DigitalOcean fully manages Regional Load Balancers and Global Load Balancers, ensuring they are highly available load balancing services. Load balancers distribute traffic to groups of backend resources in specific regions or across different regions, which prevents the health of a service from depending on the health of a single server, cluster, or region.
DigitalOcean offers two different types of load balancers: regional and global. Each type of load balancer has different pricing and inclusions.
Global load balancers cost $15.00 per month and include the following:
Regional load balancers cost $12.00 per month for each node they contain.
Each additional node increases the load balancer’s maximum:
You can add up to 200 nodes to a load balancer if your account limits allow it. To request a limit increase, contact support.
You can scale regional load balancers up or down at any time to meet your performance needs. The more nodes a load balancer has, the more simultaneous connections and requests per second (RPS) it can maintain. We prorate the load balancer’s costs by the number of hours it runs at each size. Your invoice displays the number of hours the load balancer runs at each size on a separate line. You can resize a load balancer only once per minute.
Performance may vary depending on the load balancer’s workload. Using different protocols and package management settings produce different results. Therefore, we cannot provide specific performance metrics for each load balancer size, and we strongly recommend that you run your own benchmarks to see what size works for your application’s specific needs.
There is no additional cost to use Let’s Encrypt with load balancers.
The maximum number of new SSL connections does not apply to load balancers configured for SSL passthrough.
DigitalOcean Load Balancers by themselves don’t generate bandwidth charges; they are bandwidth neutral.
The public outbound traffic that originates from your resources and passes through the load balancer counts towards your bandwidth limit. In this scenario, the aggregated bandwidth is reported as part of the load balancer and not attributed to the individual resources behind it.