Universal user input/output application layers

• US 8,819,311 B2
• Filed: 05/23/2007
• Issued: 08/26/2014
• Est. Priority Date: 05/23/2007
• Status: Expired due to Fees

• Current Assignee: RPX Corporation
• Original Assignee: RPX Corporation
• Inventors: Liao, Cheng
• Primary Examiner(s): Tsai, Henry
• Assistant Examiner(s): ROCHE, JOHN B
• Application Number:
  US11/752,914
• Publication Number:
  US 20080295118A1
• Time in Patent Office: 2,652 Days
• Field of Search: 710/50, 719/328
• US Class Current: 710/56
• CPC Class Codes: G06F 2205/066 User-programmable number or size of buffers, i.e. number of separate buffers or their size can be allocated freely
• CPC Class Codes: G06F 2209/542 Intercept
• CPC Class Codes: G06F 9/544 Buffers; Shared memory; Pipes
• CPC Class Codes: G06F 9/545 where tasks reside in different layers, e.g. user- and kernel-space

Abstract

Files on a secondary storage are accessed using alternative IO subroutines that buffer IO requests made by a user and mimic the IO subroutines provided by an operating system. The buffer used by the alternative IO subroutines is maintained by the user and not the operating system. User applications are not recompiled or relinked when using the alternative subroutines because the library that provides these subroutines intercepts requests for buffered IO made by user applications to the operating system'"'"'s IO subroutines and replaces the requests with calls to the alternative IO subroutines that utilize the buffer maintained by the user.

Claims

1. A method of executing a software application for a user, on a computer system executing an operating system, the operating system providing a standard library having buffering subroutines that buffer data transfers between user applications and a secondary storage via system calls to a kernel of the operating system in a non-user controlled buffer, the standard library also having unbuffered subroutines that transfer data, without buffering, between the user applications and the secondary storage via system calls to the kernel of the operating system, the method comprising:

defining a plurality of user subroutines in a user software library distinct from the software application and distinct from the standard library, each user subroutine in the user software library mimicking an input functionality or an output functionality of a corresponding one of the buffering subroutines in the standard library that uses the non-user controlled buffer by utilizing a user-controlled buffer;

intercepting input/output requests by the software application to the standard library buffering subroutines that use the non-user controlled buffer;

determining if the request requires the non-user controlled buffer or the user controlled buffer;

passing the request to the user subroutine that uses the non-user controlled buffer if the request requires the non-user controlled buffer; and

replacing each intercepted input/output request with a request to the user subroutine in the user software library that mimics the buffering subroutine of the intercepted input/output request that uses the non-user controlled buffer if the request requires the user controlled buffer, whereby the user subroutine performs the requested input/output functionality using the user-controlled buffer and an unbuffered subroutine of the standard library rather than the non-user controlled buffer.

2. The method as recited in claim 1, further comprising accessing the secondary storage using the user subroutines.

3. The method as recited in claim 2, wherein said accessing of the secondary storage is performed with simultaneous read and write access to contents thereof.

4. The method as recited in claim 1, wherein the buffering subroutines include open, read, and write access.

5. The method as recited in claim 1, wherein the buffering subroutines include at least one of file stream and file descriptor requests.

6. The method as recited in claim 1, wherein the buffering subroutines include POSIX compliant requests for at least one of input and output.

10. The method as recited in claim 1, wherein intercepting input/output requests includes using an LD_PRELOAD function to dynamically preload the unbuffered user subroutines into the software application before loading the buffering subroutines.

11. The method as recited in claim 1, wherein the plurality of user subroutines further comprises non-vector processing based subroutines.

14. The method as recited in claim 1, wherein the user application comprises a High Performance Computing and High Performance Technical Computing (HPC) application.

7. A computer system having secondary storage, the computer system executing an operating system that provides a standard library having buffering subroutines that buffer data transfers between a user application and the secondary storage via system calls to a kernel of the operating system in a non-user controlled buffer, the standard library also having unbuffered subroutines that transfer data, without buffering, between the user application and the secondary storage via system calls to the kernel of the operating system, the computer system comprising:

a user memory storing a user software library distinct from the application and distinct from the standard library, the software library comprising a plurality of user subroutines, each user subroutine in the software library mimicking an input functionality or an output functionality of a corresponding one of the buffering subroutines in the standard library that uses the non-user controlled buffer by utilizing a user-controlled buffer;

a central processing unit configured to;

intercept a request by the user application to the standard library to perform at least one of the buffering subroutines that use the non-user controlled buffer;

determine if the request requires the non-user controlled buffer or the user controlled buffer;

pass the request to the user subroutine that uses the non-user controlled buffer if the request requires the non-user controlled buffer; and

pass the request to the user subroutine in the user software library that mimics the buffering subroutine of the intercepted request that uses the non-user controlled buffer if the request requires the user controlled buffer, whereby the user subroutine performs the requested input/output functionality using the user-controlled buffer and an unbuffered subroutine of the standard library rather than the non-user controlled buffer;

a network interface connected to at least one remote apparatus, wherein said network interface receives commands from the remote apparatus, the commands causing the central processing unit to alter data stored in a buffer controlled by the user.

8. The computer system as recited in claim 7, wherein said secondary storage operates with simultaneous read and write access.

12. The computer system as recited in claim 7, wherein the plurality of user subroutines further comprise non-vector processing based subroutines.

15. The computer system as recited in claim 7, wherein the user application comprises a High Performance Computing and High Performance Technical Computing (HPC) application.

9. A computer program product for use with a computer system having secondary storage, the computer system executing an operating system that provides a standard library having buffering subroutines that buffer data transfers between a user application and the secondary storage via system calls to a kernel of the operating system in a non-user controlled buffer, the standard library also having unbuffered subroutines that transfer data, without buffering, between the user application and the secondary storage via system calls to the kernel of the operating system, the computer program product comprising a non-transitory computer-readable medium having program code thereon, the program code comprising:

program code for defining a plurality of user subroutines in a user software library distinct from the application and distinct from the standard library, each user subroutine in the software user library mimicking an input functionality or an output functionality of a corresponding one of the buffering subroutines in the standard library that uses the non-user controlled buffer by utilizing a user-controlled buffer;

program code for intercepting input/output requests by the software application to the standard library buffering subroutines that use the non-user controlled buffer; and

program code for determining if the request requires the non-user controlled buffer or the user controlled buffer;

program code for passing the request to the user subroutine that uses the non-user controlled buffer if the request requires the non-user controlled buffer; and

program code for replacing each intercepted input/output request with a request to the user subroutine in the user software library that mimics the buffering subroutine of the intercepted input/output request that uses the non-user controlled buffer if the request requires the user controlled buffer, whereby the user subroutine performs the requested input/output functionality using the user-controlled buffer and an unbuffered subroutine of the standard library rather than the non-user controlled buffer.

13. The computer program product as recited in claim 9, wherein the plurality of user subroutines further comprise non-vector processing based subroutines.

16. The computer program product as recited in claim 9, wherein the user application comprises a High Performance Computing and High Performance Technical Computing (HPC) application.