摘要:
Techniques for transmitting data according to at least one quality of service requirement. A message path is calculated specifying a sequence of broker computers selected from a network of interconnected broker computers. The message path is statistically estimated to fulfill the at least one quality of service requirement. Quality of service metrics are received about the network of interconnected broker computers. If the message path is determined not to fulfill the quality of service requirement, a new message path is calculated specifying a new sequence of broker computers selected from the network of interconnected broker computers. The new message path is statistically estimated to fulfill the at least one quality of service requirement.
摘要:
Techniques for transmitting data according to at least one quality of service requirement. A message path is calculated specifying a sequence of broker computers selected from a network of interconnected broker computers. The message path is statistically estimated to fulfill the at least one quality of service requirement. Quality of service metrics are received about the network of interconnected broker computers. If the message path is determined not to fulfill the quality of service requirement, a new message path is calculated specifying a new sequence of broker computers selected from the network of interconnected broker computers. The new message path is statistically estimated to fulfill the at least one quality of service requirement.
摘要:
A system may include an overlay network linking a plurality of publishers, a plurality of subscribers, and a plurality of brokers. The overlay network may include a first computer network having a first network protocol carried by a second computer network having a second network protocol. The system may also include an agent carried by each of the publishers, the subscribers, and the brokers that adjust the network's topology based upon collected runtime data of condition of each link within the network and/or broker availability to any publisher and subscriber.
摘要:
A system may include an overlay network linking a plurality of publishers, a plurality of subscribers, and a plurality of brokers. The overlay network may include a first computer network having a first network protocol carried by a second computer network having a second network protocol. The system may also include an agent carried by each of the publishers, the subscribers, and the brokers that adjust the network's topology based upon collected runtime data of condition of each link within the network and/or broker availability to any publisher and subscriber.
摘要:
A system and method are disclosed for an elastic and scalable publish/subscribe scheme. Subscription information is received at a dispatcher node. A plurality of matching nodes is selected in an overlay network to store the subscription information on a computer readable storage medium. Upon receiving an event at a dispatching node, at least one of the matching nodes with the stored subscription information is selected to process the event.
摘要:
Embodiments of the invention relate to message queuing. In one embodiment, a request from an application for retrieving a message from a queue is received. The queue is stored across multiple nodes of a distributed storage system. A preference with respect to message order and message duplication associated with the queue is identified. A message sequence index associated with the queue is sampled based on the preference that has been identified. The message is selected in response to the sampling. The message that has been selected is made unavailable to other applications for a given interval of time, while maintaining the message in the queue. The message is sent to the application.
摘要:
A system and method are disclosed for an elastic and scalable publish/subscribe scheme. Subscription information is received at a dispatcher node. A plurality of matching nodes are selected in an overlay network to store the subscription information on a computer readable storage medium. Upon receiving an event at a dispatching node, at least one of the matching nodes with the stored subscription information is selected to process the event.
摘要:
A system and method are disclosed for an elastic and scalable publish/subscribe scheme. Subscription information is received at a dispatcher node. A plurality of matching nodes is selected in an overlay network to store the subscription information on a computer readable storage medium. Upon receiving an event at a dispatching node, at least one of the matching nodes with the stored subscription information is selected to process the event.
摘要:
A system and method are disclosed for an elastic and scalable publish/subscribe scheme. Subscription information is received at a dispatcher node. A plurality of matching nodes are selected in an overlay network to store the subscription information on a computer readable storage medium. Upon receiving an event at a dispatching node, at least one of the matching nodes with the stored subscription information is selected to process the event.
摘要:
Resources disposed within a plurality of distributed and autonomous computing systems, each of which may have its own legacy resource discovery service, are identified and allocated. Resource identification servers disposed within each one of the distributed computing systems communicate resource attribute data to a tree hierarchy of dedicated servers. The resource attribute data are maintained in native formats within the distributed computing systems and are mapped to a common format provided by the dedicated servers. The resource attribute data are aggregated at each node within the tree hierarchy, communicated up through the tree hierarchy to one or more root nodes and replicated down through all of the nodes. Additional system robustness is provided through period resource checks and resource attribute data updates. Resource allocation queries are submitted to any level node within the hierarchy and forwarded to the proper computing system for processing.