DNS rendezvous localization
A method of generating a routing table containing information as to the weighted distance between client's that use a resolver and each rack gateway, taking into account how traffic to each client can egress from the CDN AS. The routing table is generated from matrix multiplication of two matrices. One matrix contains information as to the proportion of each client's use of each resolver in a first autonomous system. The second matrix contains information as to the distance between each client and each rack, with respect to an egress gateway, in a second autonomous system. The resulting routing table is used to identify a gateway from which to serve content to a client.
1. A method of serving content comprising:
determining a portion of traffic between client devices within a first autonomous system and client resolvers within the first autonomous system;
determining distance information between the client devices and edge servers within a second autonomous system, the distance information related to egress of content from at least one of the edge servers via at least one egress gateway of the second autonomous system to the first autonomous system;
determining a network location between the client resolvers and the edge servers using the determined portion of traffic and the determined distance information; and
serving content to at least one of the client devices from at least one of the edge servers based on the determined network location.
2. The method of claim 1 , wherein the portion of traffic is expressed in a client-to-resolver matrix, the distance information is expressed in a client-to-gateway distance matrix, and the network location between the client resolvers and the edge servers is created by a matrix product of the client-to-resolver matrix and the client-to gateway distance matrix.
3. The method of claim 2 , wherein the portion of traffic includes a client metric wherein the client metric is used to populate the client-to-resolver matrix.
4. The method of claim 3 , wherein the client metric is a geographic identifier for a client device.
5. The method of claim 4 , wherein the geographic identifier is a latitude and longitude attribute associated with a client device.
6. The method of claim 2 , wherein the portion of traffic is a proportion of client devices within a first autonomous system accessing the client resolvers within the first autonomous system, the client-to-resolver matrix is a representation of at least one of a client's use of at least one resolver, and the client-to-gateway distance matrix is a representation of a distance between at least one client device and at least one edge server.
7. The method of claim 6 , wherein a routing table is used to select a edge server for at least one client device.
8. The method of claim 1 , further comprising:
sorting the network location between the client resolvers and the edge servers based on a distance between each of the edge servers and the client resolvers; and
preferring the edge server based on a shortest distance of the sorted network location.
9. An apparatus comprising:
at least one computing device in communication with a non-transitory storage media including computer executable instructions to perform operations comprising:
determine a portion of traffic between client devices within a first autonomous system and client resolvers within the first autonomous system;
determine distance information between the client devices and edge servers within a second autonomous system, the distance information related to egress of content from at least one of the edge servers via at least one egress gateway of the second autonomous system to the first autonomous system;
determine a network location between the client resolvers and the edge servers using the determined portion of traffic and the determined distance information; and
serve content to at least one of the client devices from at least one of the edge servers based on the determined network location.
10. The apparatus of claim 9 , wherein the portion of traffic is expressed in a client-to-resolver matrix, the distance information is expressed in a client-to-gateway distance matrix, and the network location between the client resolvers and the edge servers is created by a matrix product of the client-to-resolver matrix and the client-to gateway distance matrix.
11. The apparatus of claim 10 , wherein the portion of traffic is a proportion of client devices within a first autonomous system accessing the client resolvers within the first autonomous system, the client-to-resolver matrix is a representation of at least one of a client's use of at least one resolver, and the client-to-gateway distance matrix is a representation of a distance between at least one client device and at least one edge server.
12. The apparatus of claim 11 , wherein a routing table is used to select a edge server for at least one client device.
13. The apparatus of claim 10 , wherein the portion of traffic includes a client metric wherein the client metric is used to populate the client-to-resolver matrix.
14. The apparatus of claim 13 , wherein the client metric is a geographic identifier for a client device.
15. The apparatus of claim 14 , wherein the geographic identifier is a latitude and longitude attribute associated with the client device.
16. The apparatus of claim 9 , wherein the computer executable instructions are further configured to perform the operations of:
sort the network location between the client resolvers and the edge servers based on a distance between each of the edge servers and the client resolvers; and
prefer the edge server based on a shortest distance of the sorted relationship.
17. A computing device in a first autonomous system, the computing device comprising at least one memory for storing instructions that are executed by at least one processor to:
determine a portion of traffic between client devices within the first autonomous system and client resolvers within the first autonomous system;
determine distance information between the client devices and edge servers within a second autonomous system, the distance information related to egress of content from at least one of the edge servers via at least one egress gateway of the second autonomous system to the first autonomous system;
determine a network location between the client resolvers and the edge servers using the determined portion of traffic and the determined distance information; and
serve content to at least one of the client devices from at least one of the edge servers based on the determined network location.
18. The computing device of claim 17 , wherein the portion of traffic is expressed in a client-to-resolver matrix, the distance information is expressed in a client-to-gateway distance matrix, and the network location between the client resolvers and the edge servers is created by a matrix product of the client-to-resolver matrix and the client-to gateway distance matrix.
19. The computing device of claim 18 , wherein the portion of traffic is the proportion of client devices within a first autonomous system accessing the client resolvers within the first autonomous system, the client-to-resolver matrix is a representation of at least one of a client's use of at least one client resolver, and the client-to-gateway distance matrix is a representation of the distance between at least one client device and at least one edge server.
20. The computing device of claim 19 , wherein a routing table is used to select a edge server for at least one client device.