Application server architecture

• US 7,689,660 B2
• Filed: 06/09/2005
• Issued: 03/30/2010
• Est. Priority Date: 06/09/2005
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Kilian, Frank, Rohland, Hans-Christoph
• Primary Examiner(s): Srivastava; Vivek
• Assistant Examiner(s): Liu; Lin
• Application Number:
  US11/149,562
• Publication Number:
  US 20060282509A1
• Time in Patent Office: 1,755 Days
• Field of Search: 709/213, 709/214, 707/9, 707/100, 707/102
• US Class Current: 709/213
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066
• CPC Class Codes: H04L 67/142 Managing session states for stateless protocols; Signalling session states; State transitions; Keeping-state mechanisms
• CPC Class Codes: H04L 67/563 Data redirection of data network streams

Abstract

Embodiments of the invention are generally directed to a system and method for an application server architecture having a common connection manager for at least two heterogeneous application server software suites. One of the software suites is a standards-based application server software suite having a plurality of worker nodes. In one embodiment, the plurality of worker nodes share code and data that is stored in a shared memory.

Claims

1. A system comprising:

a processor configured to execute instructions for an application server;

a connection manager configured to receive a request from a client, the connection manager including an integrated dispatcher, the connection manager to select which one of at least two heterogeneous application server software suites to process the request and to provide the request to the selected software suite, the application server software suites including;

a first standards-based application server software suite coupled with the connection manager, the first application server software suite having a first plurality of worker nodes each configured to receive and process the request from the client, the integrated dispatcher to dispatch requests to the first plurality of worker nodes, and

a second proprietary application server software suite coupled with the connection manager, the second application server software suite having a second plurality of worker nodes each configured to receive and process the request from the client;

a shared memory coupled with the processor, the shared memory including at least a first shared memory region to support data transfers to and from the first application server software suite and to and from the second application server software suite, the first plurality of worker nodes and the second plurality of worker nodes having access to the first shared memory region to receive and store content; and

a set of request notification queues, the set including a request notification queue for each worker node of the first plurality of worker nodes, wherein the set of request notification queues and the shared memory provide for transfer of a session from a first worker node to a second worker node upon a failure of the first worker node prior to servicing a request notification.

2. The system of claim 1, wherein the shared memory includes a second shared memory region to store session objects having session state information for the first plurality of worker nodes of the first application server software suite.

3. The system of claim 2, further comprising a third shared memory region to store session state information regarding the routing of requests, the third shared memory being a session table shared memory, the session state information including a value representing a number of outstanding active requests for a session for which responses have not been generated by worker nodes.

4. The system of claim 3, wherein the dispatcher is to place information identifying a selected worker node and identifying a request notification queue for the selected worker node into a session table entry of the session table shared memory.

5. The system of claim 2, wherein the second shared memory region is a closure based shared memory.

6. The system of claim 5, wherein the first plurality of worker nodes includes a plurality of virtual machines, the plurality of virtual machines to share memory in the second shared memory region, the second shared memory region being limited to shareable closures.

7. The system of claim 6, wherein a number of application threads running on each of the one or more virtual machines is limited to a threshold value, and wherein a number of application threads per CPU (central processing unit) of the system is limited to a fixed number that is based at least in part on an input/output ratio of an application.

8. The system of claim 5, wherein the shared memory further includes a cache region to be shared by the first plurality of worker nodes.

9. The system of claim 8, wherein the cache region to be shared by the first plurality of worker nodes is a region of the shared closure based shared memory.

10. The system of claim 5, wherein the first shared memory region is a connection oriented shared memory.

11. The system of claim 10, wherein the connection oriented shared memory is based, at least in part, on a fast channel architecture.

12. The system of claim 11, wherein the connection oriented shared memory is a request/response shared memory, the request response shared memory coupled with the connection manager, the first application server software suite and the second application server software suite.

13. The system of claim 1, wherein the shared memory is to store shared class loaders, the shared class loaders accessible by the first plurality of worker nodes, wherein the shared class loaders are not modified after the shared class loaders are created.

14. The system of claim 1, wherein the shared memory is to store at least one of:

thread status information corresponding to a selected worker node of the first plurality of worker nodes; and

session status information corresponding to the selected worker node of the first plurality of worker nodes.

15. The system of claim 14, further comprising:

a monitoring console configured to obtain at least one of session status information and thread status information from the shared memory.

16. The system of claim 1, wherein any of a plurality of protocol handlers may be plugged into the connection manager to support any of a plurality of protocols.

17. The system of claim 1, wherein the first application server software suite is compatible with J2EE™

(Java™

2 Platform, Enterprise Edition).

18. A method comprising:

receiving a request from a client at a connection manager, the connection manager including an integrated first dispatcher;

selecting one of a plurality of heterogeneous application software suites for execution by a processor to receive the request, the application server software suites including;

a first standards-based application server software suite coupled with the connection manager, the first application server software suite having a first plurality of worker nodes each configured to receive and process the request from the client, wherein the integrated first dispatcher is configured to operate as a dispatcher for the first plurality of worker nodes, and

a second proprietary application server software suite coupled with the connection manager, the second application server software suite having a second plurality of worker nodes each configured to receive and process the request from the client;

transferring the request from the connection manager to a dispatcher associated with the selected application software suites;

selecting a first worker node of the selected application software suite to receive the request;

dispatching the request from the dispatcher associated with the selected application software suite to the first worker node;

storing a notification for the request in a first request notification queue of a set of request notification queues if the first worker node is a worker node of the first plurality of worker nodes, the set including a request notification queue for each worker node of the first plurality of worker nodes, wherein the set of request notification queues and the shared memory provide for transfer of a session from the first worker node to a second worker node upon a failure of the first worker node prior to servicing a request notification; and

receiving data for the request from the dispatcher at the first worker node via a shared memory, the shared memory including a first shared memory region to support data transfers to and from the first application server software suite and to and from the second application software suite, the first and second plurality of worker nodes having access to the first shared memory region to receive and store content, the first worker node to process the request.

19. The method of claim 18, wherein the first memory region is a connection oriented shared memory, and wherein dispatching the request to the first worker node comprises:

reading the request data from a network layer interface into the connection oriented shared memory.

20. The method of claim 18, wherein the shared memory further includes a second shared memory region to store session objects having session state information for the first plurality of worker nodes, the second shared memory region being a closure based shared memory, further comprising:

providing a cache from the closure based shared memory, wherein the cache is visible to the first plurality of worker nodes.

21. The method of claim 18, further comprising:

generating status information for a thread assigned to the selected first worker node; and

storing the status information in shared memory.

22. The method of claim 18, further comprising:

generating session context information for a session between the client providing the request and the first worker node; and

storing the session status information in the shared memory.

23. An article of manufacturer comprising computer-readable medium providing instructions that, when executed by a processor, cause the processor to:

receive a request from a client at a connection manager, the connection manager including an integrated first dispatcher;

select one of a plurality of heterogeneous application software suites for execution by a processor to receive the request, the application server software suites including;

a first standards-based application server software suite coupled with the connection manager, the first application server software suite having a first plurality of worker nodes each capable configured to receive and process the request from the client, wherein the integrated first dispatcher is configured to operate as a dispatcher for the first plurality of worker nodes, and

a second proprietary application server software suite coupled with the connection manager, the second application server software suite having a second plurality of worker nodes each capable configured to receive and process the request from the client;

transfer the request from the connection manager to a dispatcher associated with the selected application software suite;

select a first worker node of the selected application software suite to receive the request;

dispatching the request from the dispatcher for the selected application software suite to the first worker node;

store a notification for the request in a first request notification queue of a set of request notification queues if the first worker node is a worker node of the first plurality of worker node, the set including a request notification queue for each worker node of the first plurality of worker nodes, wherein the set of request notification queues and the shared memory provide for transfer of a session from the first worker node to a second worker node of the first plurality of worker nodes upon a failure of the first worker node prior to servicing a request notification; and

receive data for the request from the dispatcher for the selected application software suite at the first worker node via a shared memory, the shared memory including a first shared memory region to support data transfers to and from the first application server software suite and to and from the second application software suite, the first plurality of worker nodes and the second plurality of worker nodes having access to the first shared memory region to receive and store content, the first worker node to process the request.

24. The article of manufacture of claim 23, wherein the shared memory further includes a second shared memory region to store session objects having session state information for the first plurality of worker nodes, the second shared memory region being a closure based shared memory, and wherein the computer-readable medium provides further instructions that, when executed by the processor, cause the processor to:

provide a cache from the closure based shared memory, wherein the cache is visible to the first plurality of worker nodes.

25. The article of manufacture of claim 23, wherein the computer-readable medium provides further instructions that, when executed by the processor, cause the processor to:

generate status information for a thread assigned to the selected first worker node; and

store the status information in the shared memory.