METHOD AND SYSTEM FOR CLIENT-SIDE SCALING OF WEB SERVER FARM ARCHITECTURES IN A CLOUD DATA CENTER

• US 20100268764A1
• Filed: 02/01/2010
• Published: 10/21/2010
• Est. Priority Date: 04/15/2009
• Status: Active Grant

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture Global Services Limited (Accenture PLC)
• Inventors: LIU, Huan, WEE, Sewook
• Granted Patent: US 8,433,749 B2
• Time in Patent Office: Days
• US Class Current: 709/203
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: H04L 43/04 Processing captured monitoring data, e.g. for logfile generation
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• 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

Abstract

A novel web server farm architecture is provided that combines various cloud components and innovatively maximizes their strengths to achieve a scalable, adaptable, load balanced computing architecture. In one embodiment, this architecture includes a storage host for static content hosting, a cluster of virtual machines (operating as web servers) for dynamic content hosting, and a dedicated application engine for monitoring the load of the virtual machines. A browser or end-application user of a web application is first routed to a static anchor page hosted on the storage host. The browser is subsequently presented with a list of virtual machines and data corresponding to their respective current loads. To access dynamic content, the browser is able to select from the virtual machines based in large part on the reported current loads of the virtual machines. Once a virtual machine is selected, the browser may communicate directly with the selected web server.

Claims

1. A system for load-balancing a web server farm, the system comprising:

an application executing on a plurality of cloud infrastructure components, the plurality of cloud infrastructure components comprising;

a static host configured to host static content corresponding to the application;

a plurality of web servers communicatively coupled to the static host and configured to host dynamic content corresponding to the application; and

an application engine communicatively coupled to the plurality of web servers and configured to collect load information corresponding to the plurality of web servers,

wherein, based on the load information, a client browser selects an available web server from the plurality of web servers to access dynamic content corresponding to the application.

2. The system of claim 1, wherein the static host comprises a scalable online storage web service.

3. The system according to claim 1, wherein at least one of the web servers of the plurality of web servers comprises a virtual private server instance.

4. The system according to claim 1, wherein the plurality of web servers comprises a plurality of dynamic pages.

5. The system according to claim 4, wherein the static host comprises a plurality of static anchor pages corresponding to the plurality of dynamic pages.

6. The system according to claim 5, wherein load information for a web server is stored in the static host.

7. The system according to claim 6, wherein, for a dynamic page of the plurality of dynamic pages hosted on an individual web server of the plurality of web servers, load information of the individual web server is stored in a static anchor page of the plurality of static anchor pages corresponding to the dynamic page.

8. The system according to claim 5, wherein a static anchor page of the plurality of static anchor pages further comprises:

a server file; and

a load balance file.

9. The system according to claim 1, wherein the load information comprises:

a list of available web servers from the plurality of web servers;

the internet protocol (IP) addresses corresponding to the list of available web servers; and

respective current loads corresponding to individual web servers comprised in the list of available web servers.

10. The system according to claim 9, wherein a respective current load corresponding to an individual web server comprises a current usage of at least one of:

the CPU, memory, and network bandwidth.

11. The system according to claim 9, wherein a probability of a client browser selecting a web server from the plurality of web servers to access dynamic content is substantially equivalent to an inverse of the relative current load corresponding to the web server.

12. The system according to claim 1, wherein an initial contact between the client browser and the application is established in the static host.

13. The system according to claim 1, wherein the application engine is further configured to automatically make scaling decisions based on load information corresponding to the plurality of web servers.

14. The system according to claim 1, wherein the application engine is further configured to provide a visualization of a performance history of the plurality of web servers.

15. The system according to claim 1, wherein, in response to being selected by the client browser, the available web browser communicates directly with the client browser.

16. A method for hosting an application in a cloud data center to facilitate client-side scaling, the method comprising:

hosting static content for an application in a storage host;

hosting dynamic content for the application in a plurality of web servers;

collecting load information corresponding to the plurality of web servers, the load information being collected by a dedicated application engine;

receiving an initial contact request from a client browser;

presenting load information for the plurality of web servers to the client browser.

receiving a selection, based on the load information, from a client browser of a selected web server of the plurality of web servers to access the dynamic content; and

facilitating a communication between the client browser and the selected web server,

wherein the storage host, the plurality of web servers and the dedicated application engine comprise components in a cloud computing infrastructure.

17. The method according to claim 16, further comprising:

automatically scaling the plurality of web servers based on the load information.

18. The method according to claim 16, wherein the presenting load information comprises:

providing a list of available web servers comprising at least a portion of the plurality of web servers to the client browser;

presenting internet protocol (IP) addresses corresponding to the list of available web servers; and

relaying respective current loads corresponding to individual web servers comprised in the list of available web servers.

19. The method according to claim 16, wherein the facilitating a communication comprises providing an IP address corresponding to the selected web server to the client browser.

20. The method according to claim 16, wherein the hosting dynamic content comprises hosting a plurality of dynamic pages.

21. The method according to claim 20, wherein the hosting the static content comprises hosting a plurality of static anchor pages, wherein a static anchor page of the plurality of static anchor pages corresponds to a dynamic page of the plurality of dynamic pages.

22. The method according to claim 16, wherein the receiving an initial contact between the client browser and the application comprises receiving a communication request from a client browser in the storage host;

23. The method according to claim 16, further comprising:

providing a visualization of a performance history of the plurality of web servers.

24. The method according to claim 16, wherein the collecting load information is performed by a dedicated application engine communicatively coupled to the plurality of web servers.

25. A method for establishing a connection between a client-side browser and a target web server to facilitate client-side scaling of a web server farm architecture in a cloud data center, the method comprising:

receiving a communication request from a client browser in a static anchor page of a website corresponding to an application and comprised in a storage host;

in response to receiving the communication request, loading the static anchor page in the client browser, the static anchor page comprising a server file and a load balance file;

determining, based on load information generated by a dedicated application engine, a target web server from a plurality of web servers to send a dynamic content request, the target web server hosting a dynamic content page corresponding to the application;

sending a data request corresponding to the dynamic content request to the target web server;

processing the data request in the target web server; and

returning the result of processing the data transfer request to the client browser,

wherein the storage host, the plurality of web servers and the dedicated application engine comprise components in a cloud computing infrastructure.

26. The method according to claim 25, wherein the determining a target web server comprises:

examining the load balance file to derive respective current loads of the plurality of web servers;

selecting a web server from the plurality of web servers as the target web server based on the respective current loads.

27. The method according to claim 25, wherein the determining a target web server comprises randomly choosing a web server from the plurality of web servers, wherein the probability of choosing any web server from the plurality of web servers is weighted to correspond to relative load of the web server.

28. The method according to claim 27, wherein the probability of choosing any web server of the plurality of web servers is substantially equivalent to an inverse proportion of the relative load of the web server.

29. The method according to claim 25, wherein the sending the data request comprises:

sending the data request to a proxy hosted on the target web server;

constructing a new data request from data comprised in the data request;

sending the new data request to the target web server;

30. The method according to claim 29, wherein the returning the result of processing comprises sending the result from the web server to the proxy.

31. The method according to claim 30, wherein the returning the result of the processing comprises sending the result from the proxy to the client browser.

32. The method according to claim 29, wherein the proxy is hosted on the virtual machine.

33. The method according to claim 25, wherein the sending the data request comprises sending a data request comprising:

a browser cookie associated with the website; and

a plurality of parameters corresponding to the browser.

34. The method according to claim 25, wherein the data request comprises an http request.

35. The method according to claim 34, wherein the sending the data request further comprises sending POST data corresponding to the http request.

36. The method according to claim 35, wherein the POST data corresponding to the request comprises:

a URL corresponding to the website; and

a message body specifying information for an action being requested.

37. The method according to claim 25, wherein the processing the data transfer request comprises invoking a dynamic script processor.

38. The method according to claim 25, wherein the result of processing comprises a javascript file.

39. The method according to claim 25, further comprising executing the result of processing in the client browser.

40. The method according to claim 39, wherein the executing the result in the client browser comprises:

in the client browser, executing a javascript file comprised in the result; and

updating a page display corresponding to the website in the client browser.

41. The method according to claim 40, wherein the result of processing further comprises a set-cookie header.

42. The method according to claim 41, wherein executing the result in the client browser further comprises:

sending the set-cookie header to the client browser; and

updating a browser cookie corresponding to the website in response to receiving a set-cookie header.

43. The method according to claim 25, wherein the target web server comprises an instance of a virtual machine.