A simple guide to servers, Databases, & CDNs

What is Cloud Infrastructure?

Cloud infrastructure refers to the combination of hardware and software resources, enabling businesses to store data, run applications, and manage services without having to maintain physical servers on-site. Instead of buying and managing their own servers, companies can rent computing power, storage, and other resources from cloud providers like Amazon Web Services (AWS) or Microsoft Azure. This approach is essential for modern businesses because it offers flexibility, scalability, and cost savings, enabling them to easily adapt to changing needs and focus on their core operations rather than IT maintenance.

Servers and Serverless

Servers
Servers are powerful computers that store, process, and provide data or services to other computers over a network, like a website hosting server that delivers webpages to your browser. They can be scaled up or down depending on how much demand there is, meaning more or fewer resources can be added to handle varying amounts of traffic. This flexibility reduces the need for businesses to maintain and manage large amounts of hardware themselves, saving time, money, and effort.

Serverless
Serverless computing is a way to build and run applications without worrying about managing the underlying servers. Instead of setting up and maintaining servers yourself, a cloud provider takes care of all that behind the scenes, allowing you to focus solely on writing and deploying your code. This approach simplifies managing infrastructure since the cloud provider automatically handles scaling, updates, and maintenance. It's also cost-effective, especially for applications with unpredictable or infrequent usage patterns, as you only pay for what you use. However, there are some limitations, such as challenges with frequent database connections and restrictions on the size of data that can be returned, making it less ideal for serving large files like PDFs.

Recommendation
For most use cases, we recommend using traditional servers (EC2 instances on AWS). Serverless functions can be a great tool for specific use-cases where the scope is well-defined. For early stage companies, the scope and requirements can often change and it’s much easier to handle dynamic requirements on a traditional server.

Databases

Databases are essential tools for storing and organizing data. Relational databases (SQL), like MySQL and PostgreSQL, use structured tables to manage data, making them great for tasks that need consistency and clear relationships between data points. On the other hand, NoSQL databases, like MongoDB and DynamoDB, are more flexible and don’t require a fixed structure, which makes them ideal for handling unstructured data like social media content or large-scale data streams.

SQL (Structured Query Language) databases, like MySQL and PostgreSQL, are relational databases that organize data into structured tables with predefined schemas, making them great for handling consistent and structured data. They are ideal when relationships between data points are important, and they use SQL for querying data.

NoSQL databases, like MongoDB and DynamoDB, are more flexible, allowing for storage of unstructured or semi-structured data without requiring a fixed schema. They are well-suited for handling large-scale, dynamic data, such as social media content or real-time analytics. While SQL databases excel in complex queries and transactional consistency, NoSQL databases offer scalability and flexibility for handling diverse data types.

Recommendation: For many use cases, a traditional SQL database will solve most requirements. For use cases where you have unstructured data that is in a JSON format, databases like PostgresSQL have a JSON column type where you can store that data. In some scenarios, we have seen companies whose primary data format is JSON - in those scenarios, we have recommended using Mongo.

Content Delivery Networks (CDNs)

A CDN is a system of servers spread out across different geographic locations. These servers store copies of your website's content and deliver it to users from the server closest to them, which helps your website load faster. By reducing the distance between the content and the user, CDNs lower the time it takes for the content to reach the user, known as latency, and improve the overall speed of the site.

For example, a website that has a lot of images, might be slowed down because all of those images need to be sent from the main server. It’s much more efficient to have several servers spread throughout the region so that users requests don’t travel as far.

In the image below, you can see a global CDN network. In this scenario, each user’s request is sent to the local CDN server rather than travelling to the main server.

Executive Summary

Cloud infrastructure provides the foundation for modern business technology, offering resources like storage and computing power over the internet rather than on physical servers. This flexible and scalable system allows companies to focus on their core operations while leaving the heavy lifting of IT management to cloud providers. Within cloud infrastructure, businesses can choose between traditional servers, where computing power is scaled based on demand, or serverless computing, which offloads server management entirely to cloud providers.

Additionally, understanding databases is crucial, as relational (SQL) and NoSQL databases serve different business needs based on structure and scalability. Finally, using Content Delivery Networks (CDNs) helps businesses with global audiences deliver content faster, enhancing the user experience and reducing operational costs. Overall, cloud infrastructure offers flexibility, scalability, and efficiency to help businesses adapt quickly.