| 摘要 |
Further preferred embodiments of the present invention include methods directed to (a) Active Data Structures, (b) Mobile Devices, (c) Ad-Hoc Device Collections, and (d) Concurrent Massively Parallel Supercomputers. Therein, a distributed, object-oriented database engine utilizing independent, intelligent processing nodes as a cooperative, massively parallel system with redundancy and fault tolerance. Instead of using traditional methods of parallelism as found in most distributed databases, the invention utilizes a messaging system and a series of message processing nodes to determine where attributes and data files associated with objects are stored. The architecture is loosely coupled, each node independently determining if it manages or routes storage and retrieval requests. |
| 主权项 |
1. A method of assembling a catalog of objects and object references from objects stored within a network, the network including a plurality of interconnected computers with each computer storing a plurality of objects and a plurality of object catalogs, the method comprising: executing a plurality of agent programs on each computer, the plurality of agent programs containing a plurality of virtual machine environments, a plurality of collections of machine interpreted functions of which a plurality are suitable for communication with a respective plurality of the computers, a plurality of collections of machine interpreted functions of which a plurality are suitable for communication with a respective plurality of the agents on a plurality computers, and a plurality of machine interpreted functions for managing objects and sending and receiving messages; compiling a plurality of messages, each message including source and destination meta data, meta data describing properties of the message, object data embedded within the message, and a plurality of object data payloads; receiving at least one first message from at least one a computer, a program executed on a computer receiving the at least one first message, the at least one first message including source and destination meta data, meta data describing properties of the message, object data embedded within the message, and a plurality of object data payloads; transmitting at least one second message to at least one of a computer and a program executed on a computer transmitting the at least one second message, the at least one second message including source and destination meta data, meta data describing properties of the message, object data embedded within the message, and a plurality of object data payloads; determining the disposition of at least one of the plurality of messages; and creating a plurality of catalogs containing a plurality of references to a plurality of object data and a plurality of object properties, each catalog comprising a plurality of indices and a plurality of object storage areas;
and further wherein the agent computer program uses a plurality of algorithms to store event objects organized in a structure within a storage device such that the structure organization is determined by a plurality of algorithms which assess the relationship of the agent computer program with a plurality of other agent computer programs and the content of the event objects. |
