Transparent Header Modification for Reducing Serving Load Based on Current and Projected Usage

• US 20120030341A1
• Filed: 07/28/2010
• Published: 02/02/2012
• Est. Priority Date: 07/28/2010
• Status: Active Grant

• Current Assignee: Daedalus Blue LLC
• Original Assignee: International Business Machines Corporation
• Inventors: Marzorati, Mauro, O'Connell, Brian M., Walker, Keith R., Jensen, Brian W.
• Granted Patent: US 9,363,312 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer
• CPC Class Codes: H04L 67/561 Adding application-functional data or data for application control, e.g. adding metadata
• CPC Class Codes: H04L 67/61 taking into account QoS or priority requirements
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers

Abstract

A method and system for dynamically altering the delivery of web content to end users based on server load. Responsive to receiving a client request for web content, utilization metrics are collected from each device involved in delivering the web content. A device load level is determined for each device based on the utilization metrics, a system load level is determined for a subset of the devices having the same device type based on the device load levels, and a service level to provide to the client is determined based on the system load level. The request header is modified to specify the service level to provide to the client, wherein the service level indicates the web content variant to deliver. The request is sent with the modified header to a selected device which serves the web content according to the service level to the client.

Claims

1. A computer implemented method of delivering web content responsive to a request from a client, comprising:

collecting, in a memory of a computer, utilization metrics from each device in a plurality of devices involved in delivering web content to end users;

determining, by the computer, a device load level for each device based on the utilization metrics;

determining, by the computer, a system load level for a subset of the plurality of devices of a same device type based on the device load level for each device in the subset of devices;

determining, by the computer, a service level to provide to the client based on the system load level for the subset of devices;

modifying, by the computer, a header in the request to specify the determined service level to provide to the client, wherein the determined service level indicates a variant of the web content to deliver to the client; and

sending, by the computer, the request with the modified header to a selected device in the subset of devices, wherein the selected device is directed to serve the web content according to the determined service level to the client.

2. The computer implemented method of claim 1, wherein modifying the header in the request is performed by a layer 7 load balancer that receives the request from the client.

3. The computer implemented method of claim 1, wherein the determining steps of determining a device load level, determining a system load level, and determining a service level is performed on a periodic basis at defined time intervals by a policy engine.

4. The computer implemented method of claim 1, wherein determining a device load level for each device based on the device load metrics further comprises:

applying a device policy to the collected utilization metrics of a device by assigning, based on a value of the utilization metric, a weight value to each utilization metric;

determining the device load level for the device based on the weight values of the utilization metrics and a threshold level specified in the device policy, wherein the threshold level in the device policy specifies whether the device load level is one of a highest assigned weight value, a lowest assigned weight value, or an average of the assigned weight values of the utilization metrics; and

writing the device load level to memory.

5. The computer implemented method of claim 4, wherein determining a system load level for a subset of devices of a same device type based on the device load level for each device in the subset further comprises:

applying a type policy to the device load level determined for each device in the subset of devices by assigning, based on a value of the device load level, a weight value each device load level;

determining the system load level for the subset of devices based on the weight values of the device load levels and a threshold level specified in the type policy, wherein the threshold level in the type policy specifies whether the system load level is one of a highest assigned weight value, a lowest assigned weight value, or an average of the assigned weight values of the device load levels; and

writing the system load level to memory.

6. The computer implemented method of claim 5, wherein determining a service level of web content to deliver to the client based on the system load level further comprises:

applying a service policy to the system load level determined for the subset of devices by assigning, based on a value of the system load level, a variant of the web content; and

writing the service level to memory.

7. The computer implemented method of claim 2, wherein the load balancer modifies the header by:

reading the service level stored in memory;

generating, using transparent content negotiation protocol, a modified header based on the service level, wherein the modified header specifies a variant of the web content to deliver to the client; and

inserting the modified header into the request.

8. The computer implemented method of claim 7, wherein the modified header comprises media feature tags.

9. The computer implemented method of claim 1, wherein the variant of the web content delivered to the client comprises web content delivered at a reduced bit rate stream.

10. The computer implemented method of claim 1, further comprising:

receiving, at the selected device, the request with the modified header;

decoding the modified header;

fetching the variant of the web content specified in the modified header; and

serving the variant of the web content to the client.

11. A computer program product of delivering web content responsive to a request from a client, the computer program product comprising:

a computer readable storage medium;

first program instructions to collect, in a memory of a computer, utilization metrics from each device in a plurality of devices involved in delivering web content to end users;

second program instructions to determine, by a processing unit in the computer, a device load level for each device based on the utilization metrics;

third program instructions to determine a system load level for a subset of the plurality of devices of a same device type based on the device load level for each device in the subset of devices;

fourth program instructions to determine a service level to provide to the client based on the system load level for the subset of devices;

fifth program instructions to modify a header in the request to specify the determined service level to provide to the client, wherein the determined service level indicates a variant of the web content to deliver to the client; and

sixth program instructions to send the request with the modified header to a selected device in the subset of devices, wherein the selected device is directed to serve the web content according to the determined service level to the client,

wherein the first, second, third, fourth, fifth, and sixth program instructions are stored on the computer readable storage medium.

12. The computer program product of claim 11, wherein the fifth program instructions to modify the header in the request is executed by a layer 7 load balancer that receives the request from the client.

13. The computer program product of claim 11, wherein the second program instructions to determine a device load level for each device based on the device load metrics further comprises:

first program sub-instructions to apply a device policy to the collected utilization metrics of a device by assigning, based on a value of the utilization metric, a weight value to each utilization metric;

second program sub-instructions to determine the device load level for the device based on the weight values of the utilization metrics and a threshold level specified in the device policy, wherein the threshold level in the device policy specifies whether the device load level is one of a highest assigned weight value, a lowest assigned weight value, or an average of the assigned weight values of the utilization metrics; and

third program sub-instructions to write the device load level to memory.

14. The computer program product of claim 13, wherein the third program instructions to determine a system load level for a subset of devices of a same device type based on the device load level for each device in the subset further comprises:

first program sub-instructions to apply a type policy to the device load level determined for each device in the subset of devices by assigning, based on a value of the device load level, a weight value each device load level;

second program sub-instructions to determine the system load level for the subset of devices based on the weight values of the device load levels and a threshold level specified in the type policy, wherein the threshold level in the type policy specifies whether the system load level is one of a highest assigned weight value, a lowest assigned weight value, or an average of the assigned weight values of the device load levels; and

third program sub-instructions to write the system load level to memory.

15. The computer program product of claim 14, wherein the fourth program instructions to determine a service level of web content to deliver to the client based on the system load level further comprises:

first program sub-instructions to apply a service policy to the system load level determined for the subset of devices by assigning, based on a value of the system load level, a variant of the web content; and

second program sub-instructions to write the service level to memory.

16. The computer program product of claim 12, wherein the fifth program instructions to modify the header further comprises:

first program sub-instructions to read the service level stored in memory;

second program sub-instructions to generate, using transparent content negotiation protocol, a modified header based on the service level, wherein the modified header specifies a variant of the web content to deliver to the client; and

third program sub-instructions to insert the modified header into the request.

17. The computer program product of claim 16, wherein the modified header comprises media feature tags.

18. The computer program product of claim 11, wherein the variant of the web content delivered to the client comprises web content delivered at a reduced bit rate stream.

19. The computer program product of claim 11, further comprising:

seventh program instructions to receive, at the selected device, the request with the modified header;

eighth program instructions to decode the modified header;

ninth program instructions to fetch the variant of the web content specified in the modified header; and

tenth program instructions to serve the variant of the web content to the client.

20. A computer system for managing web content responsive to a request from a client, comprising:

a processor, a computer readable memory, and a computer readable storage medium;

first program instructions to collect utilization metrics from each device in a plurality of devices involved in delivering web content to end users;

second program instructions to determine a device load level for each device based on the utilization metrics;

third program instructions to determine a system load level for a subset of the plurality of devices of a same device type based on the device load level for each device in the subset of devices;

fourth program instructions to determine a service level to provide to the client based on the system load level for the subset of devices;

fifth program instructions to modify a header in the request to specify the determined service level to provide to the client, wherein the determined service level indicates a variant of the web content to deliver to the client; and

sixth program instructions to send the request with the modified header to a selected device in the subset of devices, wherein the selected device is directed to serve the web content according to the determined service level to the client.

21. The computer system of claim 20, wherein the computer system comprises a load balancer that performs the first, second, third, fourth, fifth, and sixth program instructions, and wherein the computer system comprises the subset of devices, wherein the selected device serves the web content according to the determined service level to the client.

22. A computer implemented method of delivering web content, comprising:

receiving, at a load balancer, a request for web content from a client;

responsive to receiving the request, modifying, by the load balancer, a header in the request to specify a service level to provide to the client, wherein the service level indicates a variant of the web content to deliver to the client; and

sending, by the load balancer, the request with the modified header to a selected device in the subset of devices, wherein the selected device is directed to serve the web content according to the determined service level to the client.