A load balancer is a traffic management implementation that optimizes resource use by distributing network traffic across different, redundant servers to ensure application availability and performance.
Load balancing is focused on efficiently distributing traffic to the most appropriate servers. Load balancing can occur at multiple levels. In cloud applications, the most common levels are Layer 4 and Layer 7. Different algorithms have been developed over time for load balancing, including round robin, weighted least request, random, least connections and more.
Load balancers can implement a number of additional capabilities, including:
- health checks that verify traffic is being routed to healthy instances
- observability to provide visibility into metrics such as latency, error rates, and throughput
- TLS termination, to secure encrypted connections
Load balancers are a critical part of cloud-native infrastructure and serve as the initial entry point for external traffic to business applications.