68 results about "Local congestion" patented technology

Methods and apparatus operating in a stream processing network perform load shedding and dynamic resource allocation so as to meet a pre-determined utility criterion. Load shedding is envisioned as an admission control problem encompassing source nodes admitting workflows into the stream processing network. A primal-dual approach is used to decompose the admission control and resource allocation problems. The admission control operates as a push-and-pull process with sources pushing workflows into the stream processing network and sinks pulling processed workflows from the network. A virtual queue is maintained at each node to account for both queue backlogs and credits from sinks. Nodes of the stream processing network maintain shadow prices for each of the workflows and share congestion information with neighbor nodes. At each node, resources are devoted to the workflow with the maximum product of downstream pressure and processing rate, where the downstream pressure is defined as the backlog difference between neighbor nodes. The primal-dual controller iteratively adjusts the admission rates and resource allocation using local congestion feedback. The iterative controlling procedure further uses an interior-point method to improve the speed of convergence towards optimal admission and allocation decisions.

The Domain-based Congestion Management method and apparatus detects and regulates congestion in a Diff-serv network. It uses an improvRED method for congestion detection at the core routers and token bucket filters for traffic regulation at the ingress nodes. In addition, improvRED also provides feedback control. ImprovRED uses three thresholds for detecting congestion: a minth, a maxth and a FeedbackThreshold, which takes a value between the minth and the maxth thresholds. Whenever the average queue size is greater than minth and less than Feedback-Threshold, all outgoing packets are marked appropriately to indicate a potential onset of a congestion period. When the average queue size is greater than FeedbackThreshold (but less than maxth) packets are dropped probabilistically and all the outgoing packets are marked appropriately to denote the dropping phase. When the average queue size is greater than the maximum threshold, all incoming packets are dropped. Feedback, in the form of a Local Congestion Notification (LCN) message, is used to notify the ingress nodes of a likely onset of congestion. Ingress nodes immediately respond to the congestion notification by appropriately regulating their respective traffic rates (i.e., the amount of packets they inject into the Diff-serv network). The amount of traffic (or data packets) injected into the core of the Diff-serv domain is controlled by a token bucket filter at each of the ingress nodes.

The present invention relates to the field of congestion control, and discloses a method for handling local link congestion and an apparatus. The method includes: negotiating, by a local link control entity, a local congestion notification capability with a local terminal; setting, by the local link control entity, a congestion identifier in a downlink data packet corresponding to the local congestion notification capability when a local link is congested between the local link control entity and the local terminal; and performing, by the local link control entity, congestion control on the local link, or performing, by the local terminal, congestion control on the local link according to the congestion identifier. The apparatus includes a local link control entity and a terminal device.