Dynamic advertisement routing

A method of network latency optimization, the method comprising:

• storing, in a routing pool, a plurality of internet protocol addresses including a first internet protocol address, a second internet protocol address, and a third internet protocol address, and the first internet protocol address is associated with a first target node of a plurality of target nodes, the second internet protocol address is associated with a second target node of the plurality of target nodes, and the third internet protocol address is associated with a third target node of the plurality of target nodes;
• pinging each of the plurality of internet protocol addresses in the routing pool, including at least the first internet protocol address, the second internet protocol address, and the third internet protocol address, through each of a first load balancer network interface and a second load balancer network interface;
• selecting, at least one of a first network route to the first target node via the first load balancer network interface, a second network route to the second target node via the first load balancer network interface, and a third network route to the third target node based on a first plurality of ping responses via the first load balancer network interface;
• establishing a first communication session with the first target node through the first network route, a second communication session with the second target node through the second network route, and a third communication session with the third target node through the third network route;
• pinging the first internet protocol address, the second internet protocol address, and the third internet protocol address;
• updating a first latency, a second latency, and a third latency in a latency cache based on a second plurality of ping responses, wherein the third latency exceeds a latency threshold;
• receiving a first request directed to the plurality of target nodes;
• determining, based on the latency cache, to send the first request to the first target node;
• forwarding the first request to the first target node via the first network route;
• sending a second request directed to the plurality of target nodes to one of the first target node and the second target node;
• and
• sending a third request directed to the plurality of target nodes to the third target node after the latency service updates the latency cache with a fourth latency of a fourth network route to the third target node via the second load balancer network interface, wherein the fourth latency is under the latency threshold.

A computer-readable non-transitory storage medium storing executable instructions of network latency optimization, which when executed by a computer system, cause the computer system to:

• store, in a routing pool, a plurality of internet protocol addresses including a first internet protocol address, a second internet protocol address, and a third internet protocol address, and the first internet protocol address is associated with a first target node of a plurality of target nodes, the second internet protocol address is associated with a second target node of the plurality of target nodes, and the third internet protocol address is associated with a third target node of the plurality of target nodes;
• ping each of the plurality of internet protocol addresses in the routing pool, including at least the first internet protocol address, the second internet protocol address, and the third internet protocol address, through each of a first load balancer network interface and a second load balancer network interface;
• select at least one of a first network route to the first target node via the first load balancer network interface, a second network route to the second target node via the first load balancer network interface, and a third network route to the third target node via the first load balancer network interface based on a first plurality of ping responses;
• establish a first communication session with the first target node through the first network route, a second communication session with the second target node through the second network route, and a third communication session with the third target node through the third network route;
• ping the first internet protocol address, the second internet protocol address, and the third internet protocol address;
• update a first latency, a second latency, and a third latency in a latency cache based on a second plurality of ping responses, wherein the third latency exceeds a latency threshold;
• receive a first request directed to the plurality of target nodes;
• determine, based on the latency cache, to send the first request to the first target node;
• forward the first request to the first target node via the first network route;
• send a second request directed to the plurality of target nodes to one of the first target node and the second target node;
• and
• send a third request directed to the plurality of target nodes to the third target node after the latency service updates the latency cache with a fourth latency of a fourth network route to the third target node via the second load balancer network interface, wherein the fourth latency is under the latency threshold.

A system of network latency optimization, the system comprising:

• a memory storing a latency cache and a routing pool, wherein the routing pool stores a plurality of internet protocol addresses including a first internet protocol address, a second internet protocol address, and a third internet protocol address, and the first internet protocol address is associated with a first target node of a plurality of target nodes, the second internet protocol address is associated with a second target node of the plurality of target nodes, and the third internet protocol address is associated with a third target node of the plurality of target nodes;
• one or more processors;
• a plurality of load balancer network interfaces including a first load balancer network interface and a second load balancer network interface;
• and
• a load balancer including:
• a routing service;
• and
• a latency service,
• the load balancer service executing on the one or more processors to:
• ping, by the routing service, each of the plurality of internet protocol addresses in the routing pool, including at least the first internet protocol address, the second internet protocol address, and the third internet protocol address, through each of the first load balancer network interface and the second load balancer network interface;
• select, by the routing service, at least one of a first network route to the first target node via the second load balancer network interface, a second network route to the second target node via the first load balancer network interface, and a third network route to the third target node via the first load balancer network interface based on a first plurality of ping responses;
• establish a first communication session with the first target node through the first network route, a second communication session with the second target node through the second network route, and a third communication session with the third target node through the third network route;
• ping, by the latency service, the first internet protocol address, the second internet protocol address, and the third internet protocol address;
• update, by the latency service, a first latency, a second latency, and a third latency in a latency cache based on a second plurality of ping responses;
• receive, by the load balancer, a first request directed to the plurality of target nodes;
• determine, based on the latency cache, to send the first request to the first target node;
• forward, by the load balancer, the first request to the first target node via the first network route;
• select, by the routing service, a fourth network route, wherein a fourth latency to the first internet protocol address through the fourth network route via the first load balancer network interface is lower than a fifth latency to the first internet protocol address through a fifth network route via the second load balancer network interface;
• and
• send, by the load balancer, a second request to the first target node through the fourth network route.

A method of network latency optimization, the method comprising:

• storing, in a routing pool, a plurality of internet protocol addresses including a first internet protocol address, a second internet protocol address, and a third internet protocol address, and the first internet protocol address is associated with a first target node of a plurality of target nodes, the second internet protocol address is associated with a second target node of the plurality of target nodes, and the third internet protocol address is associated with a third target node of the plurality of target nodes;
• pinging each of the plurality of internet protocol addresses in the routing pool, including at least the first internet protocol address, the second internet protocol address, and the third internet protocol address, through each of a first load balancer network interface and a second load balancer network interface;
• selecting, at least one of a first network route to the first target node via the second load balancer network interface, a second network route to the second target node via the first load balancer network interface, and a third network route to the third target node via the first load balancer network interface based on a first plurality of ping responses;
• establishing a first communication session with the first target node through the first network route, a second communication session with the second target node through the second network route, and a third communication session with the third target node through the third network route;
• pinging the first internet protocol address, the second internet protocol address, and the third internet protocol address;
• updating a first latency, a second latency, and a third latency in a latency cache based on a second plurality of ping responses;
• receiving a first request directed to the plurality of target nodes;
• determining, based on the latency cache, to send the first request to the first target node;
• forwarding the first request to the first target node via the first network route;
• selecting a fourth network route, wherein a fourth latency to the first internet protocol address through the fourth network route via the first load balancer network interface is lower than a fifth latency to the first internet protocol address through a fifth network route via the second load balancer network interface;
• and
• sending a second request to the first target node through the fourth network route.

A computer-readable non-transitory storage medium storing executable instructions of network latency optimization, which when executed by a computer system, cause the computer system to:

• store, in a routing pool, a plurality of internet protocol addresses including a first internet protocol address, a second internet protocol address, and a third internet protocol address, and the first internet protocol address is associated with a first target node of a plurality of target nodes, the second internet protocol address is associated with a second target node of the plurality of target nodes, and the third internet protocol address is associated with a third target node of the plurality of target nodes;
• ping each of the plurality of internet protocol addresses in the routing pool, including at least the first internet protocol address, the second internet protocol address, and the third internet protocol address, through each of a first load balancer network interface and a second load balancer network interface;
• select at least one of a first network route to the first target node via the second load balancer network interface, a second network route to the second target node via the first load balancer network interface, and a third network route to the third target node via the first load balancer network interface based on a first plurality of ping responses;
• establish a first communication session with the first target node through the first network route, a second communication session with the second target node through the second network route, and a third communication session with the third target node through the third network route;
• ping the first internet protocol address, the second internet protocol address, and the third internet protocol address;
• update a first latency, a second latency, and a third latency in a latency cache based on a second plurality of ping responses;
• receive a first request directed to the plurality of target nodes;
• determine, based on the latency cache, to send the first request to the first target node;
• and
• forward the first request to the first target node via the first network route;
• select a fourth network route, wherein a fourth latency to the first internet protocol address through the fourth network route via the first load balancer network interface is lower than a fifth latency to the first internet protocol address through a fifth network route via the second load balancer network interface;
• and
• send, by the load balancer, a second request to the first target node through the fourth network route.