• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    An interconnection device between first and second networks, between which critical data flows having a need for bandwidth reservation to ensure flow and latency constraints, performs, for each critical data stream, the device interconnection: measuring (409) an effective rate of said critical data stream; verifying (410) whether said measured actual rate is greater than the sum of a previously reserved bandwidth and a tolerance bandwidth; verifying (410) whether said measured actual rate is less than the previously reserved bandwidth, to which is subtracted the sum of a margin bandwidth and a tolerance bandwidth; and in the case of positive verification, adjusting (411) the bandwidth reservation for said critical data stream to a value equal to the sum of the actual measured rate of flow for said critical data stream and the margin bandwidth.
    公开号:FR3039729A1
    申请号:FR1557419
    申请日:2015-07-31
    公开日:2017-02-03
    发明作者:Hatim Fadle；Anthony Reuche
    申请人:Sagemcom Broadband SAS；
    IPC主号:
    专利说明:

    The present invention relates to a dynamic reservation of bandwidth reservation for data flows transiting between first and second communication networks via an interconnection device.
    Among the interconnection devices of first and second communication networks, there are in particular residential gateway type equipment. Such equipment makes it possible to interconnect a first LAN ("Local Area Network") or WLAN ("Wireless LAN") network and a second Wide Area Network (WAN) type communication network. ), in particular to allow communication devices connected to the LAN or WLAN to access the Internet, to receive or make telephone calls, to receive television (TV) over IP ("Internet Protocol" in English). , as defined in the normative document RFC 791). In the context of service implementations, data flows transit between the first and second communication networks, via said interconnection device. These data streams are typically categorized according to the service to which said data streams are attached. For each type (or category) of data flow corresponds flow and / or latency constraints that must be respected according to the Quality of Service (QoS) required by the service to which the service relevant data stream is attached. The evolution of technologies favoring the increase of the bit rate is completely dissociated between the WAN and LAN communication networks. Indeed, currently increasing the capacity of LAN communication networks is much more important than increasing the capacity of WAN communication networks. In addition, the fact that domestic gateways have LAN connectivity options that increases rapidly (eg Gigabit Ethernet, Wi-Fi, Bluetooth, ...) while the WAN side connectivity is moving more slowly. Thus, the need to regulate and manage the different data flows that pass through the interconnection device is essential. All the challenge of the current systems is to be able to offer a maximum of bandwidth on the LAN side while respecting the service constraints in terms of speed and latency WAN side. For this, it is advisable to propose a payload on the WAN side as close as possible to the physical bit rate, that is to say avoiding unnecessary bandwidth reservations which impede the good management of the admission control.
    Such a situation is glaring with respect to residential gateways, but is also encountered in other communication network interconnection devices.
    It is desirable to overcome these disadvantages of the state of the art. It is thus desirable to provide a solution that makes it possible to improve the management of the bandwidth reservation for data flows transiting between a first communication network and a second communication network interconnected via an interconnection device. It is also desirable to provide a solution that is autonomous. It is also desirable to provide a solution which consumes little of treatment resources (and consequently consumes little energy resources). A bandwidth management method implemented by an interconnection device between a first communication network and a second communication network, critical data flows and non-critical data flows transiting between the first and second networks. communication via said interconnection device, the critical data streams having unlike non-critical data streams a need for bandwidth reservation to ensure flow and latency constraints. Said method is such that the interconnection device implements a critical data flow monitoring mechanism transiting between the first and second communication networks via said interconnection device so that for each critical data stream transiting between the first and second communication networks and second communication networks via said interconnection device, the interconnection device performs the following steps: measuring an effective rate of said critical data stream; performing a first check consisting in checking whether said measured actual rate is greater than the sum of a bandwidth previously reserved for said critical data stream and a tolerance bandwidth; performing a second check consisting in checking whether said measured actual rate is less than the bandwidth previously reserved for said critical data stream, to which is subtracted the sum of a margin bandwidth and a tolerance bandwidth; and in the case of a first positive check or a second positive check, adjust the bandwidth reservation for said critical data flow to a value equal to the sum of the measured actual flow rate for said critical data flow and the margin bandwidth. Thus, the management of the bandwidth reservation for the data flows transiting between the first communication network and the second communication network interconnected via an interconnection device is improved.
    According to a particular embodiment, the reservation of bandwidth for said critical data stream is adjusted within the limit of a maximum terminal associated with said critical data stream. Thus, it is ensured that the reservation of bandwidth for said data stream remains within an acceptable limit vis-à-vis other data streams that transit via said interconnection device.
    According to a particular embodiment, said interconnection device activates said monitoring mechanism, for all of said critical data flows passing between the first and second communication networks via said interconnection device, periodically. Thus, less processing resources are used by the monitoring mechanism.
    According to a particular embodiment, said interconnection device dynamically defines, for each flow of critical data transiting between the first and second communication networks via said interconnection device, a period of time between two successive activations of the monitoring mechanism. performing the following steps: checking whether a criterion of stability of the effective flow rate of said critical data flow is respected; in case of positive verification with respect to said stability criterion, to lengthen the period of time between two successive activations of the monitoring mechanism for said critical data stream; and in case of negative verification with respect to said stability criterion, reducing the period of time between two successive activations of the monitoring mechanism for said critical data stream. Thus, a compromise between reduction of processing resource consumption and responsiveness of the monitoring mechanism is found.
    According to a particular embodiment, in the event of a negative verification with respect to said stability criterion, said period of time between two successive activations of the monitoring mechanism for said critical data stream is reinitialized to a predefined minimum terminal value. Thus, the responsiveness of the monitoring mechanism is enhanced.
    According to a particular embodiment, said stability criterion is respected when each actual flow measurement of said critical data stream in the last N runs, N> 1, is less than the reserved bandwidth, insofar as said reserved bandwidth has not been adjusted during said N previous executions of the monitoring mechanism for said critical data stream. Thus, as long as the reservation of bandwidth is adequate, the reduction of consumption of processing resources is increased.
    According to a particular embodiment, said stability criterion is respected when said reserved bandwidth has not been adjusted during said N last executions, N> 1, of the monitoring mechanism for said critical data stream. Thus, the verification of the stability criterion is simple. The invention also relates to an interconnection device between a first communication network and a second communication network, critical data flows and non-critical data flows transiting between the first and second communication networks via said interconnection device. , the critical data streams having unlike non-critical data flows a need for bandwidth reservation to ensure flow and latency constraints. The interconnection device implements a mechanism for monitoring the critical data flows transiting between the first and second communication networks via said interconnection device so that for each critical data stream passing between the first and second communication networks via said interconnection device, the interconnection device implements: means for measuring an effective bit rate of said critical data stream; means for performing a first check consisting in checking whether said measured actual rate is greater than the sum of a bandwidth previously reserved for said critical data stream and a tolerance bandwidth; means for performing a second check consisting of checking whether said measured actual rate is less than the bandwidth previously reserved for said critical data stream, to which is subtracted the sum of a margin bandwidth and a tolerance bandwidth ; and means for, in the case of a first positive check or a second positive check, adjusting the bandwidth reservation for said critical data stream to a value equal to the sum of the actual measured rate of flow for said critical data stream and the band margin passerby. The invention also relates to a computer program, which can be stored on a medium and / or downloaded from a communication network, in order to be read by a processor. This computer program includes instructions for implementing the above-mentioned method according to any one of its variants, when said program is executed by the processor. The invention also relates to storage means comprising such a computer program.
    The characteristics of the invention mentioned above, as well as others, will emerge more clearly on reading the following description of an exemplary embodiment, said description being given in relation to the attached drawings, among which: Fig. 1 schematically illustrates a communication system in which the present invention can be implemented; FIG. 2 schematically illustrates an example of a hardware architecture of an interconnection device of the first and second communication networks of the communication system of FIG. 1; FIG. 3 schematically illustrates an algorithm, implemented by the interconnection device, for dynamically managing bandwidth reservations for data streams transiting between the first and second communication networks via said interconnection device; FIG. 4 schematically illustrates an algorithm, implemented by the interconnection device, for dynamically managing said bandwidth reservations, in a particular embodiment of the present invention; FIG. 5 schematically illustrates a reservation of bandwidth following the execution of the algorithm of FIG. 3 or the algorithm of FIG. 4; and - FIG. 6 schematically illustrates an algorithm, implemented by the interconnection device, to define the duration of a period between two successive activations of an effective flow monitoring mechanism vis-à-vis each said data flow.
    Fig. 1 schematically illustrates a communication system in which the present invention can be implemented. The communication system of FIG. 1 comprises a first communication network 101 and a second communication network 102 interconnected by an interconnection device 110. Each of the first 101 and second 102 communication networks comprises at least one communication device (not shown). Exchanges in the form of a data stream can thus be established between communication devices of the first communication network 101. Exchanges in the form of data streams can thus be established between communication devices of the second communication network 102. Finally, exchanges in the form of a data stream can thus be established between communication devices of the first communication network 101 and communication devices of the second communication network 102, and in this case via the interconnection device 110. The invention particularly attaches to the bandwidth reservation management of these data flows transiting via the interconnection device 110.
    "Band reservation" is understood to mean a bandwidth reservation within the first communication network 101 and / or a reservation of bandwidth within the second communication network 102 and / or a reservation of bandwidth within the device. 110 interconnection (buffer sharing ("buffer" in English)). It should be understood that this reservation of bandwidth is an indication for a good management of the admission control, namely to accept or refuse new data flows in view of a current state of occupation of the resources, knowing that the flows of critical data can exceed the reservations made, which could disrupt the functioning of the communication system, since, if new data streams are accepted in view of the current bandwidth reservation, while existing data streams consume more bandwidth than that reserved, this can lead to network overload, packet loss and thus cause degradation of service.
    In a particular embodiment, the first communication network 101 is of the LAN or WLAN type, and the second communication network 102 is of the WAN type, and the interconnection device 110 is a residential gateway. The interconnection device 110 may interconnect other types of communication networks, such as for example a LAN and a WLAN.
    The data flows transiting within the first communication network 101, such as those transiting within the second communication network 102, and as those transiting between the first communication network 101 and the second communication network 102 via the interconnection device 110, may be of different types, depending on the applications to which said data streams are derived and / or intended. Indeed, certain data flows must respect transmission and speed latency constraints, and are therefore critical in terms of bandwidth management reserved for them. Failure to meet these constraints typically results in packet loss. Other data streams do not have such latency and throughput constraints, and therefore may not require bandwidth reservation. These other data streams are then transmitted as quickly as possible according to the bandwidth actually left free by the streams for which bandwidth has been reserved, avoiding packet losses. A Deep Packet Inspection (DPI) of the data streams makes it possible to determine, in particular according to the transport protocols used, what are the respective types of said data streams.
    The data flows transiting between the first communication network 101 and the second communication network 102 via the interconnection device 110 are preferentially in the IP format.
    When the interconnection device 110 is a residential gateway, three services coexist: a VoIP service ("VoIP" for "Voice over IP" in English), which is considered as critical with respect to latency and throughput constraints to respect for the proper functioning of the service; a TV-over-IP television service, which is also considered as critical with respect to latency and speed constraints to be respected for the proper functioning of the service; and a data service ("Data"), which is considered uncritical because it has no latency and throughput constraints.
    As detailed below, the interconnection device 110 is adapted to monitor the different data flows transiting between a first interface 111 of said interconnection device 110, via which said interconnection device 110 is connected to the first communication network 101. , and a second interface 112 interconnection device 110, via which said interconnection device 110 is connected to the second communication network 102. This monitoring consists in determining, for the critical data flows, the respective actual flows of said data streams. and to verify the adequacy of the respective reserved bandwidths for said data streams, and possibly adjust the bandwidth reservations, in real time. To do this, the interconnection device 110 has a DPI packet depth inspection unit 120 placed on the data path between the first 111 and second 112 interfaces. The DPI packet depth inspection unit 120 is further connected to a control unit 130 of the interconnection device 110, in particular responsible for dynamically managing the bandwidth reservations for the critical data streams. The behavior of the interconnection device 110, and more particularly of the control unit 130, vis-à-vis the bandwidth reservations is described below in relation to FIGS. 3, 4 and 6.
    Fig. 2 schematically illustrates an example of hardware architecture of the interconnection device 110.
    According to this example of hardware architecture, the interconnection device 110 comprises, connected by a communication bus 220: a processor or CPU ("Central Processing Unit" in English) 210; a Random Access Memory (RAM) 211; ROM (Read Only Memory) 212; a storage unit 213 or a storage medium reader, such as a hard disk drive HDD ("Hard Disk Drive" in English) or a card reader SD ("Secure Digital" in English); and a set of interfaces 214 enabling said communication device to be connected to the first 101 and second 102 communication networks (ie respectively the first 111 and second 112 interfaces of the schematic representation of Fig. 1).
    The processor 210 is capable of executing instructions loaded into the RAM 211 from the ROM 212, an external memory (not shown), a storage medium, or a communication network. When the interconnection device 110 is turned on, the processor 210 is able to read instructions from RAM 211 and execute them. These instructions form a computer program causing the processor 210 to implement all or some of the algorithms and steps described hereinafter.
    Thus, all or some of the algorithms and steps described below (and hence the units, of the interconnection device 110, mentioned with reference to Fig. 1) can be implemented in software form by executing a set of instructions. by a programmable machine, such as a DSP ("Digital Signal Processor" in English) or a microcontroller. All or some of the algorithms and steps described hereinafter can also be implemented in hardware form by a machine or a dedicated component, such as an FPGA ("Field-Programmable Gâte Aray" in English) or an ASIC ("Application-Specific"). Integrated Circuit ").
    Fig. 3 schematically illustrates an algorithm, implemented by the interconnection device 110, for dynamically managing bandwidth reservations for data streams transiting between the first 101 and second 102 communication networks via said interconnection device 110.
    In a step 301, the interconnection device 110 detects an activation event of a data flow monitoring mechanism transiting between the first 101 and second 102 communication networks via said interconnection device 110. According to a first embodiment implementation, the monitoring mechanism of said data streams is activated periodically. According to a second particular embodiment, the monitoring mechanism of said data streams is activated by relying on a period of time between two successive activations, the duration of which is dynamically adjusted, as detailed below in relation to FIG. 6.
    In a next step 302, the interconnection device 110 performs effective bit rate measurements of each of the critical data streams that pass between the first 101 and second 102 communication networks via said interconnection device 110.
    The interconnection device 110 can be informed of the critical data flows, which transit between the first 101 and second 102 communication networks via said interconnection device 110, to be taken into account, by prior configuration, eg thanks to information entered by a user via a control user interface of said interconnection device 110. In a variant, the interconnection device 110 can be informed of the critical data flows, which transit between the first 101 and second 102 communication networks via said device. 110, to be taken into account, by analyzing messages exchanges that transit between the first 101 and second 102 communication networks via said interconnection device 110 to establish said critical data flows. Indeed, such critical data flows are typically based on a prior protocol of linking and configuration of the terminal devices concerned. Analyzing these messages by deep packet inspection DPI can identify subsequent critical data streams. A similar procedure is also typically implemented to terminate the data streams. In another variant, the interconnection device 110 may be informed of critical data flows, which transit between the first 101 and second 102 communication networks via said interconnection device 110, to be taken into account, by directly analyzing the data flows. data passing between the first 101 and second 102 communication networks via said interconnection device 110. Analyzing said data flows by deep packet inspection DPI makes it possible to discriminate the critical data flows from the other data streams (non-critical) . It is this latter approach that is used, illustratively, thereafter.
    To perform the actual flow measurements of each of the critical data flows, the interconnection device 110 must be able to discriminate the critical data flows among all the data flows that pass through the interconnection device 110. is done by inspection of DPI deep packets that pass through the interconnection device 110. In this case, the analysis is less expensive in terms of processing resources than to discover critical data streams not previously identified, because it can Here it is sufficient to retrieve source and destination IP addresses, as well as source and destination ports, to identify the data flow in question.
    In a next step 303, the interconnection device 110 checks whether a bandwidth reservation adjustment has to be made with respect to at least one of said critical data streams. When bandwidth has been previously reserved for the critical data stream considered, the interconnection device 110 checks whether the bandwidth reservation is still adequate, or even necessary, as a function of the actual measured bit rate for said critical data stream. . When no bandwidth has been previously reserved for the critical data stream considered, the interconnection device 110 must perform the bandwidth reservation according to the measured actual rate for said critical data stream. As detailed later in connection with FIG. 4 in a particular embodiment, the interconnection device 110 checks whether the measured actual flow rate for said critical data stream (denoted BWc) is greater than the sum of the previously reserved bandwidth (denoted by BWr) for said data stream critical and bandwidth tolerance (rated BWt); if so, a bandwidth reservation (increase) adjustment must be made for said critical data stream, and a step 304 is performed. In addition, the interconnection device 110 checks whether the actual flow rate (thus noted BWc) measured for said critical data flow is less than the previously reserved bandwidth (therefore denoted by BWr) for said critical data flow to which is subtracted the sum of a margin bandwidth (denoted BWm) and the tolerance bandwidth (therefore denoted BWt); if so, a bandwidth reservation (reduction) adjustment must be made for said critical data stream, and step 304 is performed. In any other case, no bandwidth adjustment is necessary, and a step 305 is performed, in which the data flow monitoring mechanism passes between the first 101 and second 102 communication networks via said interconnection device 110. is disabled.
    The margin bandwidth BWm serves to guard against sporadic upward variations in the actual flow BWc of the critical data stream under consideration. The margin bandwidth BWm can be predefined, for example depending on the type of service to which the critical data stream is attached. The margin bandwidth BWm may also be a predefined percentage of the effective bit rate BWc of said critical data stream.
    The tolerance bandwidth BWt serves as a trigger for adjusting the bandwidth reserved for said critical data stream considered. The tolerance bandwidth BWt can be predefined, for example depending on the type of service to which the critical data stream is attached. The tolerance bandwidth BWt can also be a predefined percentage of the effective bit rate BWc of said critical data stream.
    In step 304, the interconnect 110 performs a bandwidth reservation adjustment for each of the critical data streams for which the actual rate measurement performed in step 302 has shown that the previous bandwidth reservation is no longer adequate. As detailed later in connection with FIG. 4 in a particular embodiment, the interconnection device 110 performs a bandwidth reservation equal to the sum of the actual bit rate (therefore BWc) measured for said critical data stream and the margin bandwidth (denoted BWm) , unless the critical data stream for which a bandwidth reservation had previously been made no longer exists, in which case the interconnection device 110 releases the bandwidth previously reserved for said critical data stream. Then step 305 is performed.
    Fig. 4 schematically illustrates an algorithm, implemented by the interconnection device 110, for dynamically managing the bandwidth reservations for the critical data streams passing through said interconnection device 110, in a particular embodiment of the present invention.
    In a step 401, the interconnection device 110 activates the data flow monitoring mechanism transiting between the first 101 and second 102 communication networks via said interconnection device 110. This step is typically initiated as step 301 above. described.
    In a next step 402, the interconnection device 110 checks whether at least one critical data stream, therefore to be monitored, has an effective throughput of zero. For example, such information may have been provided to the monitoring mechanism following the detection, by deep packet inspection DPI, that closing messages of said critical data stream have been exchanged between a device of the first communication network 101 and a device of the second communication network 102. Otherwise, as apparent later, the algorithm of FIG. 4 typically includes a plurality of loops, and this step 402 is used to manage the critical data streams that have disappeared during monitoring. If at least one critical data stream has a zero effective rate, a step 403 is performed; otherwise, a step 404 is performed.
    In step 403, the interconnection device 110 cancels the bandwidth reservation BWr that had previously been made for each critical data stream having an effective bit rate BWc. In a preferred embodiment, a minimum bandwidth reservation equal to the margin bandwidth BWm is retained for each critical data stream that may pass through the interconnection device 110. This ensures that the entire the bandwidth is not consumed by non-critical data streams, which could prevent proper support of critical data streams that are enabled or reactivated later. In other words, the interconnection device 110 releases, within the limit defined by the system specifications, the bandwidth BWr that had been previously reserved for each critical data stream having an effective bit rate BWc. Then, step 404 is performed.
    In step 404, the interconnection device 110 checks whether the monitoring is complete. Indeed, the monitoring mechanism is preferably not permanently activated, so as to limit the consumption of processing resources of the interconnection device 110. The monitoring is typically activated intermittently for periods of time of the predefined duration. If the monitoring is completed, a step 405 is performed in which the data flow monitoring mechanism transiting between the first 101 and second 102 communication networks via said interconnection device 110 is disabled, as for step 305; otherwise, a step 406 is performed.
    In step 406, the interconnection device 110 waits for packet reception (s) via one or the other of the first 101 and second 102 communication networks. A watchdog mechanism can be implemented to prevent the algorithm of FIG. 4 blocks in this step, which would mean that no more critical data flow passes through the interconnection device 110.
    In a next step 407, the interconnection device 110 performs a DPI deep packet inspection so as to identify to which data stream belongs the received data packet (s), and more particularly if the (s) received data packet (s) belongs to a critical data stream.
    In a next step 408, the interconnection device 110 checks whether the received data packet (s) belongs to a critical data stream, and therefore to a data stream to be monitored. If this is the case, a step 409 is performed; otherwise, step 404 is repeated.
    In step 409, the interconnection device 110 measures the actual bit rate BWc of said critical data stream to which the received data packet (s) belongs.
    In a next step 410, the interconnection device 110 checks whether the previously reserved bandwidth BWr for said critical data stream is still adequate with respect to the actual flow BWc of said critical data stream. The interconnection device 110 checks whether the actual measured bit rate BWc for said critical data stream is greater than the sum of the previously reserved bandwidth BWr for said critical data stream and the tolerance bandwidth BWt; if so, a bandwidth reservation (increase) adjustment must be made for said critical data stream, and a step 411 is performed. In addition, the interconnection device 110 checks whether the actual measured bit rate BWc for said critical data stream BWc is less than the previously reserved bandwidth BWr for said critical data stream to which the sum of the margin bandwidth is subtracted. BWm and BWt tolerance bandwidth; if so, a bandwidth reservation (reduction) adjustment must be made for said data stream, and step 411 is performed. In any other case, no adjustment is necessary to said critical data stream, and step 402 is repeated.
    In step 411, the interconnect 110 performs a bandwidth reservation adjustment for each of the critical data streams for which the actual rate measurement performed at step 409 has shown that the previous bandwidth reservation is no longer adequate. The interconnection device 110 then performs a reservation of bandwidth equal to the sum of the actual flow rate BWc measured for said critical data flow and the margin bandwidth BWm, preferably within the limit of a maximum terminal BWmax predefined for said critical data flow. For example, the predefined maximum limit BWmax is predefined according to the type of service (e.g. VoIP, TV, Data) associated with said critical data stream. Then step 402 is repeated.
    It should be noted that, if a critical data stream is known by the interconnection device 110 prior to its transit via the interconnection device 110 (eg by configuration or analysis of prior protocol messages), then the reservation of the band initial pass-through for said critical data stream may be equal to the sum of an expected bit rate BWp for said critical data stream (eg as apparent in said configuration or in said prior protocol messages), and the margin bandwidth BWm, preferably within the limit of the maximum limit BWmax predefined for said critical data stream.
    Fig. 5 schematically illustrates a reservation of bandwidth following the execution of the algorithm of FIG. 3 or the algorithm of FIG. 4.
    In the schematic representation of FIG. 5 the actual flow rate BWc measured for said critical data flow. The reserved bandwidth BWr for said critical data stream is therefore the sum of the effective bit rate BWc plus the bandwidth of the margin BWm. The margin bandwidth BWm appears in hatching on the representation of FIG. 5. The margin bandwidth BWm appears on each side of the margin bandwidth BWm: two thresholds are thus defined, a minimum threshold (BWr - BWm - BWt) below which a next measurement of the effective bit rate of said critical data stream will result in a reduction in bandwidth reservation, and a maximum threshold (BWr + BWt) beyond which a next measurement of the actual throughput of said critical data stream will result in an increase in bandwidth reservation.
    The algorithms of Figs. 3 and 4 have been described as part of a common monitoring of all the critical data flows that pass between the first 101 and second 102 communication networks via the interconnection device 110. Such a monitoring mechanism may, however, be implemented independently for each critical data stream to be monitored (step 408 of the algorithm of Fig. 4 then being omitted). In this case, the frequency at which the monitoring mechanism is activated can be adjusted independently for each critical data stream that passes between the first 10 and second 102 communication networks via the interconnection device 110.
    Fig. 6 schematically illustrates an algorithm, implemented by the interconnection device 110, for defining the duration of a period T between two successive activations of the monitoring mechanism for a critical data stream that passes via said interconnection device 110. algorithm of FIG. 6 is preferably executed at the end of the execution of the algorithm of FIG. 3 or that of FIG. 4.
    In a step 601, the interconnection device 110 obtains actual flow information BWc from said critical data stream. This information results from the measurement performed in step 302 or step 409, at least during the previous execution of the monitoring mechanism (ie the algorithm of Fig. 3 or that of Fig. 4) .
    In a next step 602, the interconnection device 110 checks whether a criterion of stability of the effective bit rate BWc of said critical data stream is respected. According to a particular embodiment, said stability criterion is respected when each actual flow rate measurement BWc of said critical data stream during the last N runs (N> 1) are less than the reserved bandwidth BWr and greater than the reserved bandwidth. BWr to which is subtracted the margin bandwidth BWm (BWr - BWm), insofar as said reserved bandwidth BWr has not been adjusted during said N previous executions of the monitoring mechanism (ie of the algorithm of FIG. 3 or that of FIG. 4) for said critical data stream. According to another particular embodiment, said stability criterion is respected when said reserved bandwidth BWr has not been adjusted during said N last executions of the monitoring mechanism {i.e. of the algorithm of FIG. 3 or that of FIG. 4) for said critical data stream. In the case where the criterion of stability of the effective flow BWc of said flow of critical data is respected, a step 503 is performed; otherwise, a step 504 is performed.
    In step 503, the period T between two successive activations of the monitoring mechanism for said critical data stream is increased, within the limit of a predefined maximum limit Tmax. Said period T can be increased by a step of predefined increment, within the limit of said maximum terminal Tmax predefined. In a preferred embodiment, said period T is increased by a certain percentage (factor "a") by the value of said period T before increase, within the limit of the maximum terminal Tmax predefined. For example, said period T is doubled.
    In step 503, the period T between two successive activations of the monitoring mechanism for said critical data stream is decreased. Said period T can be decreased by a predefined decrement step, within the limit of a predefined minimum terminal Tmin. In a preferred embodiment, said period T is reset to the value of the minimum terminal Tmin predefined.
    At the moment when a new critical data stream to be monitored is detected by the interconnection device 110, the period T between two successive activations of the monitoring mechanism for said critical data stream is preferably fixed at the value of the predefined minimum terminal Tmin. , but could also be at the value of the predefined maximum Tmax (assuming that the critical data flow will not significantly change in terms of throughput when the critical data flow has just been established) or at an intermediate value between the predefined minimum terminal Tmin and the predefined maximum terminal Tmax.
    权利要求:
    Claims (10)
    [1" id="c-fr-0001]
    1) Bandwidth management method implemented by an interconnection device (110) between a first communication network (101) and a second communication network (102), critical data flows and non-critical data streams transiting between the first and second communication networks via said interconnection device, the critical data streams having, in contrast to the non-critical data flows, a need for bandwidth reservation to ensure rate and latency constraints, characterized in that the interconnection device implements a mechanism for monitoring critical data flows passing between the first and second communication networks via said interconnection device so that for each critical data stream passing between the first and second communication networks via said interconnection device, the interconnection device ef perform the following steps: measure (302; 409) an effective throughput of said critical data stream; performing (303; 410) a first checking of whether said actual measured rate is greater than the sum of a bandwidth previously reserved for said critical data stream and a tolerance bandwidth; performing (303; 410) a second check consisting of checking whether said measured actual rate is less than the bandwidth previously reserved for said critical data stream, to which is subtracted the sum of a margin bandwidth and a band passing of tolerance; and in the case of a first positive check or a second positive check, adjust (304; 411) the bandwidth reservation for said critical data stream to a value equal to the sum of the actual measured rate for said critical data stream and the margin bandwidth.
    [0002]
    2) Method according to claim 1, characterized in that the bandwidth reservation for said critical data stream is adjusted within the limit of a maximum terminal associated with said critical data stream.
    [0003]
    3) Method according to any one of claims 1 and 2, characterized in that said interconnection device activates said monitoring mechanism, for all of said critical data flows passing between the first and second communication networks via said device interconnection periodically.
    [0004]
    4) Method according to any one of claims 1 and 2, characterized in that said interconnection device dynamically defines, for each flow of critical data transiting between the first and second communication networks via said interconnection device, a period time between two successive activations of the monitoring mechanism by performing the following steps: checking (602) whether a criterion of stability of the effective flow of said flow of critical data is respected; in case of positive verification with respect to said stability criterion, lengthening (603) the period of time between two successive activations of the monitoring mechanism for said critical data flow; and in case of negative verification with respect to said stability criterion, reducing (604) the period of time between two successive activations of the monitoring mechanism for said critical data stream.
    [0005]
    5) Method according to claim 4, characterized in that, in the event of a negative verification with respect to said stability criterion, said period of time between two successive activations of the monitoring mechanism for said critical data stream is reset (604 ) to a predefined minimum terminal value.
    [0006]
    6) Method according to any one of claims 4 and 5, characterized in that said stability criterion is met when each actual flow measurement of said flow of critical data in the N last runs, N> 1, are less than the band reserved passerby, insofar as said reserved bandwidth has not been adjusted during said N previous executions of the monitoring mechanism for said critical data stream.
    [0007]
    7) Method according to any one of claims 4 and 5, characterized in that said stability criterion is met when said reserved bandwidth has not been adjusted during said N last executions, N> 1, of the monitoring mechanism for said flow of critical data.
    [0008]
    8) Computer program, characterized in that it comprises instructions for implementing, by a processor of an interconnection device between a first communication network and a second communication network, the method according to one any of claims 1 to 7, when said program is executed by said processor.
    [0009]
    9) Storage means, characterized in that they store a computer program comprising instructions for implementing, by a processor of an interconnection device between a first communication network and a second communication network, the Method according to any one of claims 1 to 7, when said program is executed by said processor.
    [0010]
    An interconnection device (110) between a first communication network (101) and a second communication network (102), critical data streams, and non-critical data streams transiting between the first and second communication networks via said interconnection device, the critical data streams having unlike the non-critical data streams a bandwidth reservation need to ensure rate and latency constraints, characterized in that the interconnection device implements a monitoring mechanism critical data streams passing between the first and second communication networks via said interconnection device such that, for each critical data stream passing between the first and second communication networks via said interconnection device, the device interconnection implements: means for measuring (302; 409) an actual flow rate of said critical data flows; means for performing (303; 410) a first check consisting of checking whether said measured actual rate is greater than the sum of a bandwidth previously reserved for said critical data stream and a tolerance bandwidth; means for performing (303; 410) a second check consisting of checking whether said measured actual rate is less than the bandwidth previously reserved for said critical data stream, to which is subtracted the sum of a margin bandwidth and a bandwidth of tolerance; and means for, in the case of a first positive check or a second positive check, adjusting (304; 411) the bandwidth reservation for said critical data stream to a value equal to the sum of the actual measured rate of flow for said data stream. critical and margin bandwidth.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3329645B1|2020-01-01|Method for managing bandwidth by means of a device for interconnecting communication networks
    FR2805112A1|2001-08-17|METHOD AND UNIT FOR CONTROLLING THE FLOW OF A TCP CONNECTION ON A CONTROLLED SPEED NETWORK
    EP2862324B1|2016-10-19|Method and device for quick, unobtrusive estimation of the available bandwidth between two ip nodes
    EP1513287A2|2005-03-09|Apparatus for processing measurements in a communication network
    FR2926939A1|2009-07-31|DATA TRANSMISSION METHOD WITH ACQUITTATION ANTICIPATION, INPUT DEVICE, COMPUTER PROGRAM PRODUCT, AND CORRESPONDING STORAGE MEDIUM
    US20150063347A1|2015-03-05|Systems and methods of improving the quality of voip communications
    FR2933834A1|2010-01-15|METHOD FOR MANAGING DATA STREAM TRANSMISSION ON A TUNNEL TRANSPORT CHANNEL, TUNNEL HEAD, COMPUTER PROGRAM PRODUCT, AND CORRESPONDING STORAGE MEDIUM.
    EP3704888B1|2022-01-05|Primary channel selection method for wireless communications
    EP1424832A1|2004-06-02|End-to-end measuring device of network information
    FR3023093A1|2016-01-01|METHOD FOR AUTHORIZING THE ESTABLISHMENT OF A PAIR TO BE MIXED IN A MOBILE TELECOMMUNICATIONS NETWORK
    EP1515489B1|2007-11-28|Communication network with traffic management through active configurable measures
    WO2008074959A2|2008-06-26|System for reserving a pass band for different classes of traffic
    EP1372295A1|2003-12-17|Device and method for controlling profiles, namely of data-flows, in a communications network
    WO2014135794A1|2014-09-12|Congestion control method for telecommunications networks
    EP3085035B1|2019-03-13|Technique for signalling congestion in a packet communication network
    EP3117573A1|2017-01-18|Method of managing congestion of a transmission link in a packet communication network
    FR3095729A1|2020-11-06|Methods and devices for measuring reputation in a communication network
    CA3087762A1|2021-02-28|Method for configuring a wireless communication range extender system and a wireless communication range extender system implementing said method
    WO2019133565A1|2019-07-04|Statistical traffic classification with adaptive boundaries in a broadband data communications network
    FR3055504A1|2018-03-02|METHOD FOR CONTROLLING THE LOAD OF A DATA CONCENTRATION GATEWAY FOR A WIRELESS COMMUNICATION NETWORK
    FR3047623A1|2017-08-11|DATA TRANSMISSION METHOD IN MULTI-PATH COMMUNICATION
    FR2908577A1|2008-05-16|Quality of service resource managing method for e.g. domestic network, involves sending information representing accepted value to receiver device to initialize transmission of content based on window size equal to accepted value
    EP3231137A1|2017-10-18|Overlay network for communication network connecting data centres of a cloud services provider
    WO2009122071A2|2009-10-08|Method for managing device load in a system for implementing an application service
    WO2011101576A1|2011-08-25|Service access management in a network
    同族专利:
    公开号 | 公开日
    EP3329645B1|2020-01-01|
    CN107925632A|2018-04-17|
    US10367755B2|2019-07-30|
    US20180219796A1|2018-08-02|
    EP3329645A1|2018-06-06|
    CN107925632B|2021-05-25|
    WO2017021129A1|2017-02-09|
    BR112018001984A2|2018-09-18|
    FR3039729B1|2018-07-13|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20100094989A1|2008-10-14|2010-04-15|Jian Li|Methods and apparatus to allocate bandwidth between video and non-video services in access networks|
    EP2573997A1|2011-09-26|2013-03-27|Thomson Licensing|Method for controlling bandwidth and corresponding device|
    US20130275578A1|2012-04-13|2013-10-17|CirrusWorks, Inc.|Method and apparatus for dynamic bandwidth allocation for optimizing network utilization|FR3106710A1|2020-01-28|2021-07-30|Naval Group|DATA FLOW EXCHANGE MANAGEMENT MODULE IN AN EXCHANGE ARCHITECTURE FOR MOBILE DEVICE TRAINING|
    FR3106712A1|2020-01-28|2021-07-30|Naval Group|DATA FLOW EXCHANGE ARCHITECTURE IN A MOBILE DEVICE TRAINING|US7257632B2|2001-07-30|2007-08-14|Fujitsu Limited|Method and apparatus for a bandwidth broker in a packet network|
    US7483380B2|2004-08-30|2009-01-27|Motorola, Inc.|Method and apparatus for bandwidth management|
    CN101237391B|2007-02-01|2010-12-08|中兴通讯股份有限公司|Channel bandwidth dynamic adjusting device and method|
    US7899024B2|2007-02-28|2011-03-01|Intel Corporation|Method and apparatus to support VoIP calls in an IEEE 802.16 interface|
    US20100284423A1|2008-01-24|2010-11-11|Mitsubishi Electric Corporation|Bandwidth guaranteed communication system|
    JP5082145B2|2008-09-04|2012-11-28|日本電気株式会社|Node device and bandwidth control method thereof|
    CN102201968A|2010-03-25|2011-09-28|华为技术有限公司|Method, device and system for distributing user bandwidth in Internet protocol television |
    US20130100803A1|2011-10-21|2013-04-25|Qualcomm Incorporated|Application based bandwidth control for communication networks|
    CN103326961B|2013-06-14|2016-06-29|中国人民解放军信息工程大学|Reserved bandwidth self-adaption method of adjustment based on QoS|CN108353321B|2016-11-04|2021-02-09|华为技术有限公司|Network hotspot control method and related equipment|
    US11240049B2|2019-06-05|2022-02-01|International Business Machines Corporation|Automatic recharging of data quota for successful completion of transaction|
    法律状态:
    2016-06-22| PLFP| Fee payment|Year of fee payment: 2 |
    2017-02-03| PLSC| Publication of the preliminary search report|Effective date: 20170203 |
    2017-06-21| PLFP| Fee payment|Year of fee payment: 3 |
    2018-06-21| PLFP| Fee payment|Year of fee payment: 4 |
    2020-06-23| PLFP| Fee payment|Year of fee payment: 6 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1557419|2015-07-31|
    FR1557419A|FR3039729B1|2015-07-31|2015-07-31|METHOD FOR MANAGING BANDWIDTH BY AN INTERCONNECTION DEVICE OF COMMUNICATION NETWORKS|FR1557419A| FR3039729B1|2015-07-31|2015-07-31|METHOD FOR MANAGING BANDWIDTH BY AN INTERCONNECTION DEVICE OF COMMUNICATION NETWORKS|
    CN201680050067.3A| CN107925632B|2015-07-31|2016-07-18|Bandwidth management method, storage device and interconnection equipment|
    BR112018001984-4A| BR112018001984A2|2015-07-31|2016-07-18|bandwidth management process by a communication network interconnect device|
    EP16742232.8A| EP3329645B1|2015-07-31|2016-07-18|Method for managing bandwidth by means of a device for interconnecting communication networks|
    US15/748,341| US10367755B2|2015-07-31|2016-07-18|Method for managing bandwidth by a communication network interconnection device|
    PCT/EP2016/067020| WO2017021129A1|2015-07-31|2016-07-18|Method for managing bandwidth by means of a device for interconnecting communication networks|
    [返回顶部]