Basic Global Accelerator instances use the high-quality global Internet bandwidth and the transmission network of Alibaba Cloud to provide users with point-to-point acceleration services.
Introduction to basic GA instances
You can use basic Global Accelerator instances to accelerate content delivery at Layer 3 (IP). To enable acceleration, you need to only specify an acceleration area and an endpoint group.
Access mode
You can use only accelerated IP addresses to connect to basic Global Accelerator instances. The accelerated IP addresses must be Elastic IP Address (EIP). Clients can connect to the nearest access point of the Alibaba Cloud global transmission network by sending requests to an accelerated IP address.
Acceleration network
Basic GA instances provide one-to-one acceleration services between acceleration regions and origin server regions. Basic GA instances provide acceleration services only at Layer 3 (IP protocols).
After client requests reach the Alibaba Cloud global transmission network, GA distributes the traffic to backend services based on the mappings between accelerated IP addresses and endpoints.
Origin server connection
GA forwards client requests to backend services only over private networks. The following endpoint types are supported: elastic network interfaces (ENIs), Classic Load Balancer (CLB) instances, Network Load Balancer (NLB) instances, and Elastic Compute Service (ECS) instances.
Limits on basic GA instances
Limits on acceleration areas
You can add only one acceleration area and one acceleration region for each basic GA instance. Only IPv4 clients can connect to basic GA instances.
By default, you can add up to 20 accelerated IP addresses and up to 10 idle accelerated IP addresses to a basic GA instance.
If you want to add more accelerated IP addresses or increase the quota of idle accelerated IP addresses, you can increase the gaplus_quota_basic_gaip_limit quota or the gaplus_quota_basic_gaip_idle_limit quota on the Quota Management page. For more information, see the Adjust quotas section of the Manage GA quotas topic.
Limits on the bandwidth of an acceleration region
The limits on the bandwidth of an acceleration region vary based on the billing method and bandwidth metering method of the basic Global Accelerator instance.
Instance billing method | Bandwidth metering method | Limits on the bandwidth of an acceleration region |
Pay-as-you-go | Pay-by-data-transfer | You can specify the maximum bandwidth of an acceleration region based on your business requirements. Bandwidth range: 2 Mbit/s to 20,000 Mbit/s. All accelerated IP addresses in the acceleration region share the maximum bandwidth. |
Subscription | Pay-by-bandwidth | The bandwidth of an acceleration region is throttled based on the maximum bandwidth of the basic bandwidth plan that is associated with the GA instance. You can associate only basic bandwidth plans that provide basic bandwidth or premium bandwidth with basic GA instances. |
Limits on endpoint groups
You can add only one endpoint group. By default, you can add up to 200 endpoints.
If you want to add more endpoints, you can increase the gaplus_quota_basic_endpoint_limit quota on the Quota Management page. For more information, see the "Adjust quotas" section of the Manage GA quotas topic.
Only the following endpoint types are supported: secondary ENIs, CLB instances in VPCs, internal-facing NLB instances, and Elastic Compute Service instances. ENIs, CLB instances, NLB instances, and ECS instances cannot be associated with EIPs or static public IP addresses.
If your basic Global Accelerator instance was created after August 1, 2023, you can create multiple IP addresses for an acceleration region and use internal-facing Network Load Balancer (NLB) instances as endpoints. If your basic Global Accelerator instance was created before August 1, 2023 and you want to use the preceding features, contact your account manager to upgrade your GA instance.
If your Global Accelerator instances were created before August 1, 2023, the following limits apply:
You can add only one acceleration area and one acceleration region to each GA instance.
You can add only one endpoint group and one endpoint to each GA instance.
You cannot specify an NLB instance as an endpoint.
You do not need to add an accelerated IP address or associate an accelerated IP address with an endpoint.
Acceleration areas and regions supported by basic Global Accelerator instances
The regions in the following table are for reference only. You can view the actual regions that are supported by GA instances in the GA console.
The following acceleration regions are unavailable by default. If you want to specify these acceleration regions, use the following methods:
If you want to specify China (Nanjing), Brazil (Sao Paulo), Canada (Toronto), Canada (Vancouver), South Korea (Seoul), Thailand (Bangkok), Vietnam (Ho Chi Minh), and UAE (Dubai) as acceleration regions, contact your account manager.
If you want to specify China (Heyuan) as the acceleration region, apply for the privilege in Quota Center.
Log on to the Privileges page of the Quota Center console and apply for the privilege named Accelerated regional whitelist. The ID of the privilege is ga_whitelist/region_whitelist. For more information, see the Adjust quotas section of the Manage GA quotas topic.
Subscription basic Global Accelerator instances do not support the UAE (Dubai) region.
Acceleration area | Region |
North China | China (Qingdao), China (Beijing), and China (Zhangjiakou) |
South China | China (Shenzhen), China (Heyuan), and China (Guangzhou) |
East China | China (Hangzhou), China (Shanghai), and China (Nanjing) |
Southwest China | China (Chengdu) |
North America | US (Silicon Valley), US (Virginia), Canada (Toronto), and Canada (Vancouver) |
South America | Brazil (Sao Paulo) |
Asia Pacific | China (Hong Kong), South Korea (Seoul), Singapore, Malaysia (Kuala Lumpur), Japan (Tokyo), Indonesia (Jakarta), Philippines (Manila), Thailand (Bangkok), and Vietnam (Ho Chi Minh) |
Europe | Germany (Frankfurt) and UK (London) |
Middle East | UAE (Dubai) |
Procedure
