Clustered computer system utilizing separate servers for redundancy in which the host computers are unaware of the usage of separate servers

• US 7,155,638 B1
• Filed: 01/17/2003
• Issued: 12/26/2006
• Est. Priority Date: 01/17/2003
• Status: Active Grant

• Current Assignee: Unisys Corporation
• Original Assignee: Unisys Corporation
• Inventors: Heideman, Michael J., Novak, David A., Konrad, Dennis R., Crandall, Carl R.
• Primary Examiner(s): Bonzo, Bryce P.
• Application Number:
  US10/346,933
• Time in Patent Office: 1,439 Days
• Field of Search: 714/6, 714/11, 714/12
• US Class Current: 714/11
• CPC Class Codes: G06F 11/2005 using redundant communication controllers
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2043 where the redundant components share a common memory address space
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources

Abstract

An apparatus for and method of enhancing reliability within a cluster lock processing system having a relatively large number of commodity cluster instruction processors which are managed by a cluster lock manager. Because the commodity processors have virtually no system viability features such as memory protection, failure recovery, etc., the cluster/lock processors assume the responsibility for providing these functions. The low cost of the commodity cluster instruction processors makes the system almost linearly scalable. The cluster/locking, caching, and mass storage accessing functions are fully integrated into a single hardware platform which performs the role of the master. Upon failure of this hardware platform, a second redundant hardware platform converts from slave to master role. The logic for the failure detection and role swapping is placed within software, which can run as an application under a commonly available operating system. Furthermore, the recovery is completely accomplished without assistance of the Host computer(s) or ultimate user(s) coupled via the Host computer(s).

Claims

1. An apparatus comprising:

a. describing means for describing a service request;

b. a plurality of commodity cluster instruction processors;

c. first managing means responsively coupled to said describing means and to each of said plurality of commodity cluster instruction processors for managing said plurality of commodity cluster instruction processors to honor said service request;

d. permitting means responsively coupled to said first managing means for permitting said first managing means to assign said describing means to a selected one of said plurality of commodity cluster instruction processors for said honoring of said service request;

e. redundant managing means responsively coupled to each of said plurality of commodity cluster instruction processors for managing said plurality of commodity cluster instruction processors upon failure of said first managing means; and

f. switching means responsively coupled to said first managing means and said redundant managing means for switching from said first managing means to said redundant managing means without notification to said describing means.

2. An apparatus according to claim 1 wherein each of said describing means further comprises a first host computer.

3. An apparatus according to claim 2 wherein a first portion of said switching means is resident in said first managing means and a second portion of said switching means is resident in said redundant managing means.

4. An apparatus according to claim 3 further comprising a second host computer responsively coupled to said first managing means and said redundant managing means.

5. An apparatus according to claim 4 wherein said first host computer and said second host computer further comprises Unisys OS2200 mainframe computers.

6. An apparatus comprising:

a. a plurality of commodity cluster instruction processors;

b. a primary cluster lock server having a first copy of a software cluster lock manager responsively coupled to each of said plurality of commodity cluster instruction processors;

c. a secondary cluster lock server having a second copy of said software cluster lock manager responsively coupled to each of said plurality of commodity cluster instruction processors;

d. a first host computer separate and independent of said plurality of commodity cluster instruction processors responsively coupled to said primary cluster lock server via said first copy of said software cluster lock manager and said secondary cluster lock server via said second copy of said software cluster lock manager which submits a service request; and

e. a component responsively coupled to said primary cluster lock server and secondary cluster lock server which substitutes said secondary cluster lock server for said primary cluster lock server without assistance from said host computer.

7. The apparatus of claim 6 wherein said primary cluster lock server further comprises a dedicated cache memory.

8. The apparatus of claim 7 wherein each of said primary cluster lock server and said secondary lock server further comprise Windows operating system.

9. The apparatus of claim 8 further comprising a second host computer separate and independent of said plurality of commodity cluster instruction processors responsively coupled to said primary cluster lock server via said first copy of said software cluster lock manager.

10. The apparatus of claim 9 wherein said second host computer is responsively coupled to said secondary cluster lock server via said second copy of said software cluster lock manager.

11. A method of enhancing availability of a data processing system comprising:

a. assigning a first cluster lock server the role of master;

b. assigning a second cluster lock server the role of slave;

c. transferring a first service request from a first host to said second cluster lock server via a second copy of a software cluster lock manager located within said second cluster lock server;

d. transferring said first service request from said second cluster lock server to said first cluster lock server;

e. checking operation of said first cluster lock server using software within said first cluster lock server, said second cluster lock server, and a first control platform;

f. switching said first cluster lock server to the role of slave without assistance of host; and

g. switching said second cluster lock server to the role of master without notice to said first host thereby permitting said first host to continue to communicate with said second cluster lock server via said second copy of said software cluster lock manager.

12. A method according to claim 11 further comprising transferring a second service request from said first host to said first cluster lock server via a second copy of said software cluster lock manager located within said first cluster lock server.

13. A method according to claim 12 further comprising re-transferring said second service request from said first cluster lock server to said second cluster lock server after said second switching step.

14. A method according to claim 13 further comprising transferring a third service request from a second host to said first cluster lock server.

15. A method according to claim 14 further comprising re-transferring said third service request from said first cluster lock server to said second cluster lock server after said second switching step.

16. A data processing system comprising:

a. a plurality of commodity cluster instruction processors;

b. a first cluster lock server responsively coupled to each of said plurality of commodity cluster instruction processors having a first internal cache memory and a first Input/Output processor and having a first copy of a software cluster lock manager;

c. a second cluster lock server responsively coupled to each of said plurality of commodity cluster instruction processors having a second internal cache memory and a second Input/Output processor and having a second copy of said software cluster lock manager;

d. a first host computer separate and independent of said plurality of commodity cluster instruction processors responsively coupled to said first cluster lock server via said first copy of said software cluster lock manager and said second cluster lock server via said second copy of said software cluster lock manager which initiates a first service request and transfers it to said second cluster lock server; and

e. a facility responsively coupled to said first cluster lock server and to said second cluster lock server which is capable of switching said first cluster lock server to functions of said second cluster lock server and switching said second cluster lock server to functions of said first cluster lock server.

17. The improvement according to claim 16 wherein said first cluster lock server and said second cluster lock server further comprise a Windows operating system.

18. The improvement according to claim 17 wherein said first cluster lock server and said second cluster lock server are located at different geographical sites.

19. The improvement according to claim 18 further comprising a second host computer separate and independent of said plurality of commodity cluster instruction processors responsively coupled to said first cluster lock server and said second cluster lock server.

20. The improvement according to claim 19 wherein said first host computer and said second host computer are not co-located.