Method for maintaining data coherency using thread activity summaries in a multicomputer system

• US 5,608,893 A
• Filed: 06/07/1995
• Issued: 03/04/1997
• Est. Priority Date: 07/19/1993
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: Sequent Computer Systems, Inc. (International Business Machines Corporation)
• Inventors: McKenney, Paul E., Slingwine, John D.
• Primary Examiner(s): Kim, Matthew M.
• Application Number:
  US08/480,627
• Time in Patent Office: 636 Days
• Field of Search: 364/200 MS File, 364/900 MS File, 395/200.01, 395/447, 395/448, 395/451, 395/468, 395/469, 395/476, 395/470, 395/471, 395/472, 395/473, 395/375
• US Class Current: 711/141
• CPC Class Codes: G06F 9/52 Program synchronisation; Mutual exclusion, e.g. by means of semaphores
• CPC Class Codes: G06F 9/526 Mutual exclusion algorithms

Abstract

A substantially zero overhead mutual-exclusion apparatus and method (90, 120) is provided that allows concurrent reading and updating data while maintaining data coherency. That is, a data reading process executes the same sequence of instructions that would be executed if the data were never updated. Rather than depending exclusively on overhead-imposing locks, this mutual-exclusion mechanism tracks an execution history (138) of a thread (16, 112) to determine safe times for processing a current generation (108, 130, 131) of data updates while a next generation (110, 132, 133) of data updates is concurrently being saved. A thread is any locus of control, such as a processor. A summary of thread activity (106, 122) tracks which threads have passed through a quiescent state after the current generation of updates was started. When the last thread related to the current generation passes through a quiescent state, the summary of thread activity signals a callback processor (104, 124) that it is safe to end the current generation of updates. The callback processor then processes and erases all updates in the current generation. The next generation of updates then becomes the current generation of updates. The callback processor restarts the summary of thread activity and initiates a new next generation of updates. All data-updating threads pass through a quiescent state between the time they attempt to update data and the time the data are actually updated.

Claims

1. A method for maintaining data coherency of a shared data structure having first and second copies thereof stored in respective first and second computers, comprising the steps of:

storing in the first and second computers respective first and second mutual-exclusion mechanisms each having a thread activity summary and an update processor that processes data structure updates in response to the thread activity summary;

updating the first copy of the data structure with the first mutual-exclusion mechanism, thereby rendering the first and second copies of the data structure incoherent;

transferring the updated first copy of the data structure no the second computer; and

using the second mutual-exclusion mechanism to replace the second copy of the data structure with the updated first copy of the data structure, thereby restoring the coherency of the first and second copies of the data structure.

2. The method of claim 1 further comprising the step of reading an associated copy of the data structure concurrently with either updating or replacing the associated copy of the data structure.

3. The method of claim 1 in which the transferring step is accomplished by any of a physical storage medium, a LAN, a WAN, a system bus, a multi-port memory, or a switched interconnection network.

4. The method of claim 1 in which each of the first and second mutual-exclusion mechanisms has a summary of thread activity configuration that accounts for a computer system architecture of the computer in which it is installed.

5. The method of claim 1 in which each of the first and second mutual-exclusion mechanism operates in a UNIX®

operating system environment, and in which the threads are processors and the thread activity summaries monitor a quiescent state of the respective UNIX®

operating system.

6. The method of claim 1 in which the data structure is one of LAN or WAN data packet distributing data structures.

7. The method of claim 1 in which the data structure is of a distributed lock managing data structure type.

8. The method of claim 7 in which the lock managing data structure includes a lock handle table.

9. The method of claim 8 further including a free handle list.

10. The method of claim 1 wherein the data structure is a packet distributing data structure.

11. The method of claim 1 wherein the shared data structure is adapted to map a handle to the handle'"'"'s associated data structure.

12. A computer-readable medium for storing a computer program comprising instructions for executing the method of claim 1.

13. A computer system in accordance with the method of claim 1.