摘要:
A high-speed security device for network connected industrial controls provides hybrid processing in tandem hardware and software security components. The software security component establishes state-less data identifying each packet that requires high-speed processing and loads a data table in the hardware component. The hardware component may then allow packets matching data of the data table to bypass the software component while passing other non-matching packets to the software component for more sophisticated state analysis.
摘要:
A distributed control system employs a number of autonomous cooperative units that intercommunicate with bids and counter bids to allocate the production of a product among them. Network traffic caused by these bid messages is significantly reduced by creating as a central clearinghouse, a global resource locator that can identify autonomous cooperative units more likely to be receptive to bids from other autonomous cooperative units. Portions of the data of the global resource locator are held at the particular computers implementing the autonomous cooperative units in a cache-like structure that may be referred to without network communication. Misses within this cache structure cause a refreshing of the cache from the global resource locator and an updating of a learning relation table that further improves predictions in the future. Historical successes in bidding are stored at the cache structure to further aid in selecting likely bid recipients.
摘要:
An operating system for distributed industrial controllers ensures the completion of enrolled application programs within a predetermined time span as is required for robust industrial control by preallocating dynamic and static hardware resources on a per application basis and in a manner that ensures execution of the application within the necessary time constraints. Portions of the distributed operating system may be distributed at particular hardware resources to provide necessary modeling for those hardware resources in making the commitments to resource bandwidths.
摘要:
Fault-tolerant Ethernet is provided through the use of special interfaces providing end-devices with duplicate ports that may be alternatively enabled with the same network address in the event of a fault detected by loss of the message from a beacon device. Unidirectional network errors in the direction from the end-device to the beacon devices are detected through a path check message from the end-device to the beacon device. Multiple faults may be accommodated without network interruption by adding top-level switches without increasing the number of ports on the end-devices by cross connecting a lowest level of switch.
摘要:
A network for an industrial control system employs a ring topology that is normally opened by a ring supervisor at the ring supervisor. Upon failure of the network, the ring supervisor reconnects the ring to provide an alternative transmission path around the failure point. High speed operation is reconciled with the ability to use commercial switching integrated circuits through a dual communication channel of communicating a network state as either closed or open using both high-speed hardware handled beacon frames and low-speed software processed announce frames.
摘要:
A distributed control system and methods of operating such a control system are disclosed. In one embodiment, the distributed control system is operated in a manner in which interrupts are at least temporarily inhibited from being processed to avoid excessive delays in the processing of non-interrupt tasks. In another embodiment, the distributed control system is operated in a manner in which tasks are queued based upon relative timing constraints that they have been assigned. In a further embodiment, application programs that are executed on the distributed control system are operated in accordance with high-level and/or low-level requirements allocated to resources of the distributed control system.
摘要:
A method and circuit for precisely synchronizing clocks in separate nodes on a communication network is provided by adjusting timestamps and related data in network messages. The circuit will allow a daisy-chain connection of the nodes and will forward time synchronization frames while accounting for delays in a manner that does not use boundary clocks, but does not depart from the IEEE 1588 standard protocol. The delays will be added on the fly to synchronization packets and the IP checksum and frame CRC will be adjusted. Deterministic data delivery and redundant data paths are also provided in a full duplex Ethernet network.
摘要:
A distributed control system and methods of operating such a control system are disclosed. In one embodiment, the distributed control system is operated in a manner in which interrupts are at least temporarily inhibited from being processed to avoid excessive delays in the processing of non-interrupt tasks. In another embodiment, the distributed control system is operated in a manner in which tasks are queued based upon relative timing constraints that they have been assigned. In a further embodiment, application programs that are executed on the distributed control system are operated in accordance with high-level and/or low-level requirements allocated to resources of the distributed control system.
摘要:
An interrupt handler is provided for a real-time control system that prevents interrupts which occur asynchronously with respect to control tasks from upsetting guarantees of timely execution of the control tasks. For interrupts associated with the communication of messages between portions of a control task over the distributed system, the interrupts are converted to proxy tasks that may be scheduled like any task in a multitasked-operated system. More generally, interrupts may be assigned to a predetermined interrupt window being a portion of the total processing bandwidth of the processor. In pre-allocating the processor bandwidth to the control tasks, this interrupt window may be subtracted out thereby guaranteeing adequate bandwidth for both interrupt processing and user tasks.
摘要:
Multiple gateway devices communicating between a device level ring (DLR) network and a spanning tree (ST) network may be provided a gateway protocol that cooperatively ensures that only a single gateway is active at a given time. This cooperation may be effected by the transmission of advertise messages by gateways, the advertise messages holding precedence values so that only a single gateway having a highest precedence value is active at a given time. Loss of the advertise messages may trigger a gateway held in a backup state to assume an active gateway role.