Servlet-based grid computing environment using grid engines and switches to manage resources

• US 8,499,023 B1
• Filed: 03/23/2005
• Issued: 07/30/2013
• Est. Priority Date: 03/23/2005
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Oracle America, Inc. (Oracle Corporation)
• Inventors: Harris, Lyle T.
• Primary Examiner(s): LI, GUANG W
• Application Number:
  US11/087,420
• Time in Patent Office: 3,051 Days
• Field of Search: 709/201, 709/238, 705/26, 705/30, 713/182
• US Class Current: 709/201
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]

Abstract

A method for providing a lightweight grid computing environment. The method includes providing a web server running on a computer device and a grid engine on the web server that is adapted for processing grid compute requests from user programs. The grid engine receives a grid compute request from a user program that is processed by the grid engine to identify a client program available in the grid and a path in the grid to the client program. The method includes loading and running the client program with the grid engine to generate a client program output for transmittal to the user program. The grid compute request includes a URL request string including a name of the client program and the path to the client program and may also include a name of the computer device hosting the web server and a path to the grid engine.

Claims

1. A computer-based method for operating a grid computing environment, comprising:

providing a web server running on a computer device on a digital communications network;

providing a plurality of grid switches for interswitch communication and receiving grid compute requests from a user program and routing the grid compute requests to a grid engine on the web server;

processing the grid compute request with the grid engine to identify a client program and a path to the client program;

loading and running the client program on the computer device to generate a client program output; and

transmitting the client program output to the user program,

wherein the grid compute request comprises a Uniform Resource Locator (URL) request string including a name of the client program and the path of the client program on a client system accessible by the web server via the communications network and

wherein the transmitting of the client program output comprises providing a data stream to a grid end point accessible by the user program based on a URL opened by the grid engine based on the grid compute request.

2. The method of claim 1, wherein the URL request string further includes a name of the computer device containing the web server and a path to the grid engine.

3. The method of claim 1, wherein the path to the client program is a URL.

4. The method of claim 1, wherein the grid compute request further comprises a parameter defining a data set and wherein the loading and running of the client program further comprises loading the defined data set for processing by the client program.

5. A grid computing system, comprising:

a plurality of client programs on client devices accessible on a grid network;

a plurality of data sets stored in memory accessible on the grid network;

a first web server comprising a first grid engine for loading and running a first one of the client programs to process one or more of the data sets in response to a first grid compute request from a user program;

a second web server comprising a second grid engine for loading and running a second one of the client programs to process one or more of the data sets in response to a second grid compute request from a user program; and

a plurality of grid switches for receiving the first and second grid compute requests and routing the first grid compute request to the first grid engine and the second grid compute request to the second grid engine, wherein the first and second grid engines transmit output from the first and second client programs to at least one of the plurality of grid switches, wherein the plurality of grid switches are for interswitch communication,

wherein at least one of the plurality of grid switches transmits the output from the first and second client programs to the corresponding requesting user programs as streams and

wherein the at least one of the plurality grid switches transmits the streams to a grid end point accessible by the corresponding requesting user programs based on a Uniform Resource Locator (URL) opened by the first or second grid engine based on the first or second grid compute request, respectfully.

6. The system of claim 5, wherein the first grid computer request comprises a first Uniform Resource Locator (URL) request string including a name of the first one of the client programs and a path to the first one of the client programs and the second grid computer request comprises a second URL request string including a name of the second one of the client programs and a path to the second one of the client programs.

7. The system of claim 6, wherein the first URL request string further includes a name of a computer device containing the first web server and a path to the first grid engine and the second URL request string further includes a name of a computer device containing the second web server and a path to the second grid engine.

8. The system of claim 5, further comprising a first user device linked to the grid network and running the user program to generate the first grid compute request and second user device linked to the grid network and running the user program to generate the second grid compute request.

9. The system of claim 5, further comprising a public grid domain and a private grid domain, wherein the first grid engine is contained in the public grid domain and the second grid engine is contained in the private grid domain and further wherein the routing performed by at least one of the plurality of grid switches comprises determining that the private grid domain is requested by the second grid compute request and verifying whether the user program corresponding to the second grid compute request can access the private grid domain prior to routing the second grid compute request to the second grid engine.

10. The system of claim 9, wherein the private grid domain is a pay-per-CPU commercial domain and wherein CPU usage to process the second grid compute request is tracked and a financial account corresponding to the second grid compute request is modified to reflect monetary charges associated with the tracked CPU usage.