Edge Load Balancer (ELB) distributes incoming traffic to multiple backend Edge Node Service (ENS) instances based on forwarding policies. ELB improves business processing capabilities, expands service capabilities, and increases service availability.
Benefits
Scalability: You can use ELB as traffic ingress points for web servers and application servers. This allows you to improve the performance of your service system by adding or removing ENS instances based on your workloads.
High concurrency: If a single ENS instance cannot meet processing requirements, you can configure ELB listening rules to distribute traffic to different ENS instances.
High availability: Disaster recovery is supported to eliminate single points of failure (SPOFs) and ensure continuous service availability. You can use the session persistence feature to forward requests from the same client to the same backend ENS instance to improve access efficiency.
Multi-protocol support: ELB supports TCP and UDP at Layer 4 and HTTP and HTTPS at Layer 7. You can use ELB to provide fast and secure connections to applications such as real-time audio and video applications, interactive streaming applications, and online game applications.
Pricing
You are billed for ELB based on the pay-as-you-go billing method. The fees vary based on the instance type. ELB instances that are created in the ENS console are Internal-facing instances and are not publicly accessible. To allow an ELB instance to access to Internet, you need to associate a public elastic IP address (EIP) with the ELB instance. EIP charges you for Internet data transfer fees. For more information, see Associate an EIP.
ELB charges you based on a combination of the following metrics.
Type | Name | Maximum number of connections | CPS | QPS | Instance fee (USD/month) |
elb.s1.small | Lite | 5000 | 2000 | 800 | 7.5 |
elb.s2.small | Standard I | 50000 | 4000 | 3000 | 26 |
elb.s2.medium | Standard II | 100000 | 8000 | 6000 | 52 |
elb.s3.small | Advanced I | 200000 | 16000 | 12000 | 104 |
elb.s3.medium | Advanced II | 500000 | 40000 | 18000 | 166 |
Maximum number of connections: the maximum number of connections that an instance can support. When the number of connections exceeds the upper limit, new connection requests are discarded.
Connections per second (CPS): the maximum number of new connections per second. When the number of new connections per second exceeds the upper limit, new connection requests are discarded.
Queries per second (QPS): the number of HTTP or HTTPS queries (requests) that can be completed per second. When the actual number reaches the upper limit, new connection requests are discarded.