Global Traffic Manager (GTM) is applicable to the following scenarios:
Assume that your application service has two IP addresses, including 188.8.131.52 and 184.108.40.206. In normal conditions, users access the application service by using the IP address 220.127.116.11. If the IP address 18.104.22.168 fails, you need to switch access traffic to the IP address 22.214.171.124.
You can use GTM to create two address pools, including Pool A and Pool B. Add the IP addresses 126.96.36.199 and 188.8.131.52 into the two address pools, and then configure the health check feature. When you configure an access policy, set the primary address pool to Pool A and the primary address pool to Pool B. This way, the switchover of the active/standby IP addresses can be achieved.
Assume that your application service has three IP addresses, including 184.108.40.206, 220.127.116.11, and 18.104.22.168. The three IP addresses are available to users at the same time. In normal conditions, DNS returns the three IP addresses at the same time. If one of the three IP addresses fails, the address is temporarily removed from the DNS address list and is not returned to users. If the IP address becomes available, it is added to the address list.
You can use GTM to create an address pool Pool A that contains the IP addresses 22.214.171.124, 126.96.36.199, and 188.8.131.52. Set the primary address pool set to Pool A, and then enable and configure the health check feature. This way, the active redundancy of multiple IP addresses can be achieved.
When enterprises perform promotional activities, for example, on Double 11, they must temporarily scale up cloud resources to handle a sudden increase of access requests by several times. Multiple Server Load Balancer (SLB) instances are purchased in the same region to distribute access traffic to different IP addresses.
When you use GTM, you only need to set the load balancing policy of the primary address pool set to Return all addresses. The access traffic is evenly distributed to each IP address. This way, load balancing is achieved in highly concurrent scenarios. You can also select Return addresses by weight and configure different weights for each address pool and each address. The access traffic is distributed to each IP address based on the specified weight.
Large enterprises need to provide network services in regions across a nation or the globe. Network access is affected by different network conditions in different regions. Therefore, the enterprises set access points in the core locations of the regions. Users in different regions can access the services from the nearest access points.
GTM provides the following two access policies:
Access policy based on geographical locations: GTM returns addresses in specified address pool sets to users in different regions. Users can accelerate network access by using nearest points.
Access policy based on latency: GTM detects the access latency between the location of a user and the region in which the application is deployed. Then, GTM routes user requests to the application server cluster that has the lowest latency. This way, network access is accelerated.