Abstract: A method is intended for controlling dissemination of content data in a peer-to-peer mode between peers connected to different communication networks. This method comprises the steps of: i) building, for each peer, a first peer set comprising active peers that are primary connected neighbours close to this peer in terms of network cost and a second peer set comprising active peers that are secondary connected neighbours of this peer, chosen at random irrespectively of network cost, and ii) dynamically determining into each peer a capacity of content data it is ready to receive from peers of the second peer set, taking into account its current capacity to receive content data from peers of the first peer set, in order to optimize the peer-to-peer dissemination of the content data.

Abstract: A method is intended for controlling dissemination of content in a peer-to-peer mode between peers having wireless communication capacities and comprising a cache memory for storing contents. This method consists, each time a peer, having a group of variable values each associated to a content it can store into its cache memory and representative of utility that storing of this content represents for it and for other peers, accesses to a wireless network or to another peer offering access to these contents, in downloading N contents having the N highest variable values into its group, N being a number depending on the storage capacity the peer is ready to use into its cache memory to store contents to be downloaded.

Abstract: Methods and systems for detecting and predicting congestion patterns from network feedback are described. In the inventive embodiments, network feedback in the form of binary feedback messages is received. The network feedback can comprise any suitable feedback messages and, in the illustrated and described embodiments comprise packet loss events or ECN marks, to name just a few. The inventive embodiment uses the network feedback to predict when congestion is likely to occur in the future. Based on these predictions, the behavior of end users within the system can be modified so that congestion is reduced. Specifically, the data packets that are typically transmitted by the end users can be delayed until such a time when the predicted congestion is mitigated. Thus, the inventive systems and methods take a predictive, proactive approach to addressing network congestion issues, rather than a reactive approach.

Abstract: A control device belongs to a network equipment connected to a communication network to which are coupled communication equipments of users capable of storing contents. This control device comprises i) a first analyzing means arranged for determining respective popularities of contents of a collection from at least user information, and for determining a number of replicas for each content of this collection from at least its determined content popularity, and/or ii) a second analyzing means arranged for determining content preference(s) of users from content ratings, and iii) a computation means arranged for determining location(s) for storing the replica(s) of each content, from the determined content replica numbers and/or the determined user's content preferences, in order to optimize the access to these contents by the users.

Abstract: Various actively probing techniques on a network allow an estimation of the sustainable capacity available in the network. One implementation probes the network using two probe sequences having different loads. Probe estimates associated with each sequence are used to estimate the sustainable capacity in the network. Another implementation probes the network using two probe sequences, in which either the probe packet size or the inter-probe gap are different between the sequences. Calculation of a delay trend for each sequence leads to estimate of the maximum network capacity and the background load on the network. Yet another implementation uses a windowing technique to probe the sustainable capacity in the network. The window size is increased until the network is saturated. The number of bytes transferred within the window during a given time period is an estimate of the sustainable capacity of the network.

Abstract: An end-to-end congestion control is provided that emulates a different service differentiation than the common low-normal priority. This protocol is referred to as 4CP (Competitive, Considerate Congestion Control). The target service differentiation enables provisioning of per-flow average bandwidth guarantees to “normal” traffic, but not at the expense of potentially starving the “low” priority traffic (4CP). It thus features incentive compatibility to file-transfer applications that are throughput-greedy but want to be considerate to other traffic. 4CP is implemented and configured as a sender-only adaptation of standard TCP, and requires no special network feedback. Configuration of the bandwidth guarantee is either statically configured or automatically adjusted by 4CP. The automatic mode aims to be TCP-friendly over appropriately large timescale.

Abstract: Methods and systems for detecting and predicting congestion patterns from network feedback are described. In the inventive embodiments, network feedback in the form of binary feedback messages is received. The network feedback can comprise any suitable feedback messages and, in the illustrated and described embodiments comprise packet loss events or ECN marks, to name just a few. The inventive embodiment uses the network feedback to predict when congestion is likely to occur in the future. Based on these predictions, the behavior of end users within the system can be modified so that congestion is reduced. Specifically, the data packets that are typically transmitted by the end users can be delayed until such a time when the predicted congestion is mitigated. Thus, the inventive systems and methods take a predictive, proactive approach to addressing network congestion issues, rather than a reactive approach.

Abstract: In an application-level background transport service, a receiver node infers the available network capacity between itself and a sender node over a control interval. Based on the inferred available network capacity, the receiver node adjusts its receive window size accordingly in order to conservatively optimize the bandwidth used by a background transfer without degrading performance of other foreground transfers on the network. The adjusted receive window size is communicated to the sender node, which is likely to adjust its send window size based on the adjusted receive window size.

Abstract: In an unstructured overlay network, each node in the network periodically tests logical network links among neighboring nodes to determine whether the links should be reorganized. A Metropolis scheme is used to determine the probability with which the links are reorganized. This probability is computed based on the change in link costs and/or node degrees that would be effected by a proposed reorganization. The Metropolis scheme tends to maintain a consistency among the degrees of the nodes, thereby providing strong failure resilience.

Abstract: The invention concerns a method for assigning parts of multimedia files to distributed storage devices, the number of said devices being an integer n, characterized in that it comprises the following steps: a step consisting in dividing each multimedia file in windows, which represent data for contiguous playback of a given duration; a step consisting, for each multimedia file, in determining a number k so that any original window of file can be recovered from any k distinct symbols; a step consisting, for each window of each said multimedia files, in assigning n corresponding symbols to said devices, one to each device, the n distinct symbols being created using erasure coding; and a step consisting in partially or totally filling the storage memory left on each device after said first steps with prefix windows of said multimedia files.

Abstract: An end-to-end congestion control is provided that emulates a different service differentiation than the common low-normal priority. This protocol is referred to as 4CP (Competitive, Considerate Congestion Control). The target service differentiation enables provisioning of per-flow average bandwidth guarantees to “normal” traffic, but not at the expense of potentially starving the “low” priority traffic (4CP). It thus features incentive compatibility to file-transfer applications that are throughput-greedy but want to be considerate to other traffic. 4CP is implemented and configured as a sender-only adaptation of standard TCP, and requires no special network feedback. Configuration of the bandwidth guarantee is either statically configured or automatically adjusted by 4CP. The automatic mode aims to be TCP-friendly over appropriately large timescale.

Abstract: Real time communications over a network are adjusted to improve the quality of service (QoS) under incipient congestion conditions. The system detects incipient network congestion and feeds back information regarding the incipient congestion back to the transmitter. Based on this information, the transmission rate is altered appropriately using a control algorithm, which computes the altered transmission rate based on a weight parameter, a gain parameter, and information from a congestion report. The altered transmission rate improves the transmitter's use of the available bandwidth to maintain an acceptable QoS at the receiver.

Abstract: In an application-level background transport service, a receiver node infers the available network capacity between itself and a sender node over a control interval. Based on the inferred available network capacity, the receiver node adjusts its receive window size accordingly in order to conservatively optimize the bandwidth used by a background transfer without degrading performance of other foreground transfers on the network. The adjusted receive window size is communicated to the sender node, which is likely to adjust its send window size based on the adjusted receive window size.

Abstract: In an unstructured overlay network, each node in the network periodically tests logical network links among neighboring nodes to determine whether the links should be reorganized. A Metropolis scheme is used to determine the probability with which the links are reorganized. This probability is computed based on the change in link costs and/or node degrees that would be effected by a proposed reorganization. The Metropolis scheme tends to maintain a consistency among the degrees of the nodes, thereby providing strong failure resilience.

Abstract: Methods and systems for detecting and predicting congestion patterns from network feedback are described. In the inventive embodiments, network feedback in the form of binary feedback messages is received. The network feedback can comprise any suitable feedback messages and, in the illustrated and described embodiments comprise packet loss events or ECN marks, to name just a few. The inventive embodiment uses the network feedback to predict when congestion is likely to occur in the future. Based on these predictions, the behavior of end users within the system can be modified so that congestion is reduced. Specifically, the data packets that are typically transmitted by the end users can be delayed until such a time when the predicted congestion is mitigated. Thus, the inventive systems and methods take a predictive, proactive approach to addressing network congestion issues, rather than a reactive approach.

Abstract: Various actively probing techniques on a network allow an estimation of the sustainable capacity available in the network. One implementation probes the network using two probe sequences having different loads. Probe estimates associated with each sequence are used to estimate the sustainable capacity in the network. Another implementation probes the network using two probe sequences, in which either the probe packet size or the inter-probe gap are different between the sequences. Calculation of a delay trend for each sequence leads to estimate of the maximum network capacity and the background load on the network. Yet another implementation uses a windowing technique to probe the sustainable capacity in the network. The window size is increased until the network is saturated. The number of bytes transferred within the window during a given time period is an estimate of the sustainable capacity of the network.

Abstract: Real time communications over a network are adjusted to improve the quality of service (QoS) under incipient congestion conditions. The system detects incipient network congestion and feeds back information regarding the incipient congestion back to the transmitter. Based on this information, the transmission rate is altered appropriately using a control algorithm, which computes the altered transmission rate based on a weight parameter, a gain parameter, and information from a congestion report. The altered transmission rate improves the transmitter's use of the available bandwidth to maintain an acceptable QoS at the receiver.

Abstract: A system and method for the dissemination of information to a plurality of nodes, the nodes connected in a network environment. Each node maintains a partial view of the network that identifies some of the other network nodes. The act of sending the message to a plurality of nodes further comprises delivery of the message to all nodes identified in the partial view. The partial view is created through a decentralized subscription process.