What does scaling mean?
Scaling means changing how much computer power you have. If many people visit your site at once, one server may become slow. Scaling gives you more power for that time.
What is horizontal scaling?
Horizontal means sideways. You add more servers next to the first one.
Picture a shop with one cashier. A long queue forms. You open three more checkout lines. The work is shared and the queue gets shorter.
Each extra server is often called a node or an instance. A node is just one server in a group.
What does automatic mean?
Automatic means you do not have to press any button. The cloud watches the load. Load means how hard the servers are working.
- When load goes above a limit, new servers are added. This is called scaling out.
- When load drops, extra servers are removed. This is called scaling in.
The helper that shares visitors
When you have many servers, visitors need to be spread between them. A load balancer does this. It is like a traffic officer who sends each car to a free road.
How it usually works
- You set a rule, such as "add a server when the processor is busy for a few minutes".
- You set the smallest and the largest number of servers.
- The cloud watches the load.
- When the rule is met, it starts a new server with a copy of your site.
- The load balancer begins sending visitors to it.
- When the rush ends, the extra server is shut down.
Good things about it
- Your site stays fast in busy times.
- If one server fails, others keep working.
- You use extra servers only when you need them.
Things to remember
- Your site must be built so that several servers can share it. Files and databases must be kept in a shared place.
- More servers can mean a bigger bill on many cloud services.
Quick recap
- Horizontal scaling means adding more servers side by side.
- Automatic means the cloud does it for you based on rules.
- A load balancer shares visitors between servers.
- Extra servers are removed when the rush is over.
- Ask support if your plan includes it.