Method for aiding the reservation of resources for a packet-switched network, and associated management device and aid device

Abstract

A method intended to aid in the reservation of resources within a packet-switched communication network (CR) comprising a number N of relaying nodes (S1-S4). This method consists i), in determining for some at least of the N nodes (S1-S4) at least one resource reservation plan Pik defined by a plan identifier associated with a stream identifier and at least two ports, and ii) in memorizing for each of the N nodes (S1-S4) the various plans Pik relating to them.

Claims

1. A device for aiding resource reservation of a relaying node in a packet-switched communication network, said resource reservation for said relaying node being established following a resource reservation plan, wherein the device for aiding resource reservation comprises: a storage that stores information on at least two resource reservation plans associated with said relaying node, each of said at least two resource reservation plans being defined by a plan identifier characterizing a respective one of said at least one resource reservation plan to be used by said relaying node and being associated with at least one stream identifier and at least two ports of said relaying node, a receiver for receiving a signal comprising a current reservation plan identifier characterizing a current reservation plan to be used by said relaying node among said at least two resource reservation plans, at least one processor configured for selecting said current reservation plan among said at least two resource reservation plans, and at least one output for transmitting instructions for said relaying node to apply said current reservation plan.

2. The device for aiding resource reservation according to claim 1, wherein said storage stores information on at least one of said at least two resource reservation plans associated with at least two streams.

3. The device for aiding resource reservation according to claim 1, wherein said at least one stream identifier can associate with different ones of said at least two ports of said relaying node based on the respective one of the at least two resource reservation plans associated with said at least one stream identifier.

4. A method for aiding resource reservation of a relaying node in a packet-switched communication network, said resource reservation for said relaying node being established following a resource reservation plan, comprising: storing information on at least two resource reservation plans associated with said relaying node, each of said at least two resource reservation plans being defined by a plan identifier characterizing a respective one of said at least one resource reservation plan to be used by said relaying node and being associated with at least one stream identifier and at least two ports of said relaying node, receiving a signal comprising a current reservation plan identifier characterizing a current reservation plan to be used by said relaying node among said at least two resource reservation plans, selecting by at least one processor said current reservation plan among said at least two resource reservation plans, and transmitting instructions for said relaying node to apply said current reservation plan.

5. The method for aiding resource reservation according to claim 4, wherein storing includes storing information on at least one of said at least two resource reservation plans associated with at least two streams.

6. The method for aiding resource reservation according to claim 4, wherein said at least one stream identifier associates with different ones of said at least two ports of said relaying node based on the respective one of the at least two resource reservation plans associated with said at least one stream identifier.

7. A relaying node in a packet-switched communication network, comprising: a storage that stores information on at least two resource reservation plans associated with said relaying node, each of said at least two resource reservation plans being defined by a plan identifier characterizing a respective one of said at least one resource reservation plan to be used by said relaying node and being associated with at least one stream identifier and at least two ports of said relaying node, a signal receiver for receiving a signal including a current reservation plan identifier characterizing a current reservation plan to be used by said relaying node among said at least two resource reservation plans, at least one processor configured for selecting said current reservation plan among said at least two resource reservation plans, and at least one output for transmitting instructions for said relaying node to apply said current reservation plan wherein said relaying node applies said current reservation plan included in said signal.

8. The relaying node according to claim 7, wherein said signal receiver is a broadcast receiver.

9. The relaying node according to claim 7, wherein said relaying node comprises a signal transmitter for transmitting a signal comprising a modification notification of one of said resource reservation plans to the attention of a device for managing resource reservations for said packet-switched communication network.

10. The relaying node according to claim 7, wherein said relaying node applies said current reservation plan communicated with said instructions immediately after having received said signal.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Other characteristics and advantages of the invention will appear upon examination of the detailed description hereafter, and the annexed drawings, wherein:

(2) FIG. 1 shows in a diagrammatic and functional manner a partial example of the core of a wired local network of Ethernet/IP type comprising four items of switching equipment each equipped with an embodiment of the aid device according to the invention and of ports to which are connected, according to a first configuration, nine items of communication equipment some of which are equipped with an embodiment of the management device according to the invention,

(3) FIG. 2 shows in a diagrammatic and functional manner an example of a tree structure defining the connections established between the items of communication equipment and the items of switching equipment of FIG. 1, in accordance with the first configuration,

(4) FIG. 3 is identical to FIG. 1, but with a second connection configuration between the 4 items of switching equipment and the 9 items of communication equipment, and

(5) FIG. 4 shows in a diagrammatic and functional manner an example of a tree structure defining the connections established between the items of communication equipment and the items of switching equipment of FIG. 3, in accordance with the second configuration,

(6) The annexed drawings can be used not only to complete the invention, but also to contribute to its definition, if necessary.

DETAILED DESCRIPTION

(7) The purpose of the invention is to facilitate (if possible while accelerating) the reservation of resources within a packet-switched communication network.

(8) In what follows, it is considered, as a non-restrictive example, that the packet-switched communication network is a wired local network of Ethernet/IP type (or LAN (Local Area Network)—standard IEEE 802.3). But, the invention is not limited to this type of packet-switched communication network. It relates in fact to any type of packet-switched communication network, asynchronous, and particularly the networks of type MPLS, GMPLS, ATM, Frame Relay (FR), TDMA, CDMA, IP and wireless Ethernet (or WLAN (Wireless Local Area Network)).

(9) A non-restrictive example of a local wired core network CR of type Ethernet/IP comprising four relaying nodes Si (i=1 to 4) of ports to which are connected, according to a first configuration, nine items of communication equipment (or hosts) Hj (j=1 to 9).

(10) In what follows, it is considered as a non-restrictive example that the four relaying nodes S1 to S4 are Ethernet/IP switches. But, the invention is not limited to this type of relaying node. It relates in fact to any type of equipment belonging to the communication core network and having a role in the relaying of stream packets. Consequently, it can also involve routers for example.

(11) Moreover, it is considered in what follows, as a non-restrictive example, that the items of communication equipment (or hosts) Hj are items of studio equipment implicated in the creation of multimedia contents that must be broadcast in streaming mode. Consequently, it can involve cameras, delay servers, audio and/or video data players and/or recorders, microphones, headphones, audio and/or video mixing tables, or management computers. But, the invention is not limited to this type of communication equipment. It relates in fact to any type of communication equipment that can be connected to a relaying node Si of a packet-switched communication network. Consequently, it can also involve fixed or portable computers, audio and/or video content receivers (possibly of STB (Set-Top Box) or “Home Gateway” type), fixed or mobile telephones, communicating digital personal assistants (or PDAs), or content servers.

(12) It will be noted that the decision to connect items of communication equipment Hj to switches Si is well adapted to application with a (strong) temporal constraint, such as for example audio and/or video streaming applications. In fact, the switches are items of network equipment that best rest the temporal constraints as they manage the transfer of stream data packets according to the information contained, on the one hand, in their headers that belong to the layer 2 (transport layer called the “data link” layer) of the OSI layers model, and on the other hand, in a transfer table (or “forwarding table”—a table associating a hardware address with a switch port).

(13) The invention proposes implementing within the core of the packet-switched communication network CR, comprising a number N of relaying nodes Si (here i=1 to 4, N=4), a method for aiding in the reservation of resources.

(14) This method consists, on the one hand, in determining for some at least of the N (relaying) nodes Si at least one resource reservation plan Pik (k≧1) that is defined by a plan identifier associated with a stream identifier and at least two ports, and on the other hand, in memorizing for each of the N (relaying) nodes Si the various plans Pik relating to them.

(15) The determination of a new plan Pik for a node Si is preferably made each time it receives a message requesting a new resource reservation or pre-reservation for a stream. This new plan Pik must then be adapted to this new requested resource reservation or pre-reservation.

(16) It will be noted that it is advantageous that each determination of a plan Pik is made at the level of the node Si that is able to be used. For this purpose, and as shown, each node Si can for example be equipped with a resource reservation aid device D2.

(17) It is understood here by “equipped” the fact to comprise internally, as shown non-restrictively, or to be connected to an aid device D2.

(18) It is important to note that an aid device can be considered either of centralized type and thus adapted to the determination of reservation plans Pik for several or even all nodes SI.

(19) An aid device D2, according to the invention, comprises a calculation module MC and storage means MS2.

(20) The calculation module MC is charged with determining for its (one) node Si, according to messages that this latter has received requesting a resource reservation or pre-reservation, at least one resource reservation plan Pik (as defined above).

(21) It is important to note that the aforementioned messages have not necessarily the aim of requesting immediate progression to the reservation of a resource. This may be the case. But, it is advantageous to provide for a sort of learning phase at the level of each node Si in order to provide in advance for all the configurations that it may encounter taking account of the items of communication Hj and other nodes Si′ that are paired to it, and thus to determine in advance a reservation plan Pik for each possible configuration. Consequently, a message can thus have as an objective to request a resource “pre-reservation”.

(22) It is equally important to note that each reservation plan Pik preferably also has a plan identifier that is unique within the core network CR. A reservation plan Pik is in fact determined for a given node Si. It thus corresponds to specific local resources of the node Si (as for example the filtering policies at input and output, the buffer memories at input and output, the priority rules and the shaping rules that must be used in order to ensure for the data streams a Quality of Service (QoS) in accordance with the requirements of the applicant application that is at the origin of said streams.

(23) For example, a reservation plan Pik can be defined as follows:

(24) TABLE-US-00001 Stream_spec { Integer StreamId Integer Port_number_Lst [ ] } Reservation_Plan { Integer Reserv_Plan_Id Stream_spec Data_flow_Lst [ ] },
where “Stream_spec” relates to a stream designated by a stream identifier “StreamId” and associated with a list of at least two ports of a node Si “Port_number_Lst” and “Reservation_Plan” relate to a reservation plan designated by a plan identifier “Reserv_Plan_Id” (or Pik) and relating to a stream constituted of a list of at least one stream of data packets “Dataflow_Lst”.

(25) The storage means MS2 are intended to store each plan definition Pik that was determined by the calculation module MC. Any storage means known to those skilled in the art can be used, and particularly a memory (possibly software). This storage can for example be made in the form of a table of reservation plans Pik. It will be understood that such a table comprises the definition of at least one reservation plan Pik, and preferably of several (for the reasons invoked above).

(26) In the first connection configuration that is shown in FIGS. 1 and 2, there is: a node S1 having its port 1 connected to the node S4 (with which there is no stream relaying planned), the ports 2 and 4 are respectively connected to the items of equipment H2 and H8 (source) (for the purpose of relaying the stream from the item of equipment H8 to the item of equipment H2), and the ports 3 and 5 respectively connected to the nodes S2 and S3 (for the purpose of relaying the stream from the item of equipment H6 (source) to the items of equipment H7 and H1), a node S2 having the its ports 1 and 2 respectively connected to the node S1 and the item of equipment H1 (for the purpose of relaying the streams from the item of equipment H6 and intended for the item of equipment H1) and it ports 3 and 4 respectively connected the items of equipment H3 (source) and H5 (for the purpose of relaying the stream from the item of equipment H3 to the item of equipment H5), a node S3 having its port 1 connected to the item of equipment H6 (source) (for the purpose of relaying the stream from this item of equipment H6) and the ports 2 and 3 respectively connected to the node S1 (for the purpose of relaying the streams from the item of equipment H6 and intended for the item of equipment H1) and the item of equipment H7 (for the purpose of relaying the streams from the item of equipment H6 and intended for this item of equipment H7), a node S4 having the ports 1 and 3 respectively connected to the items of equipment H4 (source) and H9 (for the purpose of relaying the stream from the item of equipment H4 to the item of equipment H9), and its port 2 connected to the node S1 (with which there is no stream relaying planned).

(27) It will be understood that in this first connection configuration there are four streams relayed via four nodes S1 to S4.

(28) The first stream F1 relates to the items of equipment H2 to H8. It is shown in FIGS. 1 and 2 by thick dotted and continuous lines.

(29) The second stream F2 relates to the items of equipment H1, H6 and H7. It is shown in FIGS. 1 and 2 by discontinuous (dashes) lines.

(30) The third stream F3 relates to the items of equipment H3 to H5. It is shown in FIGS. 1 and 2 by dotted lines.

(31) The fourth stream F4 relates to the items of equipment H4 to H9. It is shown in FIGS. 1 and 2 by discontinuous lines (alternating dashes and double-dots).

(32) But, a second connection configuration of the type shown in FIGS. 3 and 4 can also be considered.

(33) In this second connection configuration, there is: the node S1 that has its port 1 connected to the node S4 (for the purpose of relaying streams from the item of equipment H4 and intended for the item of equipment H8), the ports 2 and 4 respectively connected to the items of equipment H2 and H8 (for the purpose of relaying streams from the item of equipment H8 to the item of equipment H2), and the ports 3 and 5 respectively connected to the nodes S2 and S3 (for the purpose, on the one hand, of a relaying of streams from the item of equipment H6 and intended for the item of equipment H1, and on the other hand, a relaying of the streams from the item of equipment H3 and intended for the item of equipment H7), a node S2 that has its ports 3 and 4 respectively connected to the items of equipment H3 (source) and H5 (for the purpose of relaying streams from the item of equipment H3 to the item of equipment H5), and its ports 1 and 2 respectively connected to the node S1 and the item of equipment H1 (for the purposes, on one hand, of relaying the streams from the item of equipment H6 and intended for the item of equipment H1, and on the other hand, of relaying the streams from the item of equipment H3 and intended for the item of equipment H7), a node S3 having its port 1 connected to the item of equipment H6 (source) (for the purpose of relaying the streams from this item of equipment H6 and intended for the item of equipment H1), and the ports 2 and 3 respectively connected to the node S1 (for the purpose of relaying the streams from the item of equipment H6 and intended for the item of equipment H1) and to the item of equipment H7 (for the purpose of relaying the streams from the item of equipment H3 and intended for this item of equipment H7), a node S4 having the ports 1 and 3 respectively connected to the items of equipment H4 (source) and H9 (for the purpose of relaying the stream from the item of equipment H4 to the item of equipment H9), and its port 2 connected to the node S1 (for the purpose of relaying the streams from the item of equipment H4 and intended for the item of equipment H8).

(34) It will be understood that in this second connection configuration there are also four streams relayed via four nodes S1 to S4.

(35) The first stream F1 relates to the items of equipment H2 to H8. It is shown in FIGS. 3 and 4 by thick dotted and continuous lines.

(36) The second stream F2 relates to the items of equipment H1 and H6. It is shown in FIGS. 3 and 4 by discontinuous (dashes) lines.

(37) The third stream F3 relates to the items of equipment H3 H5 and H7. It is shown in FIGS. 3 and 4 by dotted lines.

(38) The fourth stream F4 relates to the items of equipment H4, H8 and H9. It is shown in FIGS. 3 and 4 by discontinuous lines (alternating dashes and double-dots).

(39) Due to these first and second possible connection configurations, the to following tables of reservation plans Pik (or Reserv_Plan_Id) can be established: for the node S1:

(40) TABLE-US-00002 Reserv_Plan_Id StreamId Port_Number_Lst 1 1 2, 4 2 3, 5 5 1 2, 4 2 3, 5 3 3, 5 4 1, 4 for the node S2:

(41) TABLE-US-00003 Reserv_Plan_Id StreamId Port_Number_Lst 2 2 1, 2 3 3, 4 6 2 1, 2 3 1, 3, 4 for the node S3:

(42) TABLE-US-00004 Reserv_Plan_Id StreamId Port_Number_Lst 3 2 1, 2, 3 7 2 1, 2 3 2, 3 for the node S4:

(43) TABLE-US-00005 Reserv_Plan_Id StreamId Port_Number_Lst 4 4 1, 3 8 4 1, 2, 3

(44) It is deduced from the preceding tables, that at a given instant there is a set of eight reservation plans Pik that have been determined for the four nodes S1 to S4 of the core network CR (but only four of them are used (and therefore active) at this given instant). The reservation plans having the plan identifiers (Reserv_Plan_Id) 1 and 5 correspond respectively to P11 and P15 and can be used by the node S1. The reservation plans having the plan identifiers (Reserv_Plan_Id) 2 and 6 correspond respectively to P22 and P26 and can be used by the node S2. The reservation plans having the plan identifiers (Reserv_Plan_Id) 3 and 7 correspond respectively to P33 and P37 and can be used by the node S3. The reservation plans having the plan identifiers (Reserv_Plan_Id) 4 and 8 correspond respectively to P44 and P48 and can be used by the node S4.

(45) The attribution of plan identifiers (Reserv_Plan Id) to the different reservation plans Pik can be carried out as indicated hereafter.

(46) When a new reservation plan Pik must be determined for one of the N nodes Si, a plan identifier can be attributed to it for which the value is higher than a unit of the highest value known of a plan identifier existing in the core network CR. For example, if it is known that the highest plan identifier value is equal to 5, a plan identifier value equal to 6 (or 5+1) will be attributed to the new reservation plan Pik.

(47) It will be understood that in the presence of an aid device D2 in a node Si, it is its calculation module MC that will determine the value that it must attribute to the plan identifier of each new reservation plan Pik.

(48) To do this, the aid device D2 can be informed by the core network CR of all the plan identifier values that have already been attributed. But, it may proceed in a different way in the absence of such information. In fact the calculation module MC of a node Si (having need of a new plan Pik) can for example begin by attributing to this latter a provisional plan identifier whose value is higher than a unit at the highest plan identifier value attributed to a reservation plan Pik of the table (local) that is stored in the storage means MS2 of its aid device D2. Then, it can command its node Si to interrogate each of the N-1 other nodes Si′ (i′≠i) of the core network CR, in order to determine if one of them already has a plan Pi′k that is associated with this provisional plan identifier. Thus, in the negative it can then, after having possibly waited the extent of a selected timeout, it can definitively attribute the provisional plan identifier to this new plan Pik. However, in the affirmative, after having possibly waited the extent of the selected timeout, it attributes to the new plan Pik a plan identifier for which the value is higher than a unit of the highest value known of existing plan identifiers within the core network CR.

(49) Any message type, in accordance with a protocol used in the core network CR, can be used by a node Si to interrogate the N-1 other nodes Si′. Thus, it may for example involve a BPDU (Bridge Protocol Data Unit) type message containing the value of the provisional plan identifier that is intended to be attributed to a new plan Pik.

(50) Likewise, each interrogated node Si′ can respond to the node Si that interrogated it using any message type, as long as it is in accordance with a protocol used within the core network CR. Thus, each interrogated node Si′ can for example respond to a BPDU message with a “NACK BPDU” type message that indicates the highest value of a plan identifier value stored in the storage means MS2 of its aid device D2 if it is higher than that contained in the received BPDU message.

(51) It will be noted that an aid device D2, according to the invention, can be realized in the form of a combination of software modules and hardware, or of hardware or of software.

(52) At a given instant the N nodes Si use respectively N reservation plans Pik of a combination that was determined at a preceding instant from among the reservation plans previously determined and for which the definitions were for example stored at this preceding instant in the tables of the storage means MS2 of the aid devices D2 of the N nodes Si.

(53) A combination of N plans Pik for N nodes Si of the network core CR constitutes a global configuration (that itself can be seen as a type of global reservation plan). More specifically, a global configuration CG can be defined as the stable set of all the reservation plans that can be simultaneously active over a non-null period of time. It is in fact important to note that a combination is constituted of N mutually compatible plans.

(54) For example, each determination of a new combination of N plans Pik can be carried out following the transmission by one of the N nodes Si of a modification notification of one of the related plans Pik. It will be noted that such a modification may result in the reception beforehand by the node Si of a message requesting a new resource reservation or pre-reservation for a stream that it must relay or the end of a resource reservation for a stream that it has just relayed.

(55) It is important to note that a new global configuration (or combination) CG is constituted, with respect to a new global configuration (or combination) previously calculated, of at least the new plan Pik that is the origin of the transmission by one of the N nodes Si of a modification notification, and possibly of one or several other plans Pi′k (previously identified).

(56) The determinations of (new) combinations of N plans Pik, constituting global configurations CG, that can for example be made centrally in a master item of equipment, equipped for this purpose with a management device D1 according to the invention.

(57) It is understood here by “equipped” the fact of comprising internally, as shown non-restrictively, or to be connected to a management device D1.

(58) In the non-restrictive example shown in FIGS. 1 and 3, the management devices D1 have been implanted in some items of communication equipment (H1, H2, H4, H5, H6 and H8). But, management devices D1 can be implanted in at least one of the items of communication equipment Hj and/or in at least one of the nodes of the core network CR (and for example in a relaying node Si). Consequently, one at least of the nodes Si can possibly be equipped both with an aid device D2 and a management device D1.

(59) When there is only one item of equipment that has a management device D1, it is thus automatically the master item of equipment. However, when several items of equipment (Hj and/or Si) have a management device D1, it is preferable that one among them is selected to assure the role of master item of equipment.

(60) For example, this selection can be made among the items of equipment that are candidates. A candidate item of equipment is for example an item of equipment equipped with a management device D1 and that transmits in broadcast mode, and thus to all the other items of equipment, a specific message, for example of PDU (Protocol Data Unit) type, to notify them of its desire to assure the role of master.

(61) As a non-restrictive example, that which has the lowest IP address can be selected from among the candidates. In a variant, the master candidate can be selected according to the “weight” (shortest path, maximal bandwidth, etc.) of the set of links that associates it with other nodes of the network or items of equipment Hj. In another variant, the master candidate can be selected arbitrarily and “manually”, for example using a parameter of a “MIB SNMP”.

(62) A management device D1, according to the invention, comprises a management module MG and storage means MS1.

(63) Following the transmission by one of the N nodes Si of a modification notification of one of the plans Pik relating to it, the management module MG determines a new combination of N resource reservation plans Pik for N nodes Si. To do this, it determines the N plans of the new combination from among the set of definitions of all the resource reservation plans Pik that are associated with these N nodes Si, taking account of N plans that they are using respectively, and thus that are active, and taking account of said new plan.

(64) For example, the storage means MS1 store the set of definitions of all the resource reservation plans Pik that are associated with these N nodes Si. They are for example updated, possibly periodically, by the management device D1, following the interrogation of N nodes Si by its master item of equipment.

(65) Moreover, in order to know precisely the (active) plans Pik that are used by the N nodes Si at a given instant, the management module MG can for example command its master item of equipment to address to the N nodes Si, in broadcast mode, a message requesting that each of them transmit the definition of the (active) plan Pik that it is using.

(66) It will be noted that a new combination (or global configuration) CG can possibly comprise one or several plans identical to those used in a combination (or global configuration) previously determined (and possibly being used) by the N nodes Si.

(67) If the new determined combination (or global configuration) CG must be applied, then it proceeds to the switching from the old combination (or global configuration), to this new combination (or global configuration). In this case, the management device D1 of the master item of equipment commands this latter to address to the N nodes Si, in broadcast mode, a message (for example of PDU type) containing the list of N plan identifiers that they must immediately apply. When a node Si receives this message, it locally manages its internal resources (particularly its input and output buffer memories) so that they are adapted to the plan Pik of the new combination that it must use (it will be understood that a node Si has nothing to adapt when it is requested to use the same plan as it has just been using).

(68) It will be noted that a management device D1, according to the invention, can be realized in the form of a combination of software modules and hardware, or of hardware, or again of software modules.

(69) The invention is not limited to the embodiments of the management device, aid device, communication equipment, relaying node and aid method described above, only as an example, but it encompasses all the variants that those skilled in the art can envisage within the framework of the claims that follow hereafter.