Processing by use of synchronized tuple spaces and assertions

Abstract

A system for controlling and coordinating activities among entities in an information and process environment comprising a communications pathway for transmitting and receiving communications of the entities, and a shared memory connected to the communications pathway for maintaining a tuple space on which the entities post and receive messages synchronized to discrete time intervals.

Description

CROSS-REFERENCE TO RELATED APPLICATION [0001] This application is a Continuation of U.S. application Ser. No. 09 / 642,340, filed Aug. 21, 2000, which claims priority to United Kingdom Application No. 9920051.1 filed on Aug. 24, 1999.FIELD OF THE INVENTION [0002] This invention relates in general to the field of distributed information processing, resource allocation, computing and communications and more specifically to a system and method of organizing tuple spaces by synchronization to coordinate activities among multiple entities. BACKGROUND OF THE INVENTION [0003] Distributed information processing involves two or more processing entities under autonomous control or purpose which operate cooperatively to resolve a processing need. Such entities frequently operate within a system and are frequently subsets of larger systems tailored to address a specific processing aspect, but may originate remotely making demands on the local system. In complex systems, entities frequently compet...

AI Technical Summary

Benefits of technology

This approach enables safe interaction and addition of new features or components in complex systems, preventing system degradation by ensuring entities are aware of potential impacts before proceeding with actions, thus maintaining system stability and efficiency.

Problems solved by technology

In complex systems, entities frequently compete and share resources within the environment and the operation of one entity may affect, degrade, inhibit, or even destabilize the operation of another.

One of the problems in distributed information processing and computing is the coordination of activities among the various entities in the environment.

While the coordination problem manifests itself in distributed environments, it also appears in other contexts such as integrated multi-component computing systems where coordination is required between multiple components which may reside within the same cabinet or in service environments where resources must be shared.

One example of this problem is the creation of features in telephone switching systems.

Telephone switching systems are typically very large.

This presents the problem that the additional functions of new features interact with the functioning of features already present in the system.

In the worst case, the newly added features can disable the functioning of an older feature, destabilize the system or create conditions which will confuse the user.

Because of this problem, previous approaches have expended considerable effort during the development of new features to determine all possible interactions with all of the other features in the system.

A second example of the problem is the coordination and allocation of resources in the provisioning of a service, especially in real time environments where there are time and qualitative constraints.

Where there are numerous applications that use a service, such as a printing service, and a number of print processes that serve requests, difficulty occurs in allocating resources to get the job done given the various priorities and quality requirements of each job and the current status and capability of each resource.

A third example of the problem is component coordination in multiple component systems.

When the number of parts becomes large, the task becomes enormous, and virtually impossible when components are not available and their behavior not known beforehand at design time, or the behavior can be changed or upgraded after the initial implementation.

However, there are problems with prior art tuple spaces.

The event based, asynchronous nature of prior art tuple spaces leads to complications in synchronization, maintaining of data integrity and potential dead-locks.

Images