Enhanced domain name translation in content delivery networks
Systems, methods, apparatuses, and software for operating content delivery networks are provided herein. In one example, a method of operating a domain name translation node in a first point-of-presence of a content delivery network is presented. The method includes receiving a translation message issued by an end user device for translation of a domain name into a content network address, and processing the translation message to identify a network address of a node that transferred the translation message. The method also includes selecting the content network address based at least in part on correlations between network addresses and performance factors to direct the end user device to a target cache node at a point-of-presence different than the point-of-presence of the domain name translation node, and transferring a response message indicating the content network address which directs the end user device to the target cache node at the second point-of-presence.
1. A method of operating a content delivery network, the method comprising:
receiving, in a first local domain name server (DNS) in a first point-of-presence (PoP), a message from a first regional DNS, wherein the message comprises a request to translate a domain name to a network address of a cache node that caches content requested by an end user device;
in the first local DNS, determining to serve the content to the end user device from a cache node in a second PoP, wherein the second PoP includes a second local DNS; and
providing the network address of the cache node from the first local DNS to the end user device.
2. The method of claim 1 , further comprising, prior to providing the end user device with the network address of the cache node, translating the domain name to the network address of the cache node.
3. The method of claim 2 , wherein providing the end user device with the network address of the cache node comprises:
in the first local DNS, redirecting the message to a second local DNS in the second PoP; and
in the second local DNS, providing the end user device with the network address of the cache node.
4. The method of claim 1 , further comprising:
identifying a network address of the first regional DNS; and
identifying the cache node based on the network address of the first regional DNS.
5. The method of claim 1 , further comprising:
evaluating performance factors of multiple PoPs based at least on latencies of domain name translations; and
generating a data structure relating network addresses of the multiple PoPs and the performance factors.
6. The method of claim 5 , wherein determining to serve the content to the end user device from a cache node in a second PoP is based on the data structure.
7. The method of claim 1 , further comprising:
embedding code in content delivered to end user device that, when executed by the end user device, causes the end user device to monitor data latency statistics;
receiving the data latency statistics from the end user device; and
determining the cache node to serve the content based on the latency statistics.
8. A method of operating a first local domain name system (DNS) in a first point-of-presence (PoP), the method comprising:
receiving, at the first local DNS, a request from an end user device to translate a domain name to a network address of a cache server that stores content requested by the end user device;
determining, at the first local DNS, that the content should be served to the end user device from a cache server in a second PoP, wherein the second PoP includes a second local DNS;
providing the network address of the cache server in the second PoP to the end user device from the first local DNS.
9. The method of claim 8 , wherein providing the network address of the cache server in the second PoP to the end user device does not involve the second local DNS.
10. The method of claim 8 , wherein the first PoP and the second PoP are in the same content delivery network and each comprise cache nodes, wherein the cache nodes cache content originating from origin servers.
11. The method of claim 8 , further comprising:
identifying a network address of the first regional DNS; and
determining that content should be served to end user device from the cache server in the second PoP based on the network address of the first regional DNS.
12. The method of claim 8 , further comprising:
evaluating relative performances of a plurality of PoPs based at least on latencies of domain name translations; and
generating a data structure relating network addresses of the plurality of PoPs and their relative performances.
13. The method of claim 12 , wherein determining that the content should be served to the end user device from the cache server in the second PoP is based on the data structure.
14. The method of claim 8 , further comprising:
embedding code in content delivered to end user device that, when executed by the end user device, causes the end user device to monitor data latency statistics;
receiving the data latency statistics from the end user device; and
determining the cache node to serve the content based on the latency statistics.
15. A content delivery network comprising:
a means for receiving a message at a first local domain name server (DNS) of a first point-of-presence (PoP), wherein the message comprises a request to translate a domain name to a network address of a cache node that caches content requested by an end user device;
a means for determining to serve the content to the end user device from a cache node in a second PoP, wherein the second PoP includes a second local DNS; and
a means for providing the network address of the cache node from the first PoP to the end user device, wherein the cache node comprises a server in the second PoP.
16. The content delivery network of claim 15 , further comprising a means for translating the domain name to the network address of the cache node prior to providing the end user device with the network address of the cache node.
17. The content delivery network of claim 16 , wherein the means for providing the end user device with the network address of the cache node comprises a means in the first PoP for providing the end user device with the network address, without involving the second local DNS.
18. The content delivery network of claim 15 , further comprising:
a means for identifying a network address of a first regional DNS from which the message is received at the first PoP; and
a means for identifying the cache node based on the network address of the first regional DNS.
19. The content delivery network of claim 15 , further comprising:
a means for evaluating performance factors of multiple PoPs based at least on latencies of domain name translations; and
a means for generating a data structure relating network addresses of the multiple PoPs and the performance factors.
20. The content delivery network of claim 19 , wherein determining to serve the content to the end user device from a cache node in a second PoP is based on the data structure.