Achieving autonomic behavior in an operating system via a hot-swapping mechanism

• US 7,533,377 B2
• Filed: 09/29/2003
• Issued: 05/12/2009
• Est. Priority Date: 09/29/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Appavoo, Jonathan, Hui, Kevin Kin-Fai, Da Silva, Dilma Menezes, Auslander, Marc Alan, Krieger, Orran Yaakov, Wisniewski, Robert William
• Primary Examiner(s): VU, TUAN A
• Application Number:
  US10/673,587
• Publication Number:
  US 20050071811A1
• Time in Patent Office: 2,052 Days
• Field of Search: 717/108, 717/110, 717128-148, 717/168, 719/314, 719/321, 719328-332, 714/4, 714/5, 714/10, 718100-107, 711/114, 711/152, 711/200, 711/202, 710/260, 710/302, 713/2
• US Class Current: 717/168
• CPC Class Codes: G06F 8/656 while running

Abstract

Systems, especially operating systems, are becoming more complex to the point where maintaining them by humans is becoming nearly impossible. Many corporations have recognized this trend and have begun investing in autonomic technology. Autonomic technology allows a piece of software to monitor, diagnose, and repair itself. This can be used for improved performance, reliability, maintainability, security, etc. Disclosed herein is a mechanism to allow operating systems to hot swap a piece of operating system code, while continuing to offer to the user the service which that code is providing. This can be used, for examples, to increase the performance of an application or to fix a detected security hole live without bringing the machine down. Some autonomic ability will be mandatory in next generation operating system for without it they will collapse under their own complexity. The invention offers a key component of being able to achieve autonomic computing.

Claims

1. In a computer system using an operating system to provide access to hardware resources, wherein said operating system provides access to said hardware resources via a first code component, a method of replacing said first code component with a new code component while said operating system remains active and while said operating system provides continual availability to the hardware resources by applications operational in the computer system, and while said application call to said first component to perform defined functions, the method comprising using the computer system for:

instantiating an instance of a new code component to replace said first code component;

instantiating a mediator object;

initiating a swapping operation;

for the duration of the swapping operation, using the mediator object to intercept calls from the applications requesting the first code components to perform the defined functions, and using the mediator object to perform the requested functions;

establishing a quiescent state for the first code component;

transferring state from the first code component to the new code component; and

swapping the new code component for the first code component including the steps of;

identifying references to said first code component;

replacing the identified references to said first code component with references to said new code component; and

deallocating the first code component.

2. A method according to claim 1, wherein the method is implemented transparently to said applications.

3. A method according to claim 1, wherein the method is scalable.

4. A method according to claim 1, wherein said computer system comprises a multiprocessor system, which comprises a plurality of processors, and said method is implemented across each of said processors independently.

5. A method according to claim 1, wherein the replacing step includes the steps of:

transferring said identified references at said established quiescent state from the first code component to the new code component; and

after transferring said identified references to the new code segment at said established quiescent state, swapping the first code component with the new code component.

6. A method according to claim 1, wherein the step of identifying references includes the steps of:

separating the first code component into objects; and

grouping said objects into a table, and arranging said table to identify said references to said objects, which identified references to said objects are entered in the table.

7. A method according to claim 1, wherein:

the establishing step includes the step of;

establishing a quiescent state for the first code component, without locking the first code component, by tracking active threads to the first code component, and identifying the active threads as said identified references; and

the replacing step includes the steps of;

transferring said identified references, during the quiescent state, from the first code component to the new code component; and

after transferring said identified references, swapping the first code component with the new code component.

8. A method according to claim 1, wherein the replacing step includes the steps of:

establishing a quiescent state for the first code component that includes the identified references;

transferring the identified references at the quiescent state from the first code component to the new code component by providing an infrastructure that operates to negotiate a best transfer algorithm; and

after transferring said identified references at the quiescent state, swapping the first code component with the new code component.

9. A method according to claim 1, wherein:

the step of identifying references includes the steps of separating the first code component into objects, and

grouping said objects into a table, and arranging said table to identify said references to said objects, which identified references to said objects are entered in the table; and

the replacing step includes the steps of

establishing a quiescent state for the first code component that includes said identified references;

transferring said references identified to said objects for the first code component to the new code component by providing an infrastructure to negotiate a best transfer algorithm; and

after transferring said references to the new code segment, swapping the first code component with the new code component.

10. A system for swapping code in a computer system including an operating system, said operating system including at least one code component and providing continual availability of hardware resources by applications operational in the computer system, the system comprising one or more processors configured for:

instantiating an instance of a new source code component to replace said at least one code component;

instantiating a mediator object;

initiating a swapping operation;

for the duration of the swapping operation. using the mediator object to intercept calls from the applications requesting the first code components to perform the defined functions, and using the mediator object to perform the requested functions;

establishing a quiescent state for the at least one code component;

transferring state from the at least one code component to the new source code component; and

swapping the new source code component for the at least one code component, including;

identifying, while said operating system is active and providing continual access to said resources, references to the at least one code component of the operating system; and

transferring the identified references to the new source code component, while said operating system is active and providing continual access to said resources, and then executing the swapping.

11. A system according to claim 10, wherein the system operates transparently to said applications and the system is scalable.

12. A system according to claim 10, wherein said computer system comprises a multiprocessor system, which comprises a plurality of processors, and said method is implemented across each of said processors independently.

13. A system according to claim 10, wherein the transferring is done by:

transferring said references at the established quiescent state from the at least one code component to the new code component; and

means for swapping the at least one code component with the new code component after said identified references have been transferred.

14. A system according to claim 10, wherein the identifying references is done by:

separating the at least one code component into objects; and

grouping said objects into a table, and arranging said table to identify said references to said objects, which identified references to said objects are entered in the table.

15. A system according to claim 10, wherein the transferring is done by:

establishing a quiescent state for the at least one code component, without locking the at least one code component, by tracking active threads to the at least one code component, and identifying the active threads as said identified references;

transferring said identified references, during the quiescent state, from the at least one code component to the new code component; and

swapping the at least one code component with the new code component after said identified quiescent state references have been transferred.

16. A system according to claim 10, wherein the transferring is done by:

establishing a quiescent state for the at least one code component that includes the identified references;

transferring the identified references, during the quiescent state, from the at least one code component to the new code component by providing an infrastructure to negotiate a best transfer algorithm; and

swapping the at least one code component with the new code component after said identified references have been transferred.

17. A system according to claim 10, wherein:

the identifying references is done by;

i) separating the at least one code component into objects, and

ii) grouping said objects into a table, and arranging said table to identify said references to said objects, which identified references to said objects are entered in the table; and

the transferring is done by;

i) establishing a quiescent state for the at least one code component;

ii) transferring the identified references during the quiescent state, from the at least one code component to the new code component by providing an infrastructure to negotiate a best transfer algorithm; and

iii) swapping the at least one code component with the new code component after said identified references have been transferred.

18. A program storage device, for use with a computer system including an operating system to provide access to hardware resources, wherein said operating system provides access to said resources via a first code component, said program storage device being readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for replacing said first code component with a new code component while said operating system remains active and while said operating system provides continual availability to said resources by applications operational in the computer system, and while said applications call to said first component to perform defined functions, the method steps comprising:

instantiating an instance of a new code component to replace said first code component;

instantiating a mediator object;

initiating a swapping operation;

for the duration of the swapping operation, using the mediator object to intercept calls from the applications requesting the first code components to perform the defined functions, and using the mediator object to perform the requested functions;

establishing a quiescent state for the first code component;

transferring state from the first code component to the new code component; and

swapping the new code component for the first code component including the steps of;

identifying references to said first code component;

replacing the identified references to said first code component with references to said new code component; and

deallocating the first code component.

19. A program storage device according to claim 18, wherein the method is implemented transparently to said applications, the method is scalable, and said computer system comprises a multiprocessor system, which comprises a plurality of processors, and said method is implemented across each of said processors independently.

20. A program storage device according to claim 18, wherein the replacing step includes the steps of:

establishing a quiescent state for the first code component;

transferring said identified resources at said established quiescent state from the first code component to the new code component; and

after transferring said identified references to the new code segment at said established quiescent state, swapping the first code component with the new code component.

21. A program storage device according to claim 18, wherein the step of identifying references includes the steps of:

separating the first code component into objects; and

grouping said objects into a table, and arranging said table to identify said references to said objects, which identified references to said objects are entered in the table.

22. A program storage device according to claim 18, wherein the replacing step includes the steps of:

establishing a quiescent state for the first code component, without locking the first code component, by tracking active threads to the first code component, and identifying the active threads as said identified references;

transferring said identified references, during the quiescent state, from the first code component to the new code component; and

after transferring said identified references, swapping the first code component with the new code component.

23. A program storage device according to claim 18, wherein the replacing step includes the steps of:

transferring the identified references at the quiescent state from the first code component to the new code component by providing an infrastructure that operates to negotiate a best transfer algorithm; and

after transferring the identified at said quiescent state, swapping the first code component with the new code component.

24. A program storage device according to claim 18, wherein:

the step of identifying references includes the steps of

i) separating the first code component into objects, and

ii) grouping said objects into a table, and arranging said table to identify references to said objects, which are entered in the table; and

the replacing step includes the steps of

i) establishing a quiescent state for the first code component that includes said identified references;

ii) transferring said references identified to said objects for the first code component to the new code component by providing an infrastructure to negotiate a best transfer algorithm; and

iii) after transferring said identified references, swapping the first code component with the new code component.