Cisco Technology, Inc.

Fuji Xerox Co., Ltd.

Cheng-Jia Lai

Jing Li

Laurent Denoue

Jonathan James Trevor

Gene Golovchinsky

Prasad Miriyala

Din-I Tsai

Joji Thomas Mekkattuparamban

Ming Chi Chen

Shih-cheng Stan Yang

Wenyi Wang

In one embodiment, a router receives a packet, and determines an intra-flow packet priority level of the packet. The router may then map the intra-flow packet priority level to a weighted random early detection (WRED) marking based on running statistics of intra-flow packet priority levels across received flows, and marks the packet with the mapped WRED marking. By placing the marked packet into an outgoing network queue for transmission, the router may then forward or drop the marked packet based on the network queue, accordingly.

Adaptive marking for WRED with intra-flow packet priorities in network queues

Messages are sent from a first network device to a second network device across a network. The network includes a network portion with an expected random packet loss rate. The actual packet loss rate for packets sent across the network is compared to the expected random packet loss rate. A determination is made that the actual packet loss rate is greater than the expected random packet loss rate. Compensation for network congestion is performed in response to the determination that the actual packet loss rate exceeds the expected random packet loss rate.

Network congestion control with awareness of random packet losses

In an example embodiment, there is described herein an apparatus comprising an upstream communication interface configured to receive a multicast stream from a source encoded by a first codec, a downstream communication interface, and routing and transcoding logic coupled to the upstream communication interface and downstream communication interface. The routing and transcoding logic is configured to forward the multicast stream encoded by the first codec unchanged to a first downstream subscriber of the multicast stream. The routing and transcoding logic is configured to replicate the multicast steam and transcode the multicast stream to a second codec for a second downstream subscriber, the replicated, transcoded multicast stream is forwarded to the second downstream subscriber.

Source specific transcoding multicast

Techniques are provided herein to add multicast media streams to teleconferencing sessions without the unwanted side effects that occur when a speaker's own media is echoed back to the speaker. Multicast and unicast media streams are generated by a network device, e.g., a conference bridge, and distributed to any number of endpoints, e.g., conference endpoints. A media selector is provided that selects the appropriate multicast or unicast media stream for forwarding to an associated network endpoint.

Switchable conference multicast streaming with dynamic asymmetry

A Digital Signal Processor (DSP) cloud architecture for clustering DSP resources across multiple integrated media-services gateways. The control plane components use peer-to-peer overlay connections for DSP resource management. The data plane components use a Virtual Local Area Network (VLAN) for media stream packet processing.

Digital signal processor clustering in integrated media-services gateways

Embodiments of the present invention combine the flexibility of the computer in segmenting, indexing, and retrieving data with the strengths of people in analyzing scenes. They provide a mechanism that allows a group of analyzers at distributed locations to collaborate in generating pieces of semantic information of a data stream, integrates these pieces into a summary representation that can be used for subsequent computation, and allows related hypotheses and information requests to be presented back to the analyzers. This description is not intended to be a complete description of, or limit the scope of, the invention. Other features, aspects, and objects of the invention can be obtained from a review of the specification, the figures, and the claims.

System and method for collaborative analysis of data streams

A high-performance queueing method to implement a shared queue for collaborative clusters of servers, where each cluster of servers maintains a local queue, and those queues are networked to form a unified (or shared) queue for all those servers in clusters. Requests in an overcrowded queue are forwarded to some other queues, with a randomized algorithm and a message-exchange algorithm between the networked queues. Requests may be forwarded between clusters multiple times. A request is to be accepted with a higher probability by a cluster if the cluster is visited earlier in the order and if the cluster itself has a higher probability to accept the received request. This technique ensures that the requests are forwarded to a queue that has a shorter length (thus a shorter waiting time for first-in-first-out queues) and is closer to the original cluster with a higher probability.

Growing community of inventors

Mountain View, CA, United States of America

Cheng-Jia Lai