Web request broker controlling multiple processes

• US 6,845,505 B1
• Filed: 02/03/1997
• Issued: 01/18/2005
• Est. Priority Date: 02/03/1997
• Status: Expired due to Term

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Ng, Raymond, Pang, Robert, Anand, Mala, Sharma, Ankur, Chou, Tsung-Jen, Bookman, Matthew, Adunuthula, Seshu, Nakhoda, Shehzaad
• Primary Examiner(s): LAO, SUE X
• Application Number:
  US08/794,269
• Time in Patent Office: 2,906 Days
• Field of Search: 395/674, 395/675, 395/682, 395/684, 395/200.56, 395/200.33, 707/10, 709/104, 709/105, 709/302, 709/304, 709/226, 709/203, 709/102, 709/201, 709/206, 709/219, 709/225, 709/229, 709/238, 709/239, 718/104, 718/105, 718/102
• US Class Current: 718/105
• CPC Class Codes: G06F 21/00 Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
• CPC Class Codes: G06F 2209/5018 Thread allocation
• CPC Class Codes: G06F 9/4843 by program, e.g. task dispatcher, supervisor, operating system
• CPC Class Codes: G06F 9/5055 considering software capabilities, i.e. software resources associated or available to the machine
• CPC Class Codes: H04L 61/00 Network arrangements, protocols or services for addressing or naming
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]

Abstract

A web server configured to respond to client requests over a network such as the World Wide Web includes a web listener having a Hypertext Transfer Protocol (HTTP) daemon, a plurality of extension programs configured to perform respective operations, and a web request broker configured to identify one of the programs for responding to a client request, and determine the availability of an instance of the identified program. The web request broker maintains control of multiple instances of each server extension program to provide enhanced server operation without overwhelming server resources. The web request broker maintains a minimum number of instances of the identified program in memory, each executed in its own address space. The web request broker determines whether an available instance of the identified program is available from an existing number of instances, and selectively initiates a new instance of the program if no other instance is available. If no instance is available and the existing number of instances exceeds the maximum prescribed number, then the web request broker returns the reply to the web listener to send a reply over the network that the request was not processed.

Claims

1. A method for responding to a request issued to a server from a client over a network system, the method comprising the steps of:

said server obtaining the request over a network;

said server identifying, from a plurality of program types available to said server, a program type that corresponds to the request, wherein each program type of said plurality of program types implements different functionality than the other program types of said plurality of program types;

determining whether an instance of said program type currently exists that is not currently processing a previously received request;

if no currently executing instance of said program type exists that is not currently processing a previously received request, then said server selectively initiating an instance of the program type based on a comparison between how many instances of said program type are currently executing and a prescribed number of instances of said program type;

said server dispatching the request to said instance of said program type;

executing said instance to cause said instance to process said request; and

responding to the request based on execution of said instance.

2. The method of claim 1, wherein:

the step of obtaining the request over a network is performed by a server process executing in a first address space; and

the step of selectively initiating an instance of the program type includes initiating said instance in a second address space that is separate from the first address space.

3. The method of claim 1, further comprising the step of, after executing said instance, maintaining said instance in memory for executing a subsequent request.

4. The method of claim 1, further comprising the steps of:

detecting a fault in the instance during execution thereof; and

aborting the instance in response to detecting said fault.

5. The method of claim 1, wherein the prescribed number of said instances specifies at least one of a maximum and a minimum number of instances, wherein each instance executes within an address space that is separate from the address spaces used by other instances of said program type.

6. The method of claim 5, wherein:

the step of selectively initiating an instance of the program type includes the step of initiating a specified minimum number of instances of said program type prior to obtaining the request over the network; and

the method further includes the step of, after obtaining the request over the network, determining if one of the minimum number of instances is available for processing said request.

7. The method of claim 6 further including the step of, after obtaining the request over the network, initiating a new instance of the program type if none of the minimum number of instances is available for processing said request and the number of instances is less than the maximum number of instances.

8. The method of claim 6, wherein the step of executing said instance includes executing an available instance from the minimum number of instances.

9. The method of claim 1, wherein the step of selectively initiating an instance of the program type includes the step of returning the request to the network without processing said request when a current number of instances of said program type exceeding a maximum value associated with said program type.

10. The method of claim 1 further comprising the step of registering a plurality of program types with the server, wherein registration of each of said program types includes specifying a maximum number of instances for said program type, and a virtual path for said program type.

11. The method of claim 10, wherein:

the request identifies a virtual path; and

the step of identifying a program type that corresponds to the request includes the step of identifying the program type based on the virtual path identified in the request.

12. The method of claim 1, wherein:

the step of obtaining the request over a network comprises receiving the request from a transport protocol process operating according to a first protocol; and

the method includes the step of converting the request to a second protocol independent from the first protocol.

13. A method, for execution by a server, for responding to a request for performance of an operation, the method comprising:

obtaining the request over a network;

forwarding the request to a dispatcher executed by the server; and

processing the request by causing the dispatcher to perform the steps of;

A) determining whether an available instance exists of a program type configured to handle the request, wherein the available instance is among an existing number of instances of the program type and is not currently processing a previously received request,

B) if an available instance is available, then dispatching the request for execution by the available instance,

C) if no instance is available, then initiating a new instance of the program type for execution of the request if the existing number of instances does not exceed a maximum prescribed number,

D) if no instance is available and the existing number of instances exceeds the maximum prescribed number, then sending said reply over the network indicating the request was not processed, and

E) registering with the dispatcher a plurality of program types configured to handle respective types of requests, the step of registering with the dispatcher a plurality of said program types including the step of storing in the dispatcher for each of said program types a maximum number of instances and a virtual path specifying an address location associated with the corresponding program type.

14. The method of claim 13, further comprising the step of:

receiving, by the dispatcher, a reply from the instance executing the request; and

sending information contained in the reply over the network from the dispatcher to a client that issued the request.

15. The method of claim 13, further comprising the step of:

detecting by the dispatcher a fault in the instance executing the request; and

terminating by the dispatcher the instance executing the request in response to the detected fault.

16. The method of claim 15, further comprising the step of sending a reply over the network indicating the request was not processed.

17. The method of claim 13, wherein each instance is executed within an address space that is separate from the address spaces used by other instances of said program type.

18. The method of claim 17, wherein the step of initiating a new instance of the program type comprises the step of initiating the new instance within an address space that is separate from the address spaces used by the other instances of said program type.

19. A method, for execution by a server, for responding to a request for performance of an operation, the method comprising:

obtaining the request over a network;

forwarding the request to a dispatcher executed by the server; and

processing the request by causing the dispatcher to perform the steps of;

A) determining whether an available instance of a program type configured to handle the request is available among an existing number of instances of the program type, wherein the available instance is not currently processing a previously received request,

B) if an available instance is available, then dispatching the request for execution by the available instance,

C) if no instance is available, then initiating a new instance of the program type for execution of the request if the existing number of instances does not exceed a maximum prescribed number,

D) if no instance is available and the existing number of instances exceeds the maximum prescribed number, then sending said reply over the network indicating the request was not processed, and

E) delaying deallocation of said new instance at least a predetermined time interval after processing the request for processing a subsequent request, and automatically deallocating said new instance if no subsequent request has arrived after said predetermined time interval.

20. A method, for execution by a server, for responding to a request for performance of an operation by a particular program type, the method comprising:

obtaining the request over a network;

forwarding the request to a dispatcher executed by the server; and

processing the request by causing the dispatcher to perform the steps of;

A) determining whether an available instance exists of a program type configured to handle the request, wherein the available instance is among an existing number of instances of the program type and is not currently processing a previously received request,

B) if an available instance is available, then dispatching the request for execution by the available instance,

C) if no instance is available, then initiating a new instance of the program type for execution of the request if the existing number of instances does not exceed a maximum prescribed number,

D) if no instance is available and the existing number of instances exceeds the maximum prescribed number, then sending said reply over the network indicating the request was not processed,

wherein the step of obtaining the request over the network comprises receiving the request from a transport protocol process operating according to a first protocol, said first protocol being one of a plurality of protocols; and

the method further comprises the step of converting the request to a second protocol independent from the first protocol by a transport adapter configured to convert to said second protocol requests received according to any of said first plurality of protocols.

21. The method of claim 20, wherein the existing number of said instances is at least a prescribed minimum number of said instances.

22. The method of claim 20, further comprising the steps of:

causing the dispatcher to determine if a program type is configured to handle the request based on the operation specified in the request; and

if no program type is configured to handle the request, then sending a reply over the network indicating the request was not processed.

23. The method of claim 20, wherein:

the step of obtaining the request over the network comprises the step of executing by the server a server process; and

the step of processing the request by causing the dispatcher to perform said steps comprises the step of executing by the server a plug-in routine, added to the server process, to cause the dispatcher to perform said steps.

24. A computer readable medium having stored thereon sequences of instructions for responding to a request for performance of an operation received by a server, the sequences of instructions including instructions for performing the steps of:

obtaining the request over a network;

identifying a program type, from a plurality of program types, a program type that corresponds to the request, wherein each program type of said plurality of program types implements different functionality than the other program types of said plurality of program types;

determining whether an instance of said program type currently exists that is not currently processing a previously received request;

if no currently executing instance of said program type exists that is not currently processing a previously received request, then selectively initiating an instance of the program type based on a comparison between how many instances of said program type are currently executing and a prescribed number of instances of said program type;

dispatching the request to said instance of said program type;

executing said instance to cause said instance to process said request; and

responding to the request based on execution of said instance.

25. The computer readable medium of claim 24, further comprising sequences of instructions for performing the steps of:

detecting a fault in the instance during execution thereof; and

aborting the instance in response to detecting said fault.

26. The computer readable medium of claim 24, wherein the prescribed number of said instances specifies at least one of a maximum and a minimum number of instances, wherein each instance executes within an address space that is separate from the address spaces used by other instances of said program type.

27. A computer server configured to respond to a request for performance of an operation, comprising:

a network listener configured to receive the request over a network and send a response to the request over the network, the request having a prescribed object type specifying an operation to be performed;

a plurality of program types, each program type configured to perform an operation that generates an output in response to receiving a request having a corresponding object type specifying the operation performed by the program type, each program type having a prescribed number of instances executing at respective address spaces; and

a dispatcher plug-in configured to identify one of the program types for responding to the request based on the prescribed object type, the dispatcher plug-in selectively dispatching the request to an available instance of the identified program type based upon the corresponding prescribed number of instances, wherein the available instance of the identified program type is selected to receive said request based on a set of factors that includes that said available instance is not currently processing a previously received request, the dispatcher plug-in sending said response to the network listener based on execution of the request by the available instance.

28. The server of claim 27, wherein the network listener receives and sends the request and the response based on Hypertext Transfer Protocol (HTTP).

29. The server of claim 28, wherein the network listener includes an HTTP daemon configured to output a static Hypertext Markup Language (HTML) page in response to the prescribed object type specifying sending said static HTML page.

30. The server of claim 27, further comprising a transport adapter configured to convert the request received by the network listener from a first protocol to a second protocol independent from the first protocol, the transport adapter converting the response output by the dispatcher plug-in from said second protocol to the first protocol before sending by the network listener.

31. The server of claim 27, further comprising a plurality of execution engines, each execution engine configured to control execution of the instances of the corresponding program types and provide an Application Programming Interface (API) specifying predetermined operations to be performed by the instances of the corresponding program type, the execution engine corresponding to the identified one program receiving the request from the dispatcher plug-in and controlling execution of the available instance processing the request.

32. The server of claim 31, wherein the predetermined operations specified by the API includes at least one of initialization, execution of the request, and shutdown by at least one of the instances of the corresponding program.

33. The server of claim 27, wherein each execution engine initiates a prescribed minimum number of said instances of the corresponding program type.

34. The server of claim 27, wherein the dispatcher plug-in includes a configuration library identifying for each of the program types the corresponding object type and the prescribed number of instances.

35. The server of claim 27, wherein the object type of the request includes a virtual path specifying the identified one program type.