Técnicas de escalado horizontal

Imagine your SaaS app suddenly gets thousands of new users! How do you handle the extra demand without your service slowing down or crashing?

This is where horizontal scaling comes in. It's about adding more machines to share the workload, rather than making a single machine more powerful.

There are two main ways to scale your application:

• Vertical Scaling (Scaling Up): Increase the resources of a single server (e.g., adding more CPU, RAM). This has limits and can be expensive.
• Horizontal Scaling (Scaling Out): Add more servers to your existing pool, distributing the load across them. This is often more flexible and cost-effective for SaaS growth.

When you have multiple servers, how do you ensure incoming user requests are sent to an available server, and not just overload one?

This is the job of a load balancer. It acts as a traffic cop, sitting in front of your servers and distributing incoming network traffic evenly across them.

A load balancer ensures no single server becomes a bottleneck. If one server is busy or fails, the load balancer intelligently redirects traffic to healthy, less busy servers.

This improves application responsiveness, increases availability, and enhances overall reliability for your users.

Load balancers use various algorithms to decide where to send traffic:

• Round Robin: Sends requests to servers in a rotating sequence.
• Least Connections: Directs traffic to the server with the fewest active connections.
• IP Hash: Maps a client's IP address to a specific server, useful for maintaining session affinity.

Here's a simplified Python example showing how requests might be distributed in a round-robin fashion across a set of servers:

Auto-scaling is the ability to automatically adjust the number of computing resources in a server group based on demand.

If traffic spikes, more servers are added. If traffic drops, servers are removed. This saves costs and ensures performance.

Auto-scaling systems use metrics to decide when to act:

• CPU Utilization: If average CPU usage goes above 70%, add a server.
• Network I/O: If network traffic exceeds a certain threshold, scale out.
• Queue Lengths: For message queues, if the number of pending messages grows too large, add more workers.

For effective horizontal scaling, your services should be stateless. This means each request from a client contains all the information needed to process it, and the server doesn't store any client-specific data between requests.

Why is this important? Because any server can handle any request, making it easy to add or remove servers without disrupting user sessions.

Let's compare:

• Stateless: Servers process requests independently. Example: A simple API that returns data. User session data is stored externally (e.g., in a database or cache).
• Stateful: Servers remember information from previous interactions. Example: A server holding a user's shopping cart in its memory. This makes scaling harder, as a user must always return to the same server.

For horizontal scaling, always aim for stateless services.

Great job! In this lesson, we explored core horizontal scaling techniques for SaaS:

• Horizontal Scaling: Adding more servers to distribute load.
• Load Balancers: Essential for distributing incoming traffic across multiple servers.
• Auto-Scaling: Automatically adjusting server count based on demand.
• Stateless Design: Crucial for services to be easily scaled out, ensuring any server can handle any request.

These techniques are fundamental for building scalable and resilient SaaS platforms.