摘要:
A method for estimating a round trip time, RTT, of a packet flow, comprising the steps of: - computing a periodogram P N based on N samples h 1 , h 2 ..h N of an inter-arrival time signal X between adjacent packets, X(t k ) = h k = t k - t k-1 , using a Lomb-Scargle method, - extracting a fundamental pulsation ω 0 out of said periodogram P N using a pattern matching technique, - determining the RTT out of said fundamental pulsation.
摘要:
The invention concerns a method and a system for splitting a multiplex packet stream (1) received in a packet processing node (100) into N packet streams on N corresponding processing channels (c1, c2, ..., cN). According to the invention, said method comprises splitting said multiplex packet stream (1) by distributing each packet of the multiplex packet stream to one intermediary channel among a set of M intermediary channels (i1, i2, ..., iM) according to a hash function and measuring load on each of said N corresponding processing channels (c1, c2, ..., cN). Additionnaly, for each intermediary channel (i1, i2, ..., iM), the method further comprises performing a mapping of said intermediary channel to one of said N processing channels for transmitting a packet carried on said intermediary channel to an associated processing channel, wherein, for a current packet carried on said intermediary channel, said mapping can be dynamically updated as a function of the N processing channel's measured loads and the time difference between a reception time of said current packet and the reception time of the last preceding packet carried on said intermediary channel.
摘要:
Scheduling method of packets data flow between input and output line cards of a switching node, wherein a synchronous scheduling is implemented between all pairs of input and output line cards, such synchronous scheduling being combined with asynchronous scheduling which is activated upon request by an input line card as a function of taffic load conditions in a queue of the corresponding input line card.
摘要:
The present invention refers to a method for reducing energy consumption in a packet processing linecard of a packet transmission network, said packet processing linecard comprising a plurality of microprocessors aimed at processing packet traffic wherein the number of active microprocessors is dynamically adjusted as a function of the computation of a traffic estimator based on a recurrent estimation of at least two statistical parameters including the average and a parameter representative of the statistical distribution of the packet traffic.
摘要:
A method for packet flow-rate control within a network node (4) comprised within a packet-switched communication network connection linking a sender (1) to a receiver (2), the said method comprising the following steps - computing the average arrival rate of packets since the last packet mark or drop event; - sending a congestion signal to the sender as soon as the computed packet arrival rate is above a predefined target rate; - repeating congestion signal sending until the sender receives at least a congestion signal and consequently reduces the sending rate.
摘要:
The invention concerns a method for scheduling packets belonging to a plurality of flows received at a router. It is also provided the system for carrying out the method. According to the invention, a single packet queue is used for storing said packets, said single packet queue being adapted to be divided into a variable number of successive sections which are created and updated dynamically as a function of each received packet, each section being of variable size and a section load threshold for each flow of said plurality of flows being allocated to each section. The method further comprises insertion (S11; S22; S210 ; S222 ; S230) of each received packet of a given flow in one of said successive sections as a function of said given flow and of the corresponding section load threshold.
摘要:
Pour aiguiller sélectivement via une matrice de commutation (1) des paquets en attente dans des modules d'entrée (IMi) vers des modules de sortie (OMj) destinataires, des fonctions d'arbitrage reparties sont exécutables par cycles d'arbitrage successifs. Chaque cycle comporte : - une première phase exécutable par chaque contrôleur d'entrée (ICi) pour envoyer à chaque contrôleur de sortie (OCj) des requêtes représentatives de quantités de paquets en attente "requises", - une seconde phase exécutable par chaque contrôleur de sortie (OCj) pour déterminer en fonction des requêtes des quantités de paquets "admissibles", - une troisième phase exécutable par une unité d'arbitrage central (CSC) pour déterminer en fonction de toutes les quantités admissibles des quantités "agrégées admises", - une quatrième phase exécutable par chaque contrôleur d'entrée (ICi) pour déterminer en fonction des quantités admissibles et des quantités agrégées admises des quantités de paquets admises. Application aux réseaux de télécommunication, notamment multiservice.