I've built a small web page that makes yfinance requests to retrieve stock data, and I'd like to put it online as a working product. I'm still in the early development stage, so the traffic and computing requirements should be modest. What are the most reliable free or inexpensive ways to host it, and how would I deploy a Python application to a hosted server? I'd also like to understand when I would need to upgrade as usage increases.
5 Answers
You could also consider an edge-worker platform if the application can be rewritten around lightweight HTTP requests. That can be inexpensive for simple endpoints, but it won’t run every Python library or support long-running processes. Since yfinance and its dependencies may not fit that environment directly, a conventional Python host or serverless function is likely the simpler starting point.
For a beginner, a managed deployment service is often easier than configuring a virtual private server yourself. You connect the project, specify the Python version and start command, and the platform builds and runs it. Free plans commonly sleep when unused and may have limits on runtime, bandwidth, or background jobs. Once the app gets regular traffic, a small paid instance—often only a few dollars per month—can provide more predictable performance.
If your app is compatible with a serverless model, an option such as AWS Lambda can be very inexpensive and may remain within the free allowance for low traffic. Each request starts a short-lived function instead of keeping a Python process running continuously. That approach works well for small API endpoints, but it may require adapting a traditional Python web app and checking execution-time, memory, and outbound-request limits.
For a basic stock-data application, you probably don’t need much CPU, but you do need to consider how often it calls the data provider and how many visitors use it. A free hosting platform can work during development, although some free services put inactive apps to sleep. That means the first request after a period of inactivity may be slow. This is fine for a prototype, but less ideal for a production service that must respond immediately.
The best option depends on how your app is structured, but a small web app can often fit within a cloud provider’s free tier. Services from providers such as Azure, AWS, Google Cloud, or Oracle can give you a small virtual machine or application environment. A virtual machine is essentially a Linux server where you install Python, upload your code, install its dependencies, configure a web server, and keep the application running. Be careful with billing limits and verify what is genuinely free, since free tiers and quotas can change.
So with a service like Azure, I could create a small virtual machine and run the Python web app on it? I’m mainly trying to learn the deployment process now and scale the machine later if traffic grows.
That sounds more suitable for getting the first version online. I can start with a sleeping free service, then move to a paid instance once the application has consistent users.