Content delivery acceleration system

• US 10,015,280 B2
• Filed: 08/18/2016
• Issued: 07/03/2018
• Est. Priority Date: 08/18/2016
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: Greenberg, Don
• Primary Examiner(s): Salad, Abdullahi E
• Application Number:
  US15/241,012
• Publication Number:
  US 20180054499A1
• Time in Patent Office: 684 Days
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/5681 Pre-fetching or pre-delivering data based on network characteristics
• CPC Class Codes: H04L 67/63 Routing a service request depending on the request content or context

Abstract

This document describes a content delivery acceleration system including a communication interface configured to communicate with a remote device over a wireless network; a cache; and a controller configured to perform operations including: pre-computing one or more evaluation results for a webpage associated with a selectable network reference of a webpage, caching the one or more pre-computed evaluation results; retrieving, from the cache, the one or more pre-computed evaluation results; and pre-loading of one or more content tags for the one or more digital component slots, with the one or more content tags being preloaded with uniform resource locators (URLs) from which to load the digital components represented in the pre-computed evaluation results.

Claims

1. A content delivery acceleration system comprising:

a communication interface configured to communicate with a remote device over a wireless network;

a cache; and

a controller configured to perform operations comprising;

pre-computing one or more evaluation results for a webpage, with the one or more pre-computed evaluation results representing one or more digital components for populating one or more digital component slots on the webpage;

caching the one or more pre-computed evaluation results in the cache; and

responsive to receiving a request from a system configured to serve the webpage;

retrieving, from the cache, the one or more pre-computed evaluation results; and

pre-loading one or more content tags for the one or more digital component slots with one or more uniform resource locators (URLs) from which to retrieve the one or more digital components represented in the one or more pre-computed evaluation results.

2. The content delivery acceleration system of claim 1, wherein pre-loading the one or more content tags comprises loading the one or more content tags with the one or more URLs in advance of receipt of a request from a publisher system for one or more evaluation results.

3. The content delivery acceleration system of claim 1, wherein pre-computing comprises computing the one or more evaluation results in advance of receipt of a request from a publisher system for the one or more evaluation results.

4. The content delivery acceleration system of claim 1, the operations further comprising:

requesting, based on the one or more pre-computed evaluation results, a digital component of the one or more digital components represented by the one or more pre-computed evaluation results for populating the one or more digital component slots on the webpage;

receiving the digital component; and

pre-loading the digital component for a digital component slot of the one or more digital components slots on the webpage so that a latency in serving the webpage is reduced, relative to a latency in serving the webpage independent of pre-loading the digital component.

5. The content delivery acceleration system of claim 4, wherein pre-loading of the digital component enables the digital component and webpage data of the webpage to be served in a single service.

6. The content delivery acceleration system of claim 4, wherein the digital component comprises an advertisement.

7. The content delivery acceleration system of claim 1, wherein the webpage comprises a first webpage and wherein the operations further comprise:

receiving webpage data from a remote system, the webpage data being for a second webpage and comprising one or more selectable network references to one or more third webpages and the first webpage;

parsing the received webpage data for the one or more selectable network references; and

for a particular one of the one or more selectable network references, generating, based on the parsed webpage data, prediction data representing a likelihood that the particular one of the one or more selectable network references representing the first webpage is selected.

8. The content delivery acceleration system of claim 7, wherein generating the prediction data comprises:

receiving a user identifier and a webpage identifier for the first webpage;

receiving data comprising one or more of user click data for the first webpage, webpage data for the first webpage, and content provider data;

analyzing the received data using one or more machine-learning algorithms; and

returning a prediction value associated with a user for the first webpage, the prediction data comprising the prediction value.

9. The content delivery acceleration system of claim 7, wherein the webpage data for the second webpage comprises one or more search results based on a search query of a user, and wherein the one or more selectable network references comprise one or more clickable links to one or more additional webpages represented by the one or more search results.

10. The content delivery acceleration system of claim 7, wherein pre-computing the one or more evaluation results comprises:

determining a threshold value for comparing the prediction data, the threshold value representing a specified likelihood the particular one of the one or more selectable network references is selected; and

comparing the threshold value to the prediction data for the particular one of the one or more selectable network references.

11. The content delivery acceleration system of claim 1, wherein pre-computing the one or more evaluation results representing the one or more digital components comprises:

generating callout parameters data using prior evaluation data; and

performing an evaluation for each of the one or more digital component slots of the webpage based on the callout parameters data.

12. The content delivery acceleration system of claim 1, the operations further comprising:

associating, with a user identifier, the one or more pre-computed evaluation results representing the one or more digital component slots associated with the webpage;

caching the user identifier;

returning, in response to a request for the one or more pre-computed evaluation results for the webpage from a user associated with the user identifier, the one or more pre-computed evaluation results associated with the user identifier; and

deleting, from the cache, the one or more pre-computed evaluation results associated with the user identifier.

13. A method for acceleration of content delivery, the method comprising:

pre-computing one or more evaluation results for a webpage, with the one or more pre-computed evaluation results representing one or more digital components for populating one or more digital component slots on the webpage;

caching the one or more pre-computed evaluation results in a cache; and

responsive to receiving a request from a system configured to serve the webpage;

retrieving, from the cache, the one or more pre-computed evaluation results; and

pre-loading one or more content tags for the one or more digital component slots with one or more uniform resource locators (URLs) from which to retrieve the one or more digital components represented in the one or more pre-computed evaluation results.

14. The method of claim 13, wherein pre-loading the one or more content tags comprises loading the one or more content tags with the one or more URLs in advance of receipt of a request from a publisher system for one or more evaluation results.

15. The method of claim 13, wherein pre-computing comprises computing the one or more evaluation results in advance of receipt of a request from a publisher system for the one or more evaluation results.

16. The method of claim 13, the operations further comprising:

requesting, based on the one or more pre-computed evaluation results, a digital component of the one or more digital components represented by the one or more pre-computed evaluation results for populating the one or more digital component slots on the webpage;

receiving the digital component; and

pre-loading the digital component for a digital component slot of the one or more digital components slots on the webpage so that a latency in serving the webpage is reduced, relative to a latency in serving the webpage independent of pre-loading the digital component.

17. The method of claim 16, wherein pre-loading of the digital component enables the digital component and webpage data of the webpage to be served in a single service.

18. The method of claim 16, wherein the digital component comprises an advertisement.

19. The method of claim 13, wherein the webpage comprises a first webpage and wherein the operations further comprise:

receiving webpage data from a remote system, the webpage data being for a second webpage and comprising one or more selectable network references to one or more third webpages and the first webpage;

parsing the received webpage data for the one or more selectable network references; and

for a particular one of the one or more selectable network references, generating, based on the parsed webpage data, prediction data representing a likelihood that the particular one of the one or more selectable network references representing the first webpage is selected.

20. The method of claim 19, wherein generating the prediction data comprises:

receiving a user identifier and a webpage identifier for the first webpage;

receiving data comprising one or more of user click data for the first webpage, webpage data for the first webpage, and content provider data;

analyzing the received data using one or more machine-learning algorithms; and

returning a prediction value associated with a user for the first webpage, the prediction data comprising the prediction value.

21. The method of claim 19, wherein the webpage data for the second webpage comprises one or more search results based on a search query of a user, and wherein the one or more selectable network references comprise one or more clickable links to one or more additional webpages represented by the one or more search results.

22. The method of claim 19, wherein pre-computing the one or more evaluation results comprises:

determining a threshold value for comparing the prediction data, the threshold value representing a specified likelihood the particular one of the one or more selectable network references is selected; and

comparing the threshold value to the prediction data for the particular one of the one or more selectable network references.

23. The method of claim 13, wherein pre-computing the one or more evaluation results representing the one or more digital components comprises:

generating callout parameters data using prior evaluation data; and

performing an evaluation for each of the one or more digital component slots of the webpage based on the callout parameters data.

24. The method of claim 13, the operations further comprising:

associating, with a user identifier, the one or more pre-computed evaluation results representing the one or more digital component slots associated with the webpage;

caching the user identifier;

returning, in response to a request for the one or more pre-computed evaluation results for the webpage from a user associated with the user identifier, the one or more pre-computed evaluation results associated with the user identifier; and

deleting, from the cache, the one or more pre-computed evaluation results associated with the user identifier.

25. One or more non-transitory machine-readable media comprising instructions that are executable to perform operations comprising:

pre-computing one or more evaluation results for a webpage, with the one or more pre-computed evaluation results representing one or more digital components for populating one or more digital component slots on the webpage;

caching the one or more pre-computed evaluation results in a cache; and

responsive to receiving a request from a system configured to serve the webpage;

retrieving, from the cache, the one or more pre-computed evaluation results; and

pre-loading one or more content tags for the one or more digital component slots with one or more uniform resource locators (URLs) from which to retrieve the one or more digital components represented in the one or more pre-computed evaluation results.

26. The one or more non-transitory machine-readable media of claim 25, wherein pre-loading the one or more content tags comprises loading the one or more content tags with the one or more URLs in advance of receipt of a request from a publisher system for one or more evaluation results.

27. The one or more non-transitory machine-readable media of claim 25, wherein pre-computing comprises computing the one or more evaluation results in advance of receipt of a request from a publisher system for the one or more evaluation results.

28. The one or more non-transitory machine-readable media of claim 25, the operations further comprising:

requesting, based on the one or more pre-computed evaluation results, a digital component of the one or more digital components represented by the one or more pre-computed evaluation results for populating the one or more digital component slots on the webpage;

receiving the digital component; and

pre-loading the digital component for a digital component slot of the one or more digital components slots on the webpage so that a latency in serving the webpage is reduced, relative to a latency in serving the webpage independent of pre-loading the digital component.

29. The one or more non-transitory machine-readable media of claim 28, wherein pre-loading of the digital component enables the digital component and webpage data of the webpage to be served in a single service.

30. The one or more non-transitory machine-readable media of claim 25, wherein the webpage comprises a first webpage and wherein the operations further comprise:

receiving webpage data from a remote system, the webpage data being for a second webpage and comprising one or more selectable network references to one or more third webpages and the first webpage;

parsing the received webpage data for the one or more selectable network references; and

for a particular one of the one or more selectable network references, generating, based on the parsed webpage data, prediction data representing a likelihood that the particular one of the one or more selectable network references representing the first webpage is selected.

31. The one or more non-transitory machine-readable media of claim 30, wherein generating the prediction data comprises:

receiving a user identifier and a webpage identifier for the first webpage;

receiving data comprising one or more of user click data for the first webpage, webpage data for the first webpage, and content provider data;

analyzing the received data using one or more machine-learning algorithms; and

returning a prediction value associated with a user for the first webpage, the prediction data comprising the prediction value.

32. An electronic system for content delivery acceleration comprising:

a communication interface configured to communicate with a remote device over a wireless network;

a cache; and

means for pre-computing one or more evaluation results for a webpage, with the one or more pre-computed evaluation results representing one or more digital components for populating one or more digital component slots on the webpage;

means for caching the one or more pre-computed evaluation results in the cache; and

responsive to receiving a request from a system configured to serve the webpage;

means for retrieving, from the cache, the one or more pre-computed evaluation results; and

means for pre-loading one or more content tags for the one or more digital component slots with one or more uniform resource locators (URLs) from which to retrieve the one or more digital components represented in the one or more pre-computed evaluation results.