the time you want to run the load balancer (user input) and generate a full queue (usually servers * 100).
5. You will need a web server class so you can create the number of web servers requested.
6. You will need a request class (or struct) that holds a request.
7. You will need a queue of type requests. (called requestqueue?)
8. You will need a loadbalancer class that manages the webservers and the request queue. This should for creating multiple load balancers.
9. You will need to add new requests at random times to simulate new requests after the initial full queue you set up.
10. Dynamically allocate and deallocate web servers to maintain a balance with your requests.
11. Try different numbers as input to understand the capacity of your load balancer configurations.
12. All files must be commented with Doxygen using HTML output and loaded to a page on people.tamu.edu. Turn in this link. (having ChatGPT create Doxygen compensation is an excellent use of AI and acceptable in this assignment. Not many people like writing commentation of this type, but it is enjoyable seeing a tool create it for you.)
13. Demo the load balancer in your video.
14. Optional—1: Add an IP range blocker to serve as a firewall or DOS attack prevention mechanism. If everything above works and you choose this option, extra credit may be awarded.
15. Optional—2: Add a higher-level load balancer that can sort jobs by type, sending streaming data to one load balancer (and therefore specific servers for that purpose) and processing data to another load balancer (and therefore specific servers for processing data). If everything above works and you choose this option, extra credit may be awarded.

Deliverables:

All deliverables must provide a viable solution to the assigned problem. No attempt will be made to grade non-substantial submissions.

• Create documentation for each of your code files in HTML, hosted on people.tamu.edu
• Create a log of 10 servers running for 10000 clock cycles
• Turn in your git repository link.
• Demonstration of usability in your video.
• Create a make file and zip it together with all your .cpp and .h files. No executables. Turn in that zip file.

This project is worth 100 points.

Documentation 30%

Log and successful completion of load balancing 20%

Demonstration, code, and Git usage 50%