"How to Set Up a Load Balancer"

Setting up a load balancer on Data Mammoth distributes incoming traffic across multiple servers, improving your application's performance, reliability, and availability. This guide walks you through the entire setup process, from creating the load balancer to adding backend servers and verifying that everything works.

Prerequisites

Before setting up a load balancer, you need:

• At least two running servers (VPS or dedicated) hosting the same application. The load balancer distributes traffic across these servers.
• Your application deployed and working identically on each server. All backend servers should serve the same content and handle the same requests.
• If using HTTPS, an SSL/TLS certificate for your domain.

If you do not have servers yet, see How to Order Your First Server.

Step 1 — Create a Load Balancer

Log in to your Data Mammoth dashboard.

Click Network in the left sidebar.

Click Load Balancer.

Click Create Load Balancer or New Load Balancer.

Configure Basic Settings

• Name — Enter a descriptive name (e.g., "Production Web LB" or "API Load Balancer").
• Region — Select the data center region. Choose the same region where your backend servers are located for the lowest latency.
• Type — Select the load balancer type if options are available (Layer 4 or Layer 7).

Click Create to provision the load balancer. It will be assigned an IP address within a few minutes.

Step 2 — Configure Forwarding Rules

Forwarding rules define how the load balancer handles incoming traffic and routes it to backend servers.

Open your load balancer's detail page.

Click Forwarding Rules or Listeners.

Click Add Rule.

HTTP Forwarding Rule

This rule listens for HTTP traffic on port 80 and forwards it to port 80 on your backend servers.

HTTPS Forwarding Rule

This rule handles SSL termination — it decrypts HTTPS traffic and forwards the unencrypted request to backend servers on port 80. See Configuring SSL with Your Load Balancer for certificate setup details.

Custom Port Forwarding

For applications running on non-standard ports:

Click Save or Add Rule for each forwarding rule.

Step 3 — Select the Load Balancing Algorithm

Choose how the load balancer distributes traffic across backend servers:

• Round Robin — Sends requests to each server in rotation. Best for servers with equal capacity.
• Least Connections — Sends requests to the server with the fewest active connections. Best for variable-length requests.
• IP Hash — Routes requests from the same client IP to the same backend server. Useful for session persistence.

For a detailed explanation of each algorithm, see Load Balancing Algorithms Explained.

The default (Round Robin) works well for most web applications. Change the algorithm in the load balancer's settings if needed.

Step 4 — Add Backend Servers

Backend servers (also called "targets" or "nodes") are the servers that receive and process traffic from the load balancer.

In your load balancer's detail page, click Backend Servers or Targets.

Click Add Server.

Select a server from your list of active Data Mammoth servers, or enter a server's IP address manually.

Specify the port the backend server listens on (e.g., 80 for HTTP).

Optionally, set a weight (for weighted round robin — higher weight means more traffic).

Click Add.

Repeat for each backend server.

Example Backend Configuration

Equal weights distribute traffic evenly. If one server is more powerful, assign a higher weight to send it more traffic.

Step 5 — Configure Health Checks

Health checks monitor your backend servers and automatically remove unhealthy servers from the pool. This is essential for maintaining availability.

Navigate to the Health Checks section of your load balancer.

Configure the health check parameters:

Click Save.

For detailed health check configuration, see Configuring Health Checks.

Step 6 — Configure Session Persistence (Optional)

If your application requires users to stay connected to the same backend server (e.g., for session-based authentication), enable session persistence (sticky sessions):

Navigate to the load balancer's Settings or Session Persistence section.

Enable Sticky Sessions.

Choose the method:

- Cookie-based — The load balancer sets a cookie that identifies the backend server. - IP-based — Requests from the same client IP always go to the same server.

Set the session duration (e.g., 30 minutes).

Click Save.

Note: Sticky sessions can reduce the effectiveness of load balancing because traffic is not evenly distributed. Use them only when your application architecture requires it. Stateless applications do not need sticky sessions.

Step 7 — Update DNS

Point your domain to the load balancer instead of an individual server:

Go to your DNS provider's management interface.

Update or create an A record pointing your domain to the load balancer's IP address.

If you previously pointed the domain to a specific server, update it to the load balancer's IP.

Wait for DNS propagation (typically minutes to hours).

After DNS propagation, all traffic to your domain flows through the load balancer.

Step 8 — Verify the Setup

Test Traffic Distribution

Access your application through the domain or load balancer IP.

Verify the page loads correctly.

Check the access logs on each backend server to confirm that requests are being distributed.

Test Failover

Stop one backend server (or remove it from the pool).

Access your application — it should continue working, served by the remaining servers.

Start the server again (or add it back) and verify it rejoins the pool.

Monitor

Use your Data Mammoth dashboard to monitor:

• Load balancer traffic volume.
• Backend server health status.
• Request distribution across servers.
• Response times and error rates.

What to Do Next

• Load Balancing Algorithms Explained — Choose the optimal algorithm for your workload.
• Configuring SSL with Your Load Balancer — Set up HTTPS.
• Configuring Health Checks — Fine-tune health monitoring.
• What Is a Load Balancer? — Review the fundamentals.