This is the second project of CS-E4770 Designing and Building Scalable Web Applications at Aalto University. This project implements a Jamstack-like web application using Astro and measures its performance.
Check the report to see the usage and performance metrics.
The walking skeleton for Course Project Two consists of five services: (1) a frontend responsible for a user interface, (2) a backend responsible for an API, (3) a database server, (4) a flyway image used for setting up the database schema, and (5) an Nginx web server that acts as a reverse proxy.
When you start the walking skeleton, the Nginx web server is available on the port 7800. Requests to paths starting with /api are directed to the backend, while other requests are directed to the frontend.
There exists a grader image that can be used for grading submitted programming exercises. This is in folder "grader-image" -- you need to build it for it to work. Building it happens either by running the build.sh script or by running the command docker build -t grader-image . in the "grader-image" folder.
The grader essentially waits for a while and returns a random result.
The backend server uses the grader image (in grade.js) -- in brief, in grade.js, submitted code is copied to a new grader container that is then run, after which the result from grading is copied from the container. While the code has been written into a JS file, one could as well create shell scripts for the same purpose.
We strongly recommend creating a separate service that is responsible for grading of incoming submissions and moving the grade.js functionality into that service. In this case, the backend would then submit codes for grading into that service.