Network relaying method and device

BACKGROUND OF THE INVENTION

00011. Field of the Invention

0002The present invention relates to a network relaying method and device, and in particular to a network relaying method and device at the time of distributing data from a server to a client by multicast.

00032. Description of the Related Art

0004Together with a recent rapid spread of a personal computer, an intranet has been expanding, while a computer and a network themselves have become advanced and sophisticated. Also, the spread of web pages (WWW) and multimedia data such as moving images and voices have been rapidly advanced.

0005Furthermore, the spread of a broadband network has been recently advanced in accesses over the Internet including enterprise users and general users. The use of the voices and the moving images so-called multimedia, and a business form by distributing and broadcasting services using the multimedia have been increasing.

0006Presently, in the course of spread, a method of distributing data to each user by a unicast technology has been used in many cases. Especially in the moving image distribution accessed by numerous users, a mechanism in which numerous servers are set in data distribution sources, or a cache server is provided in every area and the users receive data from an adjoining cache server is adopted for the purpose of load balancing.

0007This is applied to an on-demand broadcasting and a real-time broadcasting. In order to provide quality which satisfies numerous users, numerous facilities and a large amount of investment are required.

0008On the other hand, in order to efficiently transmit large amounts of data, a data transmission by a “multicast” technology has begun to be performed. It is expected that a multimedia data transmission using the multicast is spreading more and more.

0009While the multicast can not perform an on-demand distribution basically, it has numerous merits in a real-time moving image distribution and voice distribution such as live broadcasting.

0010Namely, in the multicast distribution, as shown in <figref idref="DRAWINGS">FIG. 1</figref>, a server S<b>1</b> can make numerous clients R<b>1</b>, R<b>2</b>, . . . receive data through multicast routers <b>1</b>_<b>1</b> and <b>1</b>_<b>2</b> (hereinafter, occasionally represented by a reference numeral “<b>1</b>”) and LAN switches SW<b>1</b>, SW<b>2</b> (hereinafter, occasionally represented by a reference numeral “SW”) only by transmitting a single stream to a multicast-capable network MNW, regardless of the number of users (clients) who receive the data.

0011Accordingly, it is not necessary to provide a cache in each location, and a bandwidth in each network can be saved different from the unicast.

0012Put another way, a traffic amount of a certain path can be kept fixed regardless of the number of users, little influence is exerted on other communications, and facilities such as a network and a cache server specific to a distribution are not required different from the case of the unicast, so that it becomes possible to distribute the multimedia data very inexpensively.

0013There is a restriction that an entire network ranging from the server to the client is required to support the multicast. However, presently, the support for the multicast has almost been completed in a relaying device and a data distribution server set on the network, or all of the personal computers used by clients. It is expected that the ratio of the multicast usage becomes gradually higher especially in the moving image distribution and the voice distribution in a real-time broadcasting form.

0014Hereinafter, a summary of an environment and a mechanism of the multicast data distribution will be described.

0015Specifically, an IP multicast address (group address), a multicast address of a data link layer, an IGMP (Internet Group Management Protocol) that is a protocol between a client (host) and a multicast router, and a multicast routing protocol constructing a multicast distribution tree within the network will now be described respectively. It is to be noted that the data link layer will be described for an Ethernet.

0016An IP multicast distribution uses the multicast address, functions between clients R<b>1</b>, R<b>2</b> and a multicast router <b>1</b> as shown in <figref idref="DRAWINGS">FIGS. 28A and 28B</figref>, and is realized by operating the IGMP, which provides a function that the router <b>1</b> grasps the multicast group of the subordinate clients R<b>1</b> and R<b>2</b> and the multicast routing protocol which constructs a multicast data distribution tree from the server to a plurality of receiving clients between routers.

0017The IGMP is a protocol with which the router <b>1</b> grasps/manages whether or not a receiving client exists in a local network to which the router itself belongs by transferring an IGMP message of a packet format shown in <figref idref="DRAWINGS">FIG. 29</figref> between the clients R<b>1</b>, R<b>2</b> and the router <b>1</b>.

0018Put another way, the IGMP is a protocol between the router and the client for informing the router <b>1</b> that the clients R<b>1</b> and R<b>2</b> join a certain multicast group. Both of the router and the clients are required to mount a function specified by the IGMP (applied when a “protocol ID” in an IP header shown in <figref idref="DRAWINGS">FIG. 30</figref> is “2”). An IGMPv1 is specified in an appendix <b>1</b> of RFC 1112, an IGMPv2 is specified in RFC 2236, and an IGMPv3 is being standardized in IETF.

0019The multicast routing protocol is a protocol between routers for a path control (or arrangement of a path tree) determining to which interface of the router <b>1</b> multicast data are copied in order for the router <b>1</b> to distribute a multicast data stream to the receiving clients R<b>1</b> and R<b>2</b> in a network composed of a plurality of multicast routers <b>1</b> and the layer <b>3</b> switches SW as shown in <figref idref="DRAWINGS">FIG. 1</figref>.

0020As for its type, there are DVMRP (used in MBone, specified in RFC 1075), PIM (specified in RFC 2117; new version IETF is being reviewed), MOSPF (operable only on OSFP, specified in RFC 1584 and 1585), and the like.

0021It is to be noted that the above-mentioned RFC 1112 has a function of mapping between an IP address of a class D, i.e. a multicast IP address and a multicast physical address, and a filtering function of receiving only a packet of a specific multicast physical address and of transferring the packet to an upper layer.

0022A correspondence between the multicast IP address and the multicast physical address is specified (in RFC 1700) to put lower 23 bits of the multicast IP address into the lower 23 bits of the multicast physical address “01.00.5E.00.00.00₁₆”. For example, the multicast IP address “239.133.130.34” is the multicast physical address “01:00:5E:05:82:22”.

0023Also, the multicast address (IPv4) is specified as a class D address, in a range from “224.0.0.0” to “239.259.255.255” by a decimal notation.

0024<figref idref="DRAWINGS">FIG. 31</figref> shows the IP address of the class D. The class D address is identified with the first 4 bits “1110”. As shown in <figref idref="DRAWINGS">FIG. 31</figref>, some multicast addresses are reserved for specific usages.

0025Addresses assigned to local sites, i.e. from “239.0.0.0” to “239.255.255.255” are IP multicast addresses generally used by e.g. an enterprise network, an ISP, and the like.

0026A distribution procedure of the above-mentioned multicast data will now be simply described by referring to <figref idref="DRAWINGS">FIGS. 28A and 28B</figref>.

0000[1] Join to Multicast Group

0027(1) In order to query the clients R<b>1</b> and R<b>2</b> connected to the local network as to a join to a multicast group, the multicast router <b>1</b> periodically transmits to “224.0.0.1” (All-Systems-Group) an “HMQ (Host Membership Query)” message (see {circle over (1)} in <figref idref="DRAWINGS">FIG. 28A</figref>) in which a type value is “0x11” within an IGMP header of a packet in a format (IGMPv2) shown in <figref idref="DRAWINGS">FIG. 29</figref>.

0028(2) A client which desires the join to the multicast group responds to the above-mentioned “HMQ”. In order to inform the multicast address of a group which the client desires to join, the client transmits an “HMR (Host Membership Report)” message (see {circle over (2)} in <figref idref="DRAWINGS">FIG. 28A</figref>) whose type value is “0x12” within the IGMP header of the packet in the same format to the multicast address which the client joins.

0029Receiving this message, the multicast router <b>1</b> grasps the multicast group (identified by the above-mentioned class D address) which the client joins, and starts the transmission of multicast data (stream) to the local network.

0000[2] Multicast Data Distribution

0000(1) On the other hand, a transmitting server such as the server S<b>1</b> in <figref idref="DRAWINGS">FIG. 1</figref> of the multicast data transmits “single” data (stream) to the multicast group identified with the class D address.

0000(2) The multicast router <b>1</b> within the network copies the data stream addressed to the group as required to be transmitted along the path to each receiving client which joins the multicast group.

0030Put another way, the multicast data are distributed along the path tree, formed by a multicast routing protocol, of the multicast data distribution from the transmitting server to each receiving client. Finally, the “single” data stream transmitted by the server is distributed to “a plurality of” clients within the network.

0000[3] Leave from Multicast Group

0000(1) A client which desires to leave the multicast group, which the client has joined, transmits a “leave” message (see (I in <figref idref="DRAWINGS">FIG. 28B</figref>) as shown in <figref idref="DRAWINGS">FIG. 28B</figref> to “224.0.0.2” (All-Routers-Group) at the time of determining the leave.

0031(2) In order to confirm that there is no other client which has joined the multicast group, the multicast router <b>1</b> having received the “leave” message transmits a “GS-Q (Group Specific Query)” message (see {circle over (4)} in <figref idref="DRAWINGS">FIG. 28B</figref>) by designating the group address.

0032If there is another client which has joined the group except the client which has transmitted the “leave” message in this case, the other client transmits the “HMR” message to inform its existence to the multicast router <b>1</b>.

0033While there are numerous merits in the data and contents distribution using the multicast, the management of an individual user is difficult since a server does not transmit data to each client different from the case of the unicast, and the multicast itself uses a UDP (User Data Protocol) packet.

0034The difficulty of the management means that it is impossible to grasp e.g. the number of receiving clients at a certain time, and how long (from when until when) each client has received data, and to acquire and manage information of the client itself.

0035Accordingly, it has been impossible to collect information for reviewing which content is popular and how many clients have received the content, and to realize charging in accordance with a time of a data reception as a service form.

SUMMARY OF THE INVENTION

0036It is accordingly an object of the present invention to realize a network relaying method and device managing individual information of a client which joins/leaves multicast at the time of distributing data from a server to a client by multicast.

0037In order to achieve the above-mentioned object, a network relaying method according to the present invention comprises: a first step of discriminating join/leave information transmitted by a client to a multicast group; a second step of processing the join/leave information of the client into multicast join/leave notifying information; and a third step of transferring the multicast join/leave notifying information to a server.

0038The above-mentioned multicast join/leave notifying information may include identifying information of the client, a multicast address and a join/leave starting time.

0039Also, the above-mentioned first step may discriminate the join/leave information of the client mounted on a multicast control packet from the client, the second step may process the join/leave information of the client into the multicast join/leave notifying packet and the third step may transfer the multicast join/leave notifying packet to the server.

0040Also, when transferring the multicast join/leave notifying packet to the server and receiving join/leave information for a same multicast group from another client within a fixed time, the above-mentioned second step may generate an aggregated join/leave notifying packet in which the join/leave information for all of the clients is aggregated in a single packet and the third step may transfer the aggregated join/leave notifying packet to the server.

0041Furthermore, when the first step discriminates a reception of the multicast join/leave notifying packet from a downstream network device, the above-mentioned third step may further transfer the multicast join/leave notifying packet to the server.

0042Furthermore, when the first step discriminates a reception of the aggregated join/leave notifying packet from a downstream network device, the third step may further transfer the aggregated join/leave notifying packet to the server.

0043Furthermore, when the first step discriminates a reception of the aggregated join/leave notifying packet from a downstream network device, the above-mentioned second step may generate, when the join/leave information for a same multicast group from the client itself is aggregated, an aggregated join/leave notifying packet in which all of join/leave information is aggregated in a single packet and the third step may immediately transfer the aggregated join/leave notifying packet to the server.

0044Furthermore, the above-mentioned third step may hold the join/leave information of the client until a transfer is requested from the server.

0045Furthermore, a transmitting source address included in a packet may be used as identifying information of the above-mentioned client.

0046Furthermore, when the client does not respond to a periodical query, the third step may transmit leave notifying information of the client to the server.

0047Furthermore, a network relaying device for realizing the network relaying method according to the present invention comprises: a discriminating portion for discriminating join/leave information transmitted by a client to a multicast group; an information processing portion for processing the join/leave information of the client into multicast join/leave notifying information; and a transfer processor for transferring the multicast join/leave notifying information to a server.

0048Also, the above-mentioned multicast join/leave notifying information may include identifying information of the client, a multicast address and a join/leave starting time.

0049Also, the above-mentioned discriminating portion may discriminate the join/leave information of the client mounted on a multicast control packet from the client, the information processing portion may process the join/leave information of the client into the multicast join/leave notifying packet and the transfer processor may transfer the multicast join/leave notifying packet to the server.

0050Also, when transferring the multicast join/leave notifying packet to the server and receiving join/leave information for a same multicast group from another client within a fixed time, the above-mentioned information processing portion may generate an aggregated join/leave notifying packet in which the join/leave information for all of the clients is aggregated in a single packet and the transfer processor may transfer the aggregated join/leave notifying packet to the server.

0051Also, when the discriminating portion discriminates a reception of the multicast join/leave notifying packet from a downstream network device, the above-mentioned transfer processor may further transfer the multicast join/leave notifying packet to the server.

0052Also, when the discriminating portion discriminates a reception of the aggregated join/leave notifying packet from a downstream network device, the above-mentioned transfer processor may further transfer the aggregated join/leave notifying packet to the server.

0053Furthermore, when the above-mentioned discriminating portion discriminate a reception of the aggregated join/leave notifying packet from a downstream network device, the above-mentioned information processing portion may generate, when the join/leave information for a same multicast group from the client itself is aggregated, an aggregated join/leave notifying packet in which all of join/leave information is aggregated in a single packet and the transfer processor may immediately transfer the aggregated join/leave notifying packet to the server.

0054Also, the above-mentioned transfer processor may hold the join/leave information of the client until a transfer is requested from the server.

0055A transmitting source address included in a packet may be used as identifying information of the above-mentioned client.

0056Furthermore, when the client does not respond to a periodical query, the above-mentioned transfer processor may transmit leave notifying information of the client to the server.

0057Furthermore, a server may comprise: a discriminating portion for discriminating multicast join/leave notifying information transmitted from the above-mentioned network relay; and a holding portion for extracting information of the client in the multicast join/leave notifying information to be held.

0058The server may further comprise means charging by the hour a multicast data distribution (content reception) based on the client information.

0059The network relaying method and device according to the present invention as mentioned above will now be described referring to attached figures.

0060<figref idref="DRAWINGS">FIG. 1</figref> shows a network arrangement for illustrating a concept of the present invention. In <figref idref="DRAWINGS">FIG. 1</figref>, S<b>1</b> denotes a transmitting server of multicast data, <b>1</b>_<b>1</b> and <b>1</b>_<b>2</b> denote multicast (-capable) routers in which the network relaying method and device of the present invention are mounted, SW<b>1</b> and SW<b>2</b> denote LAN switches or layer <b>2</b> switches accommodating clients, MNW denotes a network composed of the multicast routers, R<b>1</b> denotes a client which receives normal data, R<b>2</b> denotes a client which receives multicast data, PKT<b>1</b> denotes a multicast control packet which functions between the receiving clients R<b>1</b>, R<b>2</b> and the multicast router <b>1</b>_<b>2</b>, and PKT<b>2</b> denotes a user information packet including join or leave (hereinafter, referred to as join/leave) information of the receiving clients and being transferred to the server S<b>1</b>. It is to be noted that the packets PKT<b>1</b> and PKT<b>2</b> are occasionally represented by a reference numeral PKT.

0061<figref idref="DRAWINGS">FIG. 2</figref> shows a diagram of a principle arrangement of a network relaying device <b>1</b> of the present invention which is a router in which a normal multicast function is mounted. The network relaying device <b>1</b> is composed of transmitting/receiving interfaces <b>2</b>, <b>7</b> and <b>8</b>, a discriminating portion <b>3</b> of the multicast control packet PKT<b>1</b>, a generator <b>4</b> of a client management information packet PKT<b>2</b> for notifying the user information to the server S<b>1</b>, and a transfer processor <b>5</b> which includes a routing table <b>6</b> and which performs routing processing (including multicast routing processing) or the like mounted on the normal network relay.

0062In operation, when the receiving client R<b>2</b> is connected to the transmitting/receiving interface <b>2</b> through the LAN switch SW<b>2</b>, the multicast control packet PKT<b>1</b> for notifying the multicast router <b>1</b> that the receiving client desires to receive the multicast data (desires to join multicast group) or stops receiving the multicast data although the data having been received (desires to leave multicast group) is transmitted from the receiving client R<b>2</b>, and the multicast router <b>1</b> receives this packet.

0063The multicast router <b>1</b> examines what the received packet is by the packet discriminating portion <b>3</b>. If the packet is a multicast control packet PKT<b>1</b>, it is transferred to the transfer processor <b>5</b> for performing a normal multicast data transfer control, so that the information is reflected on the routing table <b>6</b>.

0064At the same time, in order to manage the user information in the server S<b>1</b>, the packet generator <b>4</b> generates the multicast join/leave notifying packet storing the user information to be transferred to the transfer processor <b>5</b>, so that it is transferred from an interface <b>7</b> or <b>8</b> in the direction where the server S<b>1</b> exists.

0065The packet reaches the server S<b>1</b>, thereby enabling a reception user management of data which has been difficult by the prior art multicast data transfer to be realized.

BRIEF DESCRIPTION OF THE DRAWINGS

0066The above and other objects and advantages of the invention will be apparent upon consideration of the following detailed description, taken in conjunction with the accompanying drawings, in which the reference numerals refer to like parts throughout and in which:

0067<figref idref="DRAWINGS">FIG. 1</figref> is a diagram showing a network arrangement for illustrating a network relaying method and device according to the present invention;

0068<figref idref="DRAWINGS">FIG. 2</figref> is a block diagram of a principle arrangement of a network relaying device according to the present invention;

0069<figref idref="DRAWINGS">FIG. 3</figref> is a diagram showing a network arrangement for illustrating an embodiment of the present invention;

0070<figref idref="DRAWINGS">FIG. 4</figref> is a diagram showing a simplified network arrangement of the network arrangement of <figref idref="DRAWINGS">FIG. 3</figref>;

0071<figref idref="DRAWINGS">FIG. 5</figref> is a block diagram showing an arrangement of an embodiment of a network relaying device according to the present invention;

0072<figref idref="DRAWINGS">FIG. 6</figref> is a format diagram of a multicast join notifying packet;

0073<figref idref="DRAWINGS">FIGS. 7A and 7B</figref> are diagrams showing a multicast join notifying sequence to a server and a receiving client management table managed by the server;

0074<figref idref="DRAWINGS">FIG. 8</figref> is a diagram showing a network example at the time of aggregating client information;

0075<figref idref="DRAWINGS">FIG. 9</figref> is a format diagram of a multicast aggregated join notifying packet;

0076<figref idref="DRAWINGS">FIGS. 10A and 10B</figref> are diagrams showing a multicast aggregated join notifying sequence to a server and a receiving client management table;

0077<figref idref="DRAWINGS">FIG. 11</figref> is a diagram showing an aggregated network example of a multicast aggregated join notifying packet from a downstream direction and a multicast join desiring message information;

0078<figref idref="DRAWINGS">FIG. 12</figref> is a format diagram of a multicast aggregated join notifying packet to which information is added on the way to a server;

0079<figref idref="DRAWINGS">FIGS. 13A and 13B</figref> are diagrams showing an information addition sequence to a multicast aggregated join notifying packet and a receiving client management table;

0080<figref idref="DRAWINGS">FIG. 14</figref> is a block diagram showing an embodiment of a server (serving as client management server and multicast data transmitting server);

0081<figref idref="DRAWINGS">FIG. 15</figref> is a format diagram of a multicast (aggregated) join notifying packet identified with a protocol ID field;

0082<figref idref="DRAWINGS">FIG. 16</figref> is a format diagram showing an IP header in <figref idref="DRAWINGS">FIG. 15</figref>;

0083<figref idref="DRAWINGS">FIG. 17</figref> is a format diagram of a multicast leave notifying packet;

0084<figref idref="DRAWINGS">FIGS. 18A and 18B</figref> are diagrams showing a multicast leave notifying sequence to a server and a receiving client management table;

0085<figref idref="DRAWINGS">FIG. 19</figref> is a format diagram of a multicast aggregated leave notifying packet;

0086<figref idref="DRAWINGS">FIG. 20</figref> is a diagram of a receiving client management table managed by a server at the time of leaving multicast aggregation;

0087<figref idref="DRAWINGS">FIG. 21</figref> is a format diagram of a multicast aggregated leave notifying packet to which information is added on the way to a server;

0088<figref idref="DRAWINGS">FIG. 22</figref> is a diagram of a receiving client management table managed by the server concerning <figref idref="DRAWINGS">FIG. 21</figref>;

0089<figref idref="DRAWINGS">FIG. 23</figref> is a general format diagram of an ICMPv6 message;

0090<figref idref="DRAWINGS">FIG. 24</figref> is a general format diagram of an MLD message;

0091<figref idref="DRAWINGS">FIGS. 25A and 25B</figref> are diagrams showing a summary of a detection of a client fault, a server notification, and a management table update;

0092<figref idref="DRAWINGS">FIG. 26</figref> is a concept diagram of charging service in accordance with time when a multicast content is used;

0093<figref idref="DRAWINGS">FIG. 27</figref> is a diagram showing an example of multicast content distribution service and a time charging system using the present invention;

0094<figref idref="DRAWINGS">FIGS. 28A and 28B</figref> are diagrams for illustrating a general multicast join/leave procedure of an IGMP;

0095<figref idref="DRAWINGS">FIG. 29</figref> is a general format diagram of an IGMPv2 packet;

0096<figref idref="DRAWINGS">FIG. 30</figref> is a general format diagram of an IPv4 header; and

0097<figref idref="DRAWINGS">FIG. 31</figref> is a diagram illustrating an assignment according to a usage purpose of a multicast address.

DESCRIPTION OF THE EMBODIMENTS

0098Hereinafter, embodiments of a network relaying method and device according to the present invention will be described. As for a multicast control message, the above-mentioned IGMPv2 (specified in RFC 2236) message which has the most widely spread and has been mounted will now be described. However, the present invention can be also applied to a case where the multicast control message is of the IGMPv1 or IGMPv3.

0099Also, in the case of IPv6, the function corresponding to the IGMPv2 is called an MLD (Multicast Listener Discovery), and specified in RFC 2710. Also, the function corresponding to the IGMPv3 is being standardized by IETF. While the case where the network layer is of IPv4 will, be described in this embodiment, this embodiment can be also applied to the case of the IPv6, as a matter of course. The difference between the IPv6 multicast and the IPv4 multicast will be described later.

0100<figref idref="DRAWINGS">FIG. 3</figref> shows a general IP network such as an enterprise IP network, an IP network of an ISP (Internet Service Provider) and a carrier which are targets of the present invention. In the same way as <figref idref="DRAWINGS">FIG. 1</figref>, the IP network is composed of clients R<b>1</b>-R<b>3</b> such as personal computers, multicast-capable routers <b>1</b>_<b>1</b>-<b>1</b>_<b>8</b> (hereinafter, occasionally represented by a reference numeral “<b>1</b>”) which compose a network, a server S<b>1</b> which distributes and manages multicast data by a multicast routing protocol MRP and the like.

0101Also, as shown in <figref idref="DRAWINGS">FIG. 1</figref>, a form in which the clients R<b>1</b>-R<b>3</b> are accommodated in the LAN switch/layer <b>2</b> switch to be connected to the multicast router <b>1</b> is general in an enterprise and accesses over the Internet. However, this form is not shown for simplifying the figure. It is to be noted that client management information packets CMP<b>1</b>-CMP<b>3</b> in <figref idref="DRAWINGS">FIG. 3</figref> are respectively composed of the packets PKT<b>1</b> and PKT<b>2</b> of <figref idref="DRAWINGS">FIG. 1</figref>.

Embodiment [1]

0102<figref idref="DRAWINGS">FIG. 4</figref> shows a simplified embodiment [1] of the present invention in the network arrangement of <figref idref="DRAWINGS">FIG. 3</figref>. <figref idref="DRAWINGS">FIG. 4</figref> shows a state where the multicast join packet PKT<b>1</b> is transmitted to the multicast router <b>1</b> from the client R<b>1</b>, a multicast join notifying packet PKT<b>2</b>_<b>1</b> is transmitted from the multicast router <b>1</b> to the server S<b>1</b> in response to the packet PKT<b>1</b>, and a state where the server S<b>1</b> transmits multicast data DT.

0103It is to be noted that the IGMP packet format and the IGMP message field are the same as those shown in <figref idref="DRAWINGS">FIG. 29</figref>, and the IPv4 header shown in <figref idref="DRAWINGS">FIG. 30</figref> may be used for the IPv4 header.

0104Also, <figref idref="DRAWINGS">FIG. 5</figref> shows an arrangement of an embodiment such as a multicast router, a multicast-capable layer <b>3</b> switch of the network relay, whose principle is shown in <figref idref="DRAWINGS">FIG. 2</figref>, according to the present invention.

0105In <figref idref="DRAWINGS">FIG. 5</figref>, packet discriminating portions <b>3</b>_<b>1</b>-<b>3</b>_<b>3</b> (hereinafter, occasionally represented by a reference numeral “3”) detect a multicast join desiring message (IGMP Report) and a leave desiring message (IGMP Leave message) from a client to be transmitted to not only a multicast processor <b>9</b> which performs a known multicast control but also a timer <b>10</b> and a client information holding portion <b>11</b>.

0106The timer <b>10</b> is used when a join/leave information notifying message (packet) in which client information is aggregated is generated together with the client information holding portion <b>11</b> and the packet generator <b>4</b> described later.

0107Hereinafter, operation of the network relaying device <b>1</b> shown in <figref idref="DRAWINGS">FIG. 5</figref> will be described referring to <figref idref="DRAWINGS">FIGS. 3 and 4</figref>.

0108Firstly, the network relaying device <b>1</b> grasps the information of the receiving clients R<b>1</b>-R<b>3</b> of the multicast data existing in the subordinate of its own interfaces <b>2</b>, <b>7</b>, and <b>8</b>. Specifically, the packets transmitted from the clients R<b>1</b>-R<b>3</b> desiring the reception of the multicast data are examined by the packet discriminating portion <b>3</b>. If the packets include the IGMP reports (join desiring messages), it is recognized that the receiving client exists in the subordinate of the receiving interface.

0109As for the identification method of the IGMP report, {circle over (1)} an IP address is a multicast address, the {circle over (2)} “protocol No.” within the IP header is “2” indicating the IGMP, and {circle over (3)} the type value of the IGMP message is “0x16” are firstly confirmed.

0110As shown in <figref idref="DRAWINGS">FIG. 29</figref>, the multicast address (group address) which the receiving clients R<b>1</b>-R<b>3</b> desire to receive is stored in the IGMP report.

0111The network relaying device <b>1</b> generates the multicast join notifying packet PKT<b>2</b>_<b>1</b> with the destination address (see <figref idref="DRAWINGS">FIG. 30</figref>) as an address of the server S<b>1</b> at the packet generator <b>4</b> so that the server S<b>1</b> may manage that the client of transmitting source of the message desires to receive the multicast data (desire to join multicast). Then, the transfer processor <b>5</b> transfers the multicast join notifying packet PKT<b>2</b>_<b>1</b> to the server S<b>1</b> by normal routing processing.

0112At this time, the type value of the IGMP message is changed to be transmitted so that the server S<b>1</b> may identify the multicast join notifying message when the server S<b>1</b> receives the packet.

0113Namely, as for the type value of the IGMP message determined by the standard, as shown in <figref idref="DRAWINGS">FIG. 29</figref>, there are a version 1 (IGMPv1), a version 2 (IGMPv2) and a version 3 (IGMPv3) (being standardized by IETF (Internet Engineering Task Force) which is a standardization group of the Internet technology). However, in the present invention, the type value except the above-mentioned type values is set to the multicast join notifying message to be identified. This type value will be described later.

0114The IGMP report indicating that the clients R<b>1</b>-R<b>3</b> desire to receive multicast data DT<b>1</b>-DT<b>3</b> is transmitted to an address which the clients R<b>1</b>-R<b>3</b> desire to receive (in the IGMPv3 under review, it is transmitted to “224.0.0.22”). In order to enable the server S<b>1</b> to manage the information of the receiving clients of the multicast data DT<b>1</b>-DT<b>3</b>, the network relaying device <b>1</b>, when receiving the IGMP report, replaces its destination address field by the server address to be transmitted to the server S<b>1</b>.

0115Since the “maximum response time field” within the IGMP message shown in <figref idref="DRAWINGS">FIG. 29</figref> is not necessary for processing at this time, a method of setting “0” or the like is considered. Also, it is desirable to set a new calculated value to a “check sum” for integrity of the data.

0116A multicast address G<b>1</b> which the client desires to receive is included in the group address. Thus, when e.g. the same client desires to receive a plurality of different multicast addresses, the addresses can be independently managed.

0117<figref idref="DRAWINGS">FIG. 6</figref> shows a format of the multicast join notifying packet PKT<b>2</b>_<b>1</b> transmitted from the network relaying device <b>1</b> to the server S<b>1</b> as mentioned above. On the receiving client information, the address of the client R<b>1</b> desiring the multicast join and a join starting time t<b>1</b> are mounted.

0118The server S<b>1</b> having received this packet confirms the message type. When it is the join notifying message, the address of the client R<b>1</b> stored in the source address field (see <figref idref="DRAWINGS">FIG. 30</figref>) is associated with the address G<b>1</b> of the multicast data transmitted by the server S<b>1</b> and the reception time t<b>1</b> of the join notifying packet to be stored. Thus, the client R<b>1</b> can manage which multicast data (G<b>1</b> in this example) has been received and when (t<b>1</b> in this example) the data reception has started.

0119At this time, the network relaying device <b>1</b> accommodating the receiving client R<b>1</b> in its own interface can not recognize the multicast address G<b>1</b> and the address of the server S<b>1</b> until the multicast data are received.

0120Therefore, if dummy multicast data with less traffic amount are flowed before the transmission of the actual multicast contents in an actual operation, the network relaying device <b>1</b> can recognize the relationship between the multicast address G<b>1</b> and its server S<b>1</b>, and can grasp the relationship between G<b>1</b> and S<b>1</b> when the IGMP report is received from the client R<b>1</b>, so that the information R<b>1</b> and G<b>1</b> can be transmitted to the server S<b>1</b>.

0121As a matter of course, the session of the multicast is distinguished by the multicast address. Therefore, if the IGMP report for e.g. a multicast address G<b>2</b> except G<b>1</b> is received, the processing is performed to G<b>2</b> in parallel with G<b>1</b>.

0122<figref idref="DRAWINGS">FIG. 7A</figref> shows the above-mentioned operation sequence. When the IGMP report packet PKT<b>1</b> is transmitted from the client R<b>1</b> (see (<b>1</b>) in <figref idref="DRAWINGS">FIG. 7A</figref>), the multicast router <b>1</b> having received the packet PKT<b>1</b> generates the multicast join notifying packet PKT<b>2</b>_<b>1</b> (see (<b>2</b>) in <figref idref="DRAWINGS">FIG. 7A</figref>). The packet PKT<b>2</b>_<b>1</b> is further transmitted to the server S<b>1</b> through the multicast-capable network MNW (see (<b>3</b>) in <figref idref="DRAWINGS">FIG. 7A</figref>), and a table for managing clients is prepared at the server S<b>1</b> (see (<b>4</b>) in <figref idref="DRAWINGS">FIG. 7A</figref>).

0123<figref idref="DRAWINGS">FIG. 7B</figref> shows an example of the table at this time.

Embodiment [2]

0124The network relaying device usually has a plurality of interfaces <b>2</b>, <b>8</b> and <b>7</b> as shown in <figref idref="DRAWINGS">FIG. 5</figref>. The subnet is connected to each of the interfaces as shown in <figref idref="DRAWINGS">FIG. 8</figref>, and numerous clients are connected to each subnet. When the object network scale is large, numerous join notifying messages PKT<b>2</b> are to be transmitted to the server S<b>1</b>, which makes no problem if transmitted in a distributed manner, but which imposes processing load on the server S<b>1</b> if transmitted in a concentrated manner.

0125Therefore, it is desirable that each network relaying device <b>1</b> collects the information of the subordinate receiving clients and information of a plurality of clients is included in a single packet to be transmitted collectively.

0126For example, when the network relaying device <b>1</b> receives a reception request control message, i.e. the IGMP report for the multicast address G<b>1</b>, the timer <b>10</b> is operated.

0127The information of all of the IGMP reports addressed to the same (multicast) group address and received before the timer <b>10</b> expires, i.e. the source addresses of the receiving clients are included in the client information of a single packet as shown in a packet format of <figref idref="DRAWINGS">FIG. 9</figref> and the packet is transmitted to the server as an aggregated join notifying message PKT<b>2</b>_<b>2</b>.

0128As for a determination method of a timer value, when e.g. a server or an administrator who transmits multicast data or the like desires to manage information per minute for the purpose of charging, the timer value may be set with 60 sec.

0129Also at this time, in order to indicate that this message is different from a normal IGMP report and the above-mentioned join notifying message, a type value not specified in the type field of the IGMP report in conformity with the standard and different from the type value of the above-mentioned join notifying message is used.

0130<figref idref="DRAWINGS">FIG. 10A</figref> shows a schematic operation sequence among the clients R<b>1</b>-R<b>3</b>, the multicast router <b>1</b> and the server S<b>1</b>.

0131When the IGMP report packet is transmitted from the client R<b>1</b> (see (<b>1</b>) in <figref idref="DRAWINGS">FIG. 10A</figref>), the multicast router <b>1</b> having received the packet starts up the timer <b>10</b> (see (<b>2</b>) in <figref idref="DRAWINGS">FIG. 10A</figref>). After that, when the similar packets are received from the clients R<b>3</b> and R<b>2</b> (see (3) and (4) in <figref idref="DRAWINGS">FIG. 10A</figref>), the aggregated join notifying packet is generated (see (<b>6</b>) in <figref idref="DRAWINGS">FIG. 10A</figref>) on the condition that the timer <b>10</b> has expired (see (<b>5</b>) in <figref idref="DRAWINGS">FIG. 10A</figref>), and the packet is transmitted to the server S<b>1</b> (see (<b>7</b>) in <figref idref="DRAWINGS">FIG. 10A</figref>). The server S<b>1</b> prepares the management table of the client in response to the packet (see (<b>8</b>) in <figref idref="DRAWINGS">FIG. 10A</figref>).

0132<figref idref="DRAWINGS">FIG. 10B</figref> shows an example of an information table held by the server S<b>1</b> at this time. It is to be noted that “xx” and “tx” in <figref idref="DRAWINGS">FIG. 10B</figref> indicate a state in which a status concerning other multicasts and clients is possessed.

Embodiment [3]

0133As shown in <figref idref="DRAWINGS">FIG. 11</figref>, when an aggregated join notifying message PKT<b>2</b>_<b>21</b> is transmitted to the server S<b>1</b> from a certain network relaying device <b>1</b>_<b>2</b>, the message is received by the network relaying device <b>1</b>_<b>1</b> at the subsequent stage in the direction of server. The network relaying device <b>1</b>_<b>1</b> examines the packet PKT<b>2</b>_<b>21</b>. When the packet is the aggregated join notifying message, it is further transferred in the direction of server S<b>1</b>.

0134There is a possibility that the network relaying device <b>1</b>_<b>1</b> also has the receiving client on a local interface at this time in the same way as the network relaying device <b>1</b>_<b>2</b>.

0135If the network relaying device <b>1</b>_<b>1</b> receives the join desiring message concerning multicast data G<b>1</b> from receiving clients R<b>11</b>-R<b>14</b>, and receives the aggregated join notifying packet PKT<b>2</b>_<b>21</b> concerning the multicast data G<b>1</b> from the network relaying device <b>1</b>_<b>2</b> in a state where the timer <b>10</b> is operated for collecting information, the network relaying device <b>1</b>_<b>1</b> cancels the timer <b>10</b> in order to avoid delaying the information of the network relaying device <b>1</b>_<b>2</b> and being delivered to the server S<b>1</b>. Also, the network relaying device <b>1</b>_<b>1</b> generates a packet PKT<b>2</b>_<b>22</b> in which the information of the local receiving clients R<b>11</b>-R<b>14</b> which the network relaying device <b>1</b>_<b>1</b> has held for the multicast data G<b>1</b> is added to the received aggregated join notifying packet PKT<b>2</b>_<b>21</b>, and immediately transmits the packet PKT<b>2</b>_<b>22</b>.

0136<figref idref="DRAWINGS">FIG. 12</figref> shows a format of the aggregated join notifying packet PKT<b>2</b>_<b>22</b> transmitted from the network relaying device <b>1</b>_<b>1</b>.

0137If the IGMP report is not received from the local receiving clients R<b>11</b>-R<b>14</b> before the network relaying device <b>1</b>_<b>1</b> receives the aggregated join notifying packet PKT<b>2</b>_<b>21</b> from the network relaying device <b>1</b>_<b>2</b>, the network relaying device <b>1</b>_<b>1</b> transfers the packet to the interface in the direction of server as it is. Also, the operation is independently performed for every multicast address.

0138<figref idref="DRAWINGS">FIG. 13A</figref> shows a sequence at this time. When the network relaying device <b>1</b>_<b>1</b> receives the multicast join packet from the client R<b>11</b> (see (<b>1</b>) in <figref idref="DRAWINGS">FIG. 13A</figref>), the timer is started up (see (<b>2</b>) in <figref idref="DRAWINGS">FIG. 13A</figref>). Furthermore, when the network relaying device <b>1</b>_<b>1</b> receives the aggregated join notifying packet from the network relaying device <b>1</b>_<b>2</b> (see (<b>4</b>) in <figref idref="DRAWINGS">FIG. 13A</figref>) after receiving the multicast join packet from the client R<b>12</b> (see (<b>3</b>) in <figref idref="DRAWINGS">FIG. 13A</figref>), the timer is compulsorily ended (see (<b>5</b>) in <figref idref="DRAWINGS">FIG. 13A</figref>), and the aggregated join notifying packet PKT<b>2</b>_<b>21</b> from the network relaying device <b>1</b>_<b>2</b> to which join information from the clients R<b>11</b> and R<b>12</b> is added is immediately transmitted (see (<b>6</b>) in <figref idref="DRAWINGS">FIG. 13A</figref>). The server S<b>1</b> prepares the table upon receiving the packet (see (<b>7</b>) in <figref idref="DRAWINGS">FIG. 13A</figref>).

0139<figref idref="DRAWINGS">FIG. 13B</figref> shows an example of a table managed by the server S<b>1</b> at this time.

0000Embodiment of Server

0140As shown in <figref idref="DRAWINGS">FIG. 14</figref>, the server also has a packet discriminating portion <b>22</b> for examining/identifying a user information packet. When the packet is the user information packet, client information, a status, and a reception starting time e.g. per minute are recorded in a client management table <b>23</b>.

0141Based on the information, a management application <b>24</b> can grasp the number of users and information for calculating a charge in accordance with time. Also, when the number of users is large, the information can be made reference information for designing a server site such as an addition of a server having e.g. the same contents.

0142The example of <figref idref="DRAWINGS">FIG. 14</figref> shows a case where the server also performs a receiver management and also a transmission of the multicast data. Also, a communication process <b>25</b> and an application <b>26</b> drive a multicast transmission controller <b>27</b>, and perform a function of transmitting the multicast data DT from the contents <b>28</b> to the multicast-capable network through a transmitting/receiving interface <b>21</b>.

0000Packet Modification

0143Furthermore, for the multicast join notifying packet and the aggregated join notifying packet, as shown in <figref idref="DRAWINGS">FIGS. 9 and 12</figref>, a method of using the IGMP report received from the client for a necessary information transmission may be adopted, or the packets may be IP packets having an original format.

0144For example, as shown in <figref idref="DRAWINGS">FIG. 15</figref>, multicast data address information may be included in a payload portion for a data link layer header (e.g. MAC header) and an IP header. At this time, if there is a method of identifying that the packet is the join notifying packet or the aggregated join notifying packet, it is enough. For example, as shown in <figref idref="DRAWINGS">FIG. 16</figref>, a presently not specified value is used for the protocol ID of the IP header, and by confirming the value, the packet can be identified.

0145It is to be noted that since a well-known protocol ID is managed by IANA (Internet Assigned Numbers Authority) or ICANN (The Internet Corporation for Assigned Names and Numbers), the protocol ID can be used for an actual operation if a protocol ID for message is applied.

Embodiment [4]

0146When the receiving client stops the reception of the multicast data, a normal leave desiring message (packet), i.e. “IGMP Leave” is transmitted. In the IGMPv2, the IGMP leave message is used, and in the IGMPv3, the IGMP report (transmitted to “224.0.0.22” indicating ALL-IGMPv3 router) in which the multicast address desiring reception is made “empty” is used.

0147In order to manage the receiving client, information “which client has stopped the reception and when the reception has been stopped” is also managed by the server. It is to be noted that the leave information is basically the same as the case of join information mentioned above.

0148The example of IGMPv2 will now be described.

0149The leave message transmitted by the client which has received the multicast data is transmitted to “224.0.0.2” which is “All-Routers-Group address” (It is to be noted that the IGMP report of the IGMPv3 in which a multicast address desiring a reception is made “empty” is transmitted to “224.0.0.22”).

0150The network relaying device <b>1</b> having received the leave message examines whether or not the message is the IGMP leave message by the packet discriminating portion <b>3</b>. The identification method of the IGMP leave message is firstly to confirm {circle over (1)} IP address is “224.0.0.2”, {circle over (2)} the “protocol No.” within the IP header is “2” indicating the IGMP, and {circle over (3)} the type value of the IGMP message is “0x17”.

0151A multicast address which a receiving client desires to leave is similarly stored in the IGMP leave message. In order to inform to the server S<b>1</b> that the transmitting source client of the message desires to stop the reception of the multicast data which have been received, the packet generator <b>4</b> generates a leave notifying packet which has the address of the server S<b>1</b> as a destination address, and the transfer processor <b>5</b> transfers the packet to the server S<b>1</b> by general routing processing.

0152At this time, the type value of the IGMP message is changed and transmitted so that the server S<b>1</b> can identify the leave notifying message when receiving the packet. <figref idref="DRAWINGS">FIG. 17</figref> shows a format example of such a leave notifying packet.

0153The server S<b>1</b> having received the packet confirms the message type. When it is the leave notifying packet, the address of the client R<b>1</b> stored in the source address field is associated with the address G<b>1</b> of the multicast data transmitted by the server S<b>1</b>, and a reception time t<b>2</b> of the leave notifying packet is added to the packet, thereby enabling which multicast data reception (G<b>1</b> in this example) was stopped and when the reception was stopped (t<b>2</b> in this example) to be managed.

0154Furthermore, as for the client R<b>1</b>, the server S<b>1</b> holds information that the client R<b>1</b> started the data transmission to the multicast address G<b>1</b> at the time t<b>1</b> (i.e. “R<b>1</b>, G<b>1</b> and t<b>1</b>”), as mentioned above. By associating the information with the leave starting time t<b>2</b>, a time when the client R<b>1</b> has received data, “t<b>2</b>−t<b>1</b>” can be calculated. Namely, it is indicated that the client R<b>1</b> has received the multicast data G<b>1</b> for the time (t<b>2</b>−t<b>1</b>).

0155Thus, the sequence is the same as the case of the above-mentioned IGMP report. <figref idref="DRAWINGS">FIG. 18A</figref> shows procedures (<b>1</b>)-(<b>4</b>) in the same way as <figref idref="DRAWINGS">FIG. 7A</figref>. Also, <figref idref="DRAWINGS">FIG. 18B</figref> shows an example of a table managed by the server S<b>1</b>. As shown in <figref idref="DRAWINGS">FIG. 18B</figref>, leaving time information and time information indicating a reception time are added.

Embodiment [5]

0156As a matter of course, as shown in <figref idref="DRAWINGS">FIG. 5</figref>, the network relaying device <b>1</b> has a plurality of interfaces <b>2</b>, <b>7</b> and <b>8</b>. The subnet is connected to each of the interfaces, and numerous clients are connected to each subnet as shown in <figref idref="DRAWINGS">FIG. 8</figref>. When the object network scale is large, numerous leave notifying messages are transmitted to the server. When they are transmitted in a concentrated manner, there is a possibility that the processing load is imposed to the server Therefore, in the same way as the case of the multicast join information, such a method is considered that the leave information of the subordinate receiving clients is aggregated in each network relaying device <b>1</b>, and information of a plurality of clients is included in a single packet to be transmitted collectively.

0157For example, the network relaying device <b>1</b> operates the timer <b>10</b> when the leave message i.e. the IGMP leave message for the multicast address G<b>1</b> is firstly received.

0158The information of all of the IGMP leave messages i.e. the source addresses of the receiving clients received until the timer <b>10</b> expires are included in the client information of a single packet as shown in the packet format of <figref idref="DRAWINGS">FIG. 17</figref>, and it is transmitted to the server as the multicast aggregated leave notifying message.

0159As a determination method of the timer value, when e.g. a server, or an administrator who transmits multicast data desires to manage information per one minute for the purpose of charging, the timer value may be set with 60 sec. After the timer expires, the timer <b>10</b> is started up when the IGMP leave message is received again.

0160Also, at this time, in order to indicate that the multicast aggregated leave notifying message is different from a general IGMP leave message and the above-mentioned leave notifying message, an IGMP type value not specified by the standard and different from those of the above-mentioned join notifying message, aggregated join notifying message, and the above-mentioned leave notifying message may be used in the type field of the IGMP leave message.

0161Furthermore, since the “maximum response time field” of the IGMP message shown in <figref idref="DRAWINGS">FIG. 29</figref> is unnecessary in the processing, the method of setting “0” in the “maximum response time field” or the like can be considered. Also, as for the “check sum”, it is desirable to set a newly calculated value for the integrity of data.

0162By including, in a group address field, the multicast address G<b>1</b> of which reception the client requests to stop, when e.g. the same client receives a plurality of different multicast addresses and desires to stop the reception of the addresses, the addresses can be distinctively managed.

0163<figref idref="DRAWINGS">FIG. 19</figref> shows a format example of the aggregated leave notifying packet in which information in a case where three receiving clients R<b>1</b>-R<b>3</b> leave the multicast group G<b>1</b> almost at the same time is included.

0164Also, <figref idref="DRAWINGS">FIG. 20</figref> shows the client management table <b>23</b> when the server S<b>1</b> receives the aggregated leave notifying packet.

Embodiment [6]

0165In the same way as the IGMP report packet shown in <figref idref="DRAWINGS">FIGS. 11 and 13</figref>, the aggregated leave notifying message is also received by the subsequent network relaying device <b>1</b>_<b>1</b> in the direction of server after it is transmitted from the network relaying device <b>1</b>_<b>2</b> to the server S<b>1</b>. The network relaying device <b>1</b>_<b>1</b> examines the packet, and when it is the aggregated leave notifying message, it is further transferred in the direction of server.

0166At this time, there is a possibility that the network relaying device <b>1</b>_<b>1</b> also has receiving clients on the local interface in the same way as the network relaying device <b>1</b>_<b>2</b>.

0167It is supposed that the network relaying device <b>1</b>_<b>1</b> receives the IGMP leave message concerning the multicast G<b>1</b> from the receiving clients R<b>11</b>-R<b>14</b>, and receives the aggregated join notifying packet concerning the multicast data G<b>1</b> from the network relaying device <b>1</b>_<b>2</b> with the timer <b>10</b> being operated for aggregating information.

0168At this time, in order to avoid delaying the information of the network relaying device <b>1</b>_<b>2</b> and being delivered to the server S<b>1</b>, the timer <b>10</b> is cancelled, and the information of the local receiving clients R<b>11</b>-R<b>14</b> which the network relaying device <b>1</b>_<b>1</b> has held for the multicast data G<b>1</b> is added to the received aggregated join notifying packet to be immediately transmitted.

0169If the network relaying device <b>1</b>_<b>1</b> has not received the leave desiring message from the local receiving clients R<b>11</b>-R<b>14</b> before receiving the aggregated join notifying packet from the network relaying device <b>1</b>_<b>2</b>, the packet is transferred as it is to the interface in the direction of server. Also, this operation is performed independently for every multicast address.

0170This operation in <figref idref="DRAWINGS">FIG. 11</figref> is as follows: When the network relaying device <b>1</b>_<b>1</b> receives the leave desiring message from directly connected clients such as clients R<b>11</b>-R<b>14</b> and the timer <b>10</b> is started up for aggregating the information, the network relaying device <b>1</b>_<b>2</b> transmits the aggregated leave notifying packet in the direction of server. The timer <b>10</b> is stopped when the packet is received at the network relaying device <b>1</b>_<b>1</b>. The information of the clients R<b>11</b>-R<b>14</b> is further added to the aggregated leave notifying packet from the network relaying device <b>1</b>_<b>2</b> to be immediately transferred to the server S<b>1</b>.

0171<figref idref="DRAWINGS">FIG. 21</figref> shows a format example of the aggregated leave notifying packet transmitted from the network relaying device <b>1</b>_<b>1</b> in this case. Also, <figref idref="DRAWINGS">FIG. 22</figref> shows a state of the client management table <b>23</b> of the server S<b>1</b> when the aggregated leave notifying packet is received.

0000Client Identifying Information

0172The receiving client management in the multicast data distribution has been described above. The identification of each client by the IP address of the client is the most convenient. Namely, the IGMP report packet transmitted from a user, or the IP address of the source address of the IGMP leave packet can be used.

0173Since an IP address is assigned to a personal computer in a user's house from an ISP (Internet Service Provider) with which the user establishes an account in case of e.g. a dial-up user, the IP address is used. On the other hand, in a broadband access such as FTTH (Fiber to the Home) and ADSL (Asymmetric Digital Subscriber Line), an address is assigned to an outward interface of a compact router (SOHO (Small Office Home Office) router) within the user's house, so that it can be used.

0174The address assignment of the ISP is performed, as a matter of course for charging management, with information “to which user the IP address is assigned, which IP address is assigned, and how long (from when until when) the IP address has been assigned”. Accordingly, if this information, the IP address information of the receiving client of the multicast and the reception time (t<b>2</b>−t<b>1</b>) are combined, a charge calculation in accordance with time of the multicast contents is made possible.

0175When an SOHO router exists in the user's house, and a plurality of terminals exist within the router to receive the same multicast, if one terminal transmits a leave message, i.e. the IGMP leave message, the network relaying device (multicast router) receives the message. Then, the network relaying device transmits a “GS-Q (Group Specific Query)” message described in <figref idref="DRAWINGS">FIG. 28B</figref> in order to confirm whether or not other clients which desire to receive the multicast data exist under the subordinate of the interface.

0176Since the other client within the user's house having received the message through the SOHO router newly transmits the join desiring message in order to indicate the desire of a continuous reception, the same address, i.e. the address assigned to the SOHO router remains as the information of the receiving client, so that it does not occur that the reception becomes disabled. Namely, in the present Internet access form, and an broadband access environment such as ADSL and FTTH which is expected to be increased in the future, the operation can be performed without problems.

0000Collective Transmission of Management Information

0177The method and device for managing the receiving client by transferring the join/leave information of the client to the server based on the IGMP join/leave message transmitted by the client have been described.

0178Different from the above-mentioned description, a method of holding/accumulating the client information of the network relaying device which generally accommodates a plurality of clients, and of collectively transmitting the information to a management server upon necessity of the management information as a trigger can be considered. In this case, the processing load of the client management on the server side can be reduced.

0179In a network form accommodating an end user in a general intranet or in the Internet access, as shown in <figref idref="DRAWINGS">FIG. 1</figref>, the receiving clients R<b>1</b> and R<b>2</b> are accommodated in the LAN switches/layer <b>2</b> switches SW<b>1</b> and SW<b>2</b>, and are further connected to the upper network relaying device (multicast).

0180At this time, each network relaying device holds, by the above-mentioned means, the information of the subordinate receiving client of the respective interface and information of which multicast address the data are addressed, and from when until when the data have been received.

0181Namely, by the reception of the IGMP report, the IP address of the receiving client, the multicast address G<b>1</b> requested and the join time t<b>1</b> are recorded.

0182When the IGMP leave message is received from the same receiving client, the time t<b>2</b> of receiving the IGMP leave message is added to the client address information. If this operation is performed for every multicast at each interface, the information required for the management can be acquired.

0183The management information which each network relaying device has collected for the respective local subnet connected to its own network relaying device is not immediately transmitted in the direction of server, but the information of the receiving client held is collectively transmitted to a carrier composing a network and a management server such as an ISP which provides service upon e.g. the end of a multicast session as a trigger.

0184Alternatively, when the multicast session ends, the information concerning the multicast may be transmitted or it may be transmitted by a command from a server.

0185Thus, when the information of the receiving client is not required to be grasped at the server in real time, the information is collectively transmitted at a certain time after the end of the multicast data transmission, thereby making the operation and the management on the server side simple. As a matter of course, such information can be used for a purpose of a charge calculation in accordance with time per user and the examination of the number of receiving users (audience rating).

0186The above-mentioned local user information holding function may be provided in the LAN switches SW<b>1</b> and SW<b>2</b> referring to <figref idref="DRAWINGS">FIG. 1</figref>, or may be provided in the multicast routers <b>1</b>_<b>1</b> and <b>1</b>_<b>2</b>.

0000Setting of IGMP Type Value

0187One embodiment of the present invention has been described in the above. As for the IGMP type value identifying the join notifying packet, the aggregated join notifying packet, the leave notifying packet and the aggregated leave notifying packet respectively, a value not used in the standard shown in <figref idref="DRAWINGS">FIG. 29</figref> may be used as follows for example: <ul id="ul0001" list-style="none"><li id="ul0001-0001" num="0000"><ul id="ul0002" list-style="none"><li id="ul0002-0001" num="0188"> Join notifying packet: 0x31 </li><li id="ul0002-0002" num="0189"> Aggregated join notifying packet: 0x32 </li><li id="ul0002-0003" num="0190"> Leave notifying packet: 0x33 </li><li id="ul0002-0004" num="0191"> Aggregated leave notifying packet: 0x34 </li></ul></li></ul>

0192Also, the present invention can be applied to the IPv6 multicast. A main difference between the IPv6 multicast and the IPv4 multicast will now be described.

0000(1) The multicast MAC address is generated by mapping lower 32 bits of 128-bit IPv6 address to 48-bit MAC address 33: 33: xx: xx: xx: xx.

0193(2) As for the MLD corresponding to the IGMP of the IPv4, an ICMPv6 message (specified in RFC 2463) shown in <figref idref="DRAWINGS">FIG. 23</figref> is used. The type value “130” of the ICMPv6 is a query message of the IGMP, “131” is the IGMP report, “132” is a Done (same as the IGMP leave). (see <figref idref="DRAWINGS">FIG. 24</figref>).

0000(3) An IPv6 multicast address is identified by an FF0X: 0: 0: 0: 0: 0: 0: 0 (head 8 bits is 11111111).

0194Namely, by taking steps in consideration of the above-mentioned difference, the method of the present invention can be applied to the multicast in the IPv6 environment.

Embodiment [7]

0195By the above-mentioned method, the management of the receiving client is made possible. However, when the reception can not be performed due to an OS fault of the client and a hung-up of a media reproduction application or the like, the leave message, i.e. the IGMP leave message can not be transmitted. In this case, an accurate leave management of the client can not be performed. The embodiment for solving this problem will now be described.

0196Generally, the multicast router periodically transmits (default 125 sec.) the IGMP query message to “224.0.0.1” which is an “All-Systems Group” address, so that the client during reception receives the query message and transmits the IGMP report.

0197The above-mentioned IGMP report is voluntarily transmitted by the client upon determining the reception of the multicast data in many cases, and is transmitted in response to the IGMP query message periodically transmitted from the network relaying device in a few cases. However, by the IGMP report transmitted in response to the IGMP query message periodically transmitted, client's living is periodically monitored. This is an operation specified by a general standard.

0198If a certain receiving client can not receive due to an OS fault, hung-up of a media reproduction application or the like, the receiving client does not respond to the IGMP query message periodically transmitted from the network relay, and does not transmit the IGMP report.

0199The network relaying device regards the receiving client which does not respond to the IGMP query message as a reception stop, fills in a leaving time in the client management table of its own device, fills in “reception stop/error” in the status, generates the leave notifying message for client to be transmitted to the server S<b>1</b>, thereby enabling a client management corresponding to a fault to be performed.

0200In this case, as mentioned above, the information is written by the IGMP report which the client has voluntarily transmitted upon determining the reception of the multicast data in many cases. However, operation of confirming living is performed by the reception of the IGMP report in response to the IGMP query message periodically transmitted.

0201When the client management is performed per minute accurately, the timer of the query message periodically transmitted may be set with the default 125 sec. to 60 sec. to be operated. In a case where the timer is set with ten minutes, the default value is sufficient.

0202If the client starts the reception, the starting time and the leaving time afterward may be reflected in the management table.

0203Also, when the network relaying device holds the management table, the information may be reflected in the management table of the network relay, and may be collectively transmitted to the server with a certain trigger.

0204Alternatively, as another means, there is a method of confirming living of the receiving client by Ping.

0205<figref idref="DRAWINGS">FIGS. 25A and 25B</figref> show the summary.

0206Namely, <figref idref="DRAWINGS">FIG. 25A</figref> is divided into 3 phases PS<b>1</b>-PS<b>3</b>. In the phase PS<b>1</b>, for a group join, the client R<b>1</b> transmits the IGMP report (see (<b>1</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>), the multicast router <b>1</b> generates the join notifying packet (see (<b>2</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>) and transmits the packet to the server S<b>1</b> through the multicast-capable network MNW (see (<b>3</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>). The server S<b>1</b> prepares the client management table <b>23</b> from the packet (see (<b>4</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>).

0207The phase PS<b>2</b> indicates a fault detection phase. When the multicast router <b>1</b> periodically transmits the query message (see (<b>5</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>), the client R<b>1</b> can not transmit the IGMP report to the query if the client R<b>1</b> is supposed to be in a fault state (see (<b>6</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>).

0208Thus, the multicast router <b>1</b> detects that the client R<b>1</b> is in the fault state, and fills in the state in its own client management table (see (<b>7</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>). Then, the multicast router <b>1</b> generates the leave notifying packet concerning the client R<b>1</b> (see (<b>8</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>) to be transmitted to the server S<b>1</b> (see (<b>9</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>). The server S<b>1</b> updates the management information concerning the client R<b>1</b> (see (<b>10</b>) <figref idref="DRAWINGS">FIG. 25A</figref>).

0209The phase PS<b>3</b> indicates a reception restart phase. The client R<b>1</b> transmits the IGMP report packet after the fault recovery (see (<b>11</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>). In response to the packet, the multicast router <b>1</b> generates the join notifying packet (see (<b>12</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>) to be transmitted to the server S<b>1</b> (see (<b>13</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>). The server S<b>1</b> updates the client management table (see (<b>14</b>) in <figref idref="DRAWINGS">FIG. 25A</figref>).

0210As a result, the information concerning the client R<b>1</b> within the client management table <b>23</b> in the server S<b>1</b> includes the information of the time t<b>3</b> at which the reception is started, which is as shown in <figref idref="DRAWINGS">FIG. 25B</figref>.

0000Charging Service in Accordance with Time

0211The receiving client management system in the server has been described in the above. New service can be provided by using this mechanism.

0212As mentioned above, in the distribution of data such as voices and images by the multicast, information of what receiving clients exist, how many receiving clients totally exist at a certain time, how many receiving clients have totally received the multicast data could not be grasped.

0213This is because the mechanism of the multicast does not have such a concept originally, and a UDP packet is used for the data transmission. Such a background prevents the multicast having numerous advantages at the same time from being widespread.

0214Therefore, the network relaying method and device provided by the present invention enable a user management service in the multicast data distribution and charging service in accordance with time for a content usage per receiving client which has been impossible in the prior art.

0215In the present service form of a chargeable content distribution (by unicast), following two forms are mainly provided. One is a form of purchasing a right of a service usage by a fixed fee, and the other is a pay-as-you-go form. Also, these are all unicast, and a distribution form is on-demand or real-time broadcasting.

0216At this time, as for the real-time broadcasting especially, it has been already described that the multicast is effective in terms of cutting down new plant and equipment. However, if services are provided since the multicast can not manage the user itself, there has been no other method except purchasing the right of a content access. Also, as for the multicast, not an on-demand distribution but a real-time content distribution is performed. Therefore, it has been necessary to make access without fail when the reception is possible.

0217However, if the receiving client management is performed by the present invention, charging in accordance with the time when the client has accessed can be performed, so that it becomes very easy for a carrier to provide services.

0218Also, if a user can not access content during a broadcasting time, no charge is required. When a program distributed by accessing is used for a fixed time during the content broadcasting, the charge for that time is required. Thus, services with merits for the user can be provided.

0219<figref idref="DRAWINGS">FIGS. 26 and 27</figref> show examples of the service provision form.

0220In the example of <figref idref="DRAWINGS">FIG. 26</figref>, a server site ST such as an ISP/NSP is provided with a client management/authentication server S<b>11</b>, a charging server SR, and a multicast content distribution/client management server S<b>13</b>. The server S<b>13</b> performs client management/authentication to the multicast join/leave information of the client <b>20</b> or the like at the server S<b>11</b>, whereby the server S<b>13</b> collects the multicast join/leave information through a client accommodating switch SW and multicast-capable routers <b>1</b>_<b>1</b>-<b>1</b>_<b>5</b> or the like.

0221As shown in <figref idref="DRAWINGS">FIG. 27</figref>, an Internet connection/content distribution carrier <b>30</b> provides general Internet connection service <b>32</b> to the client <b>20</b> {circle over (2)}, and charges the client (user) for the Internet connection fee {circle over (3)}. Also, when the Internet connection/content distribution carrier <b>30</b> provides the content distribution service to the client <b>20</b> {circle over (1)} as a further added value service, multicast content distribution service DS by the pay-as-you-go system for the multicast usage can be provided.

0222Namely, the content distribution service DS charging in accordance with the program reception time of the client <b>20</b> can be provided, and the Internet connection/content distribution carrier <b>30</b> charges the program reception fee for the client <b>20</b> in accordance with the program reception time of the client <b>20</b>.

0223As mentioned above, the present invention enables a receiving user management in the data distribution, which has been difficult, by the multicast in the same way as the present unicast communication, thereby reliability of the multicast communication to be provided. The present invention can contribute to the spread of the multicast-capable network device, the end system, and the multicast-capable application.

0224Also, the present invention enables the number of receiving clients at a certain time and the number of receivers in total to be grasped and managed by providing a user management function in the multicast environment, and contributes to the spread of the content distribution service and the Internet real-time broadcasting by the multicast.

0225Furthermore, the present invention can provide charging service in accordance with time in the multicast content distribution which has been impossible, by providing a user management even in the multicast distribution, and contributes to spreads of a content distribution business, and a broadband network for the purpose of a content reception.