ELB Types

Deployment mode

You get exclusive access to load balancer resources. The performance of a dedicated load balancer is never affected by the loads on other load balancers. In addition, there are a wide range of specifications available for you to choose from.

They are deployed in clusters and share resources with other instances. They support guaranteed performance.

Specifications

• Elastic specifications: You are charged for how long each load balancer is running and the number of LCUs you use.
• Fixed specifications: Multiple specifications are available for you to select to best meet your needs.

For details, see Specifications of Dedicated Load Balancers.

N/A

Performance

A dedicated load balancer in an AZ can establish up to 20 million concurrent connections. If you deploy a dedicated load balancer in two AZs, the number of concurrent connections will be doubled.

For example, if you deploy a dedicated load balancer in two AZs, it can handle up to 40 million concurrent connections.

If guaranteed performance is enabled, shared load balancers can handle up to 50,000 concurrent connections, 5,000 new connections per second, and 5,000 queries per second.

AZ

You can select one or more AZs as needed.

• If requests are from the Internet, the load balancer in each AZ you select routes the requests based on source IP addresses. If you deploy a load balancer in two AZs, the requests the load balancers can handle will be doubled.
• For requests from a private network:
  • If clients are in the AZ you select when you create the load balancer, requests are distributed by the load balancer in this AZ. If the load balancer is unhealthy, requests are distributed by the load balancer in another AZ you select.

    If the load balancer is healthy but the connections that the load balancer needs to handle exceed the amount defined in the specifications, service may be interrupted. To address this issue, you need upgrade specifications. You can monitor traffic usage on private network by AZ.

  • If clients are in an AZ that is not selected when you create the load balancer, requests are distributed by the load balancer in each AZ you select based on source IP addresses.
• If requests are from a Direct Connect connection, the load balancer in the same AZ as the Direct Connect connection routes the requests. If the load balancer is unavailable, requests are distributed by the load balancer in another AZ.
• If clients are in a VPC that is different from where the load balancer works, the load balancer in the AZ where the original VPC subnet resides routes the requests. If the load balancer is unavailable, requests are distributed by the load balancer in another AZ.

N/A

Billing item

• Fixed specifications: billed by the LCUs based on the specifications you select.
• Elastic specifications: billed by how many LCUs you use and how long you use your load balancers

You are charged for how long you use each load balancer if guaranteed performance is enabled.

Capabilities

Powerful capabilities to process Layer 4 and Layer 7 requests, advanced forwarding policies, and multiple protocols.

Basic capabilities to process Layer 4 and Layer 7 requests

Application scenarios

Heavy-traffic and highly concurrent services, such as large websites, cloud-native applications, IoV, and multi-AZ disaster recovery applications

Services with low traffic, such as small websites and common HA applications

Frontend protocols

TCP, UDP, HTTP, TLS, and HTTPS

TCP, UDP, HTTP, and HTTPS

Backend protocols

TCP, UDP, HTTP, HTTPS, and QUIC

TCP, UDP, and HTTP

Forwarding capabilities

Provide powerful Layer 4 and Layer 7 processing capabilities to forward requests based on the following:

• Forwarding rules: domain name, path, HTTP request method, HTTP header, query string, and CIDR block
• Actions: forward to a backend server group, redirect to another listener, redirect to another URL, rewrite, and return a specific response body

Provide basic Layer 4 and Layer 7 processing capabilities to forward requests based on the following:

• Forwarding rules: domain name and path
• Actions: forward to a backend server group and redirect to another listener

Key functions of backend server groups

• Health check
• Sticky session
• Slow start

• Health check
• Sticky session

Load balancing algorithms

• Weighted round robin
• Weighted least connections
• Source IP hash
• Connection ID

• Weighted round robin
• Weighted least connections
• Source IP hash

Forwarding modes of backend server groups

• Load balancing
• Active/Standby

Load balancing

Backend type

• ECS
• IP as backend server
• Supplementary network interface
• BMS
• CCE Turbo cluster

• ECS
• BMS
• CCE Turbo cluster