Adaptive Packaging of Network Resources

• US 20110044354A1
• Filed: 08/18/2009
• Published: 02/24/2011
• Est. Priority Date: 08/18/2009
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Wei, Xiaoliang
• Granted Patent: US 8,874,694 B2
• Time in Patent Office: Days
• US Class Current: 370/468
• CPC Class Codes: H04L 41/0816 the condition being an adaptation, e.g. in response to network events
• CPC Class Codes: H04L 41/0826 for reduction of network costs H04L41/0833 takes precedence
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 43/0864 Round trip delays
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/83 based on usage prediction
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

In one embodiment, a method includes accessing a data store comprising a log of requests transmitted by a plurality of remote hosts, each request corresponding to one or more resources maintained by a web site; analyzing the requests and one or more attributes of the resources corresponding to the respective requests using a cost function relating to network performance, wherein a first component of the cost function is directed to a performance improvement achieved by grouping one or more of the resources for transmission to one or more remote hosts and a second component is directed to a performance inefficiency caused by grouping the one or more resources for transmission to the one or more remote hosts; and identifying one or more groupings of one or more resources that optimize the cost function for network performance.

Claims

1. A method comprising:

accessing, by one or more computing devices, a data store comprising a log of requests transmitted by a plurality of remote hosts, each request corresponding to one or more resources of a plurality of resources;

analyzing the requests and one or more attributes of the one or more resources corresponding to the respective requests using a cost function relating to network performance, wherein a first component of the cost function is directed to a performance improvement achieved by grouping one or more of the resources of the plurality of resources for transmission to one or more remote hosts and a second component is directed to a performance inefficiency caused by grouping the one or more resources of the plurality of resources for transmission to the one or more remote hosts; and

identifying one or more groupings of one or more resources of the plurality of resources that optimize the cost function for network performance.

2. The method of claim 1, further comprising:

generating a matrix based on the log of requests, wherein each element in the matrix indicates whether a particular resource of the plurality of resources is required for a particular request;

wherein analyzing the requests and one or more attributes of the one or more resources corresponding to the respective requests using a cost function relating to network performance comprises using the cost function to operate on the matrix.

3. The method of claim 2, wherein a logical 1 in a particular element indicates that the corresponding resource is required for the particular request, and wherein a logical 0 in a particular element indicates that the corresponding resource is not required for the particular request.

4. The method of claim 1, wherein the cost function operates on all possible sets of groupings of resources corresponding to the log of requests transmitted by the plurality of remote hosts over a period of time.

5. The method of claim 1, wherein the cost function when operated on a set of groupings of resources determines a collective gain for the set of groupings that represents a sum of the gains for each grouping in the set of groupings, and wherein the cost function when operated on the set of groupings of resources determines a collective cost for the set of groupings that represents a sum of the costs for each grouping in the set of groupings.

6. The method of claim 1, wherein the gain for each grouping represents a sum of the gains for each request in the grouping, and wherein the cost for each grouping represents a sum of the costs for each request in the grouping.

7. The method of claim 6, wherein the output of the cost function for each set of groupings of resources is equal to the collective gain for the set of groupings minus the collective cost for the set of groupings.

8. The method of claim 7, wherein the gain for each grouping of resources for a given request comprises a gain that represents an aggregate round trip time that would have been saved in transmitting the resources in the grouping in one package as compared to packaging each of the resources in the grouping individually if each of the resources in the grouping of resources were actually used to serve the request.

9. The method of claim 7, wherein the cost for each grouping of resources for a given request comprises a cost that represents an aggregate amount of unused data that would have been transmitted in transmitting the resources in the grouping in one package compared to packaging each of the resources in the grouping individually divided by a bandwidth value.

10. The method of claim 1, further comprising repeating the method of claim 1 for a user on a periodic or per-session basis to re-optimize the groupings for the user.

11. The method of claim 1, wherein ones of the plurality of resources comprise a resource required for a particular feature or application.

12. The method of claim 1, wherein at least one of the plurality of resources comprises a JavaScript script.

13. The method of claim 1, wherein at least one of the plurality of resources comprises a Cascading Style Sheet (CSS) resource.

14. The method of claim 1, wherein at least one of the plurality of resources comprises an image resource.

15. The method of claim 1, further comprising generating Personal Home Page (PHP) code that determines which resource or resources correspond to each request.

16. The method of claim 1, further comprising considering a likelihood that a packaged resource will be changed in the cost function.

17. One or more computer-readable tangible storage media encoding software that is operable when executed to:

access a data store comprising a log of requests transmitted by a plurality of remote hosts, each request corresponding to one or more resources of a plurality of resources;

analyze the requests and one or more attributes of the one or more resources corresponding to the respective requests using a cost function relating to network performance, wherein a first component of the cost function is directed to a performance improvement achieved by grouping one or more of the resources of the plurality of resources for transmission to one or more remote hosts and a second component is directed to a performance inefficiency caused by grouping the one or more resources of the plurality of resources for transmission to the one or more remote hosts; and

identify one or more groupings of one or more resources of the plurality of resources that optimize the cost function for network performance.

18. The media of claim 17, further operable when executed to:

generate a matrix based on the log of requests, wherein each element in the matrix indicates whether a particular resource of the plurality of resources is required for a particular request;

wherein the media operable to analyze the requests and one or more attributes of the one or more resources corresponding to the respective requests using a cost function relating to network performance comprises media operable when executed to use the cost function to operate on the matrix.

19. The media of claim 17, wherein the cost function operates on all possible sets of groupings of resources corresponding to the log of requests transmitted by the plurality of remote hosts over a period of time.

20. The media of claim 17, wherein the cost function when operated on a set of groupings of resources determines a collective gain for the set of groupings that represents a sum of the gains for each grouping in the set of groupings, and wherein the cost function when operated on the set of groupings of resources determines a collective cost for the set of groupings that represents a sum of the costs for each grouping in the set of groupings.

21. The media of claim 17, wherein the gain for each grouping represents a sum of the gains for each request in the grouping, and wherein the cost for each grouping represents a sum of the costs for each request in the grouping, and wherein the output of the cost function for each set of groupings of resources is equal to the collective gain for the set of groupings minus the collective cost for the set of groupings.

22. The media of claim 21, wherein the gain for each grouping of resources for a given request comprises a gain that represents an aggregate round trip time that would have been saved in transmitting the resources in the grouping in one package as compared to packaging each of the resources in the grouping individually if each of the resources in the grouping of resources were actually used to serve the request.

23. The media of claim 21, wherein the cost for each grouping of resources for a given request comprises a cost that represents an aggregate amount of unused data that would have been transmitted in transmitting the resources in the grouping in one package compared to packaging each of the resources in the grouping individually divided by a bandwidth value.

24. A system comprising:

one or more processors; and

logic encoded in one or more computer-readable tangible storage media that, when executed by the one or more processors, is operable to;

access a data store comprising a log of requests transmitted by a plurality of remote hosts, each request corresponding to one or more resources of a plurality of resources;

analyze the requests and one or more attributes of the one or more resources corresponding to the respective requests using a cost function relating to network performance, wherein a first component of the cost function is directed to a performance improvement achieved by grouping one or more of the resources of the plurality of resources for transmission to one or more remote hosts and a second component is directed to a performance inefficiency caused by grouping the one or more resources of the plurality of resources for transmission to the one or more remote hosts; and

identify one or more groupings of one or more resources of the plurality of resources that optimize the cost function for network performance.

25. The system of claim 24, wherein the logic is further operable when executed to:

generate a matrix wherein each column of the matrix corresponds to a particular resource and each row of the matrix corresponds to a particular logged request in the data store, wherein each element in the matrix defines whether or not the resource corresponding to the column comprising the element is required for the particular request corresponding to the row comprising the element;

wherein the media operable to analyze the requests and one or more attributes of the one or more resources corresponding to the respective requests using a cost function relating to network performance comprises media operable when executed to use the cost function to operate on the matrix.

26. The system of claim 24, wherein the cost function operates on all possible sets of groupings of resources corresponding to the log of requests transmitted by the plurality of remote hosts over a period of time.

27. The system of claim 24, wherein the cost function when operated on a set of groupings of resources determines a collective gain for the set of groupings that represents a sum of the gains for each grouping in the set of groupings, and wherein the cost function when operated on the set of groupings of resources determines a collective cost for the set of groupings that represents a sum of the costs for each grouping in the set of groupings.

28. The system of claim 24, wherein the gain for each grouping represents a sum of the gains for each request in the grouping, and wherein the cost for each grouping represents a sum of the costs for each request in the grouping, and wherein the output of the cost function for each set of groupings of resources is equal to the collective gain for the set of groupings minus the collective cost for the set of groupings.

29. The system of claim 28, wherein the gain for each grouping of resources for a given request comprises a gain that represents an aggregate round trip time that would have been saved in transmitting the resources in the grouping in one package as compared to packaging each of the resources in the grouping individually if each of the resources in the grouping of resources were actually used to serve the request.

30. The system of claim 28, wherein the cost for each grouping of resources for a given request comprises a cost that represents an aggregate amount of unused data that would have been transmitted in transmitting the resources in the grouping in one package compared to packaging each of the resources in the grouping individually divided by a bandwidth value.