We’re happy to announce the addition of the load balancing functionality to our Leafcloud suite.
Our users can now use the Leafcloud dashboard, or API to create a load balancer. Load balancers have a publicly addressable IP address and can be configured to regulate the traffic between two or more hosts. This will help you balance the connection requests to various instances, so they share the traffic load. You can find details on usage, as well as a tutorial to help you get started, below.
Work is underway on clustering and database services and we’ll be adding new tools in the coming weeks. So, sign up for our newsletter and keep an eye out.
Load balancing with Leafcloud
We’ve based our load balancing service on Octavia, “an open-source, operator-scale load balancing solution designed to work with OpenStack.” You can find everything you need to know about Octavia here.
Load balancers allow you to maintain availability and ensure uptime. This works by placing two or more identical instances behind a load balancer. All of the instances are ready to service requests and the load balancer distributes traffic based on your choice of algorithm. If the load balancer detects that an instance is unavailable, incoming requests are redirected to the others. This way, requests never go unserviced.
Load balancers give you the option to choose between several different algorithms. These allow you to manage your traffic in different ways. Choose the load balancing algorithm that best fits your needs:
- ROUND_ROBIN, to iterate over your instances one by one.
- LEAST_CONNECTIONS, to keep the number of connections equal between instances.
- SOURCE_IP, to send the user to the same instance every time, for stateful server apps.
Check out the tutorial for more info and start balancing those loads!