KR20210044730A - Apparatus and Method for forwarding a Packet in Content Centric Network - Google Patents

Abstract

The present invention relates to an apparatus for forwarding packets in a content centric network to efficiently process transit packets and a method thereof. According to the present invention, a node constituting the content centric network comprises: a plurality of interfaces configured to receive and transmit interest packets and data packets to a plurality of remote nodes; a pending interest table (PIT) storing information on a local interest packet, of which the source network or destination network corresponds to a network in which a corresponding node is located, among unanswered interest packets; a tag table storing information on a transit interest packet, of which the source network or destination network is not a network in which the corresponding node is located, among the unanswered interest packets by a group unit related to the source network; a forwarding information base (FIB) for forwarding the interest packet to the destination node; and a processor selecting the PIT or the tag table according to whether the packet is a local packet or a transit packet and performing packet forwarding on the basis of the selected table when forwarding a packet.

Description

Apparatus and Method for forwarding a Packet in Content Centric Network

The present disclosure relates to an apparatus and method for packet forwarding in a content-oriented network. In more detail, the present invention relates to a technology for efficiently processing by newly defining a tag in a data packet or an interest packet in a content-oriented network.

Currently, the scale of data services on the Internet is increasing rapidly. Meanwhile, the current Internet operates in a manner using an IP address. Data requesters, i.e. consumers, can request and receive data using only IP addresses. In such an Internet network, when there is at least one user on a network that basically requests the same data, data must be repeatedly transmitted as many as the number of users, and thus there is an inefficiency in terms of repeated data transmission.

In order to solve this problem, research and development on a network structure centered on content, not an IP-based network structure, are being actively conducted in recent years. The content-centric network in the present disclosure includes an information-based/centric network, a Content Centric Network (CCN), an Information Centric Network (ICN), and a Named Data Network (NDN).

Networks that support content-oriented technology introduce the concept of a content provider that generates data on behalf of a sender and a receiver, and a content consumer that receives and consumes the data, so that the network performs routing using the content name instead of the IP address. can do. Routers can store specific data if necessary and distribute the data to nearby new consumers who request the data.

In a content-oriented network, when an interest packet is transmitted with a data name requested by a content consumer in a header, the corresponding data can be cached through data name-based forwarding, or the owning node can respond as a data packet. have. The data packet can be delivered to the data requester along the path on which the interest packet came. On the other hand, the nodes constituting the content-centric network include CS (Content Store) for caching data packets passing through the corresponding node, FIB (Forwarding Information Base) for forwarding interest packets based on the data name, and unanswered interest packets. It may include a Pending Interest Table (PIT) that stores and manages the information of the company.

In this case, since the PIT needs to store information for each interest packet during the round-trip time (RTT) of the interest packet, the size of the PIT increases in proportion to the forwarding capacity of the node. This requires a PIT table size, which is almost impossible to implement with the current technology, which hinders the introduction of a content-oriented network.

In order to solve the problems of the prior art described above, an object of the present disclosure is to propose a tag table that can be used with PIT or can be used in place of PIT as a method of identifying packets in a content-oriented network. have.

In addition, an object of the present disclosure is to reduce the size of a PIT in a content-oriented network to enable the implementation of a core node of several Tbps or more.

In addition, an object of the present disclosure is to efficiently process a transit packet other than a source network and a destination network of a packet received by a node's network in a content-oriented network.

In addition, an object of the present disclosure is to implement symmetric forwarding of interest packets of data packets without using PIT in a content-oriented network.

Other objects and advantages of the present disclosure may be understood by the following description, and will be more clearly understood by examples of the present disclosure. In addition, it will be easily understood that the objects and advantages of the present disclosure can be realized by the means shown in the claims and combinations thereof.

According to an embodiment of the present disclosure for achieving the above object, in a node constituting a content-oriented network, a plurality of interfaces configured to receive and transmit interest packets and data packets for a plurality of remote nodes, Among the interest packets, the source network or the destination network is a PIT (Pending Interest Table) that stores the information of the local interest packet corresponding to the network where the node is located, and the source network and the destination network are among the interest packets that have not been answered. A tag table that stores information of a transit interest packet other than the network in which the node is located in a group unit associated with the source network, and FIB (Forwarding Information Base) for forwarding the interest packet to the destination node. , During packet forwarding, a processor for selecting the PIT or the tag table according to whether the packet is a local packet or a transit packet, and performing packet forwarding based on the selected table may be included. .

On the other hand, if the destination network name of the interest packet received from the interface connected to the external network is not the same as the network name in which the node is located, or a tag is included in the interest packet received by the interface connected to the internal network, the processor The received interest packet can be classified as a transit interest packet.

Meanwhile, when an interest packet received from an interface connected to the internal network does not include a tag, or an interest packet having the same destination network name as the network name in which the node is located is received from the interface connected to the external network, the processor The received interest packet can be classified as a local interest packet.

Meanwhile, when a tag is included in the received transit interest, the processor may search the tag table for an entry matching the tag of the received packet and interface-related information of the received packet.

Meanwhile, when the tag table search is successful, the processor may change the tag of the received packet to the output tag of the entry.

Meanwhile, when the search fails in the tag table, the processor may change the tag of the received packet to the output tag by allocating an output tag.

Meanwhile, when a tag is not included in the received transit interest packet, the processor may assign a tag to the received packet based on the interface-related information on which the packet is received.

Meanwhile, in the packet The assigned tag and the received interface-related information may be added to the tag table.

Meanwhile, when the received local interest packet includes a tag, the tag may be removed from the header and the tag and the reception interface information may be stored in the PIT entry of the interest.

Meanwhile, each entry of the PIT may include a tag of a received interest packet and interface information of the received interest packet.

Meanwhile, the FIB may include a network name FIB (network name FIB) and a data name prefix FIB (data name prefix FIB).

Meanwhile, when the received interest packet is a transit interest packet, the processor may forward the packet through the network name (FIB) forwarding information base (FIB) lookup.

Meanwhile, when the received interest packet is a local interest packet, the processor may perform packet forwarding through the Data Name Prefix FIB lookup.

Meanwhile, the tag table includes a tag of the received packet and an output tag for forwarding of the received packet, and a flag for best reception time information of the received packet or aging management of the received packet It may further include at least any one of.

Meanwhile, when a data packet is received through the interface and the data packet includes a tag, the processor may search the tag table for an entry having an output tag matching the tag.

Meanwhile, the processor may change the tag to the interest packet tag of the matched entry, and forward the received packet to the interface of the matched entry.

On the other hand, if the received packet is a data packet that does not contain an output tag,

The processor searches for an interest packet matching the data name of the data packet in a PIT (Pending Interest Table), and if the search is successful, adds the tag of the interest packet to the data packet, and adds the interest packet. Packets can be forwarded based on the received interface information.

To achieve the above object, a packet forwarding method performed by a node constituting a content-oriented network according to an embodiment of the present disclosure includes the steps of receiving an interest packet in a content-oriented network, wherein a tag is added to the interest packet. Reviewing whether it is included, and forwarding the interest packet, wherein the interest packet is received by an interface connected to an external network, and the destination network name of the interest packet is the network name in which the node is located. If not identical or when the interest packet is received by an interface connected to an internal network and a tag is included in the interest packet, the received interest packet may be classified as a transit interest packet.

Meanwhile, when the interest packet is received from an interface connected to the internal network and does not include a tag, or the destination network name is the same as the network name where the node is located in the interface connected to the external network, the received Interest packets can be classified into local interest packets.

A packet forwarding method performed by a node constituting a content-oriented network according to an embodiment of the present disclosure for achieving the above object includes receiving a data packet of a content-oriented network, and a tag is included in the received data packet. Checking whether or not, and forwarding the data packet, and if a tag is included in the received data packet, searching for an entry including an output tag matching the tag in a tag table, and the entry Forwarding may be performed based on the interface information of the interest packet included in.

According to an embodiment of the present disclosure, it is possible to efficiently process transit traffic in a content-oriented network.

According to an embodiment of the present disclosure, it is possible to implement a core node of several Tbps or more by reducing the PIT size in a content-oriented network.

According to an embodiment of the present disclosure, it is possible to implement symmetric forwarding in which a forwarding path of a data packet becomes a forwarding path of an interest packet in a reverse direction without using a PIT.

The effects obtainable in the present disclosure are not limited to the above-mentioned effects, and other effects not mentioned can be clearly understood by those of ordinary skill in the art from the following description. will be.

1 is a diagram illustrating a Map-n-Encap packet forwarding method in a content-oriented network to which the present disclosure can be applied.
FIG. 2 is a diagram for describing an interest packet forwarding method based on a tag table in a content-oriented network according to an embodiment of the present disclosure.
3 is a diagram for describing a data packet forwarding method based on a tag table in a content-oriented network according to an embodiment of the present disclosure.
FIG. 4 is a diagram illustrating a packet forwarding method through a plurality of core networks between content consumers and producers located in different user networks in a content-centric network according to an embodiment of the present disclosure.
5 is a diagram illustrating a packet exchange method between a content consumer and a producer located in an adjacent core network in a content-centric network according to an embodiment of the present disclosure.
6 is a diagram illustrating an example of a node constituting a content-oriented network applicable to the present disclosure.
7 is a diagram illustrating another example of a node constituting a content-oriented network applicable to the present disclosure.
8 is a diagram illustrating another example of a node constituting a content-oriented network applicable to the present disclosure.
9 is a diagram illustrating an example of a tag table in a content-oriented network applicable to the present disclosure.
10 is a diagram illustrating another example of a tag table in a content-oriented network applicable to the present disclosure.
11 is a diagram for describing an example of a PIT (Pending Interest Table) supporting a tag in a content-oriented network applicable to the present disclosure.
12 is a diagram illustrating a packet format in a content-oriented network according to an embodiment of the present disclosure.
13 is a diagram for describing an interest packet forwarding method performed by a node constituting a content-oriented network according to an embodiment of the present disclosure.
14 is a diagram illustrating a data packet forwarding method performed by a node constituting a content-oriented network according to an embodiment of the present disclosure.
15 is a diagram for explaining an interest packet forwarding method when a network name is different from a router region in a PE node of a content-oriented network according to an embodiment of the present disclosure.
16 is a diagram for explaining an interest packet forwarding method when a network name is different from a router region in a non-PE node of a content-oriented network according to an embodiment of the present disclosure
FIG. 17 is a diagram for explaining a packet forwarding method of an interest packet whose network name is the same as a router region and does not include a network name in a content-oriented network according to an embodiment of the present disclosure.
18 is a diagram for describing a data packet forwarding method in a content-oriented network according to an embodiment of the present disclosure.
19 is a diagram illustrating another example of a node constituting a content-oriented network according to the present disclosure.

Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art may easily implement the exemplary embodiments. However, the present disclosure may be implemented in various different forms and is not limited to the embodiments described herein.

In describing an embodiment of the present disclosure, when it is determined that a detailed description of a known configuration or function may obscure the subject matter of the present disclosure, a detailed description thereof will be omitted. In addition, parts not related to the description of the present disclosure in the drawings are omitted, and similar reference numerals are attached to similar parts.

In the present disclosure, components that are distinguished from each other are intended to clearly describe each feature, and do not necessarily mean that the components are separated. That is, a plurality of components may be integrated into one hardware or software unit, or one component may be distributed to form a plurality of hardware or software units. Therefore, even if not stated otherwise, such integrated or distributed embodiments are also included in the scope of the present disclosure.

In the present disclosure, components described in various embodiments do not necessarily mean essential components, and some may be optional components. Accordingly, an embodiment comprising a subset of components described in the embodiment is also included in the scope of the present disclosure. In addition, embodiments including other components in addition to the components described in the various embodiments are included in the scope of the present disclosure.

In the present disclosure, components described in various embodiments do not necessarily mean essential components, and some may be optional components. Accordingly, an embodiment comprising a subset of components described in the embodiment is also included in the scope of the present disclosure. In addition, embodiments including other components in addition to the components described in the various embodiments are included in the scope of the present disclosure.

In the present disclosure, terms such as first and second are used only for the purpose of distinguishing one component from other components, and do not limit the order or importance of the components unless otherwise noted. Accordingly, within the scope of the present disclosure, a first component in one embodiment may be referred to as a second component in another embodiment, and similarly, a second component in one embodiment is a first component in another embodiment. It can also be called.

When a component of the present disclosure is referred to as being “connected” or “connected” to another component, it may be directly connected or connected to the other component, but other components exist in the middle. It should be understood that it may be possible. On the other hand, when a component is referred to as being “directly connected” or “directly connected” to another component, it should be understood that there is no other component in the middle.

Meanwhile, in describing embodiments of the present disclosure, the content-centric network may include an information-based/centric network, a content centric network (CCN), an information centric network (ICN), and a named data network (NDN).

Meanwhile, in describing embodiments of the present disclosure, a node constituting a content-oriented network may include a Provider Edge (PE)/Non-PE node.

Meanwhile, in describing embodiments of the present disclosure, a local packet may be used to include a local interest packet and a local interest, and a local interest packet and a local interest may be mixed.

Meanwhile, in describing embodiments of the present disclosure, a transit packet may be used to mean including a transit interest packet and a transit interest packet, and a transit interest packet and a transit interest packet may be used interchangeably.

Meanwhile, in describing embodiments of the present disclosure, the packet forwarding unit may include a processor and/or a PIT/tag table selection logic.

Meanwhile, in the embodiments of the present disclosure, a transit traffic or a transit packet refers to a packet in which a source network and a destination network of a packet received from a node of a specific core network are not the corresponding network. Other packets are defined as local packets.

Meanwhile, in describing embodiments of the present disclosure, a tag is a group identifier capable of distinguishing a group of transit interest packets using the same output interface of the same edge node when the transit interest is transmitted from the source network to the core network. Means. In addition, a tag table may mean a data table including a group identifier of the transit interest packet and information on the interest packet. At this time, the tag and the tag table are arbitrary names. For example, tags can have other names such as identifiers, IDs, indexes, etc., and tag tables are auxiliary tables, identifier tables, ID tables, index tables, second PITs (pending interest tables), auxiliary PITs, and transformations. It may be called another name such as PIT, but is not limited thereto.

Hereinafter, detailed embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings.

One of the proposed schemes to solve the rapid increase in BGP (Border Gateway Protocol) routing entries in the core network in the current Internet is the Map-n-Encap scheme, which is represented by LISP (Locator ID Separation Protocol). In the IP-based Internet, the Map-n-Encap method performs tunneling by adding an IP header to the destination network's representative IP when entering the core network when the packet destination is a different network. In the core network, the destination network IP Forwarding is based on. On the other hand, when it arrives at the destination network, it forwards to the destination IP after removing the tunnel header. Through this, in the core network, routing entries are greatly reduced by allowing only routing entries for the representative IP address of the network to secure routing scalability.

)이고, 이를 계층적 이름 구조를 사용해서 애그리게이션(aggregation)을 한다고 가정하더라도 DNS(Domain Name System) 도메인 수가 O(

) 만큼 필요하다는 예측이 있어, O(

)인 IP의 코어망 라우팅 엔트리 수에 비해 콘텐츠 중심 네트워크 코어망은 IP 망보다 1000배이상 큰 라우팅 엔트리를 처리할 수 있어야 한다. 이는, 콘텐츠 중심 네트워크의 도입에 한계로서 지적된 바 있다. Content-oriented networks including NDN/CCN/ICN are characterized by forwarding packets based on the content name. In this case, for example, the number of contents on the Internet is O(

), and even assuming that aggregation is performed using a hierarchical name structure, the number of Domain Name System (DNS) domains is O(

There is a prediction that you need as much as O(

), the content-oriented network core network must be able to handle routing entries that are 1000 times larger than the IP network compared to the number of routing entries in the core network of IP. This has been pointed out as a limitation in the introduction of a content-oriented network.

To solve this problem, a technology that applies the Map-n-Encap method in a content-oriented network has been proposed. At this time, each network has a network name, and the router sets the network name to which it belongs as a region. The data owner, the producer, registers the network name to which his or her data belongs to the NRS (Name Resolution Server), and the data requester, the consumer, queries the NRS for the name of the network where the desired data is located, and then A Forwarding Hint field that stores the network name along with the data name in the packet is inserted and transmitted. If the network name of the interest packet is different from the router's region, the router forwards it to the network name. If it is the same as the region, the router forwards it to the data name as it has arrived at the destination network. In this case, in the core network of the content-oriented network, routing entries are made based on the network name rather than the data name, so that routing scalability can be secured.

However, when implementing such a core network node, it was pointed out as a limitation that it was difficult to implement a PIT (Pending Interest Table) of an actual required size. More specifically, the PIT creates an entry for each interest packet waiting for a data packet response. This entry must be maintained during the RTT (Round Trip Time) of the packet, so that the table size increases as the node's switching capacity increases. do. For this reason, it has been estimated that PIT at a core node of several Tbps or more is almost impossible to implement with the current memory technology.

On the other hand, when a data packet, which is a response of an interest packet, returns, the PIT transmits the data packet to the input interface where the interest packet was received, in the reverse direction along the forwarding path of the interest packet. Data packets can be returned. To this end, the PIT includes a data name requested by an interest packet for each interest packet, an input interface list on which the interest packet is received, a lifetime, a nonce list for loop detection, and the like. At this time, when a new interest packet requesting the data name stored in the PIT is received, it is integrated into the matching entry that has already existed in the PIT. In addition, based on the PIT, through an interest packet integration function that updates the input interface list and the nonce list, one data packet can be multicasted to multiple data consumers, Efficiency in repetitive transmission can be achieved. In addition, a function of detecting a loop of an interest packet through a nonce may be implemented through the PIT.

However, the effect of multicast through the integration of interest packets is mainly effective at the edge close to the data consumer and not large in the core network. In addition, since the interest loop can be solved by adding a HopLimit such as TTL (Time To Live) of IP to the interest packet in addition to Nonce, In the network, PIT replacement technology is also being studied, which provides only the function of allowing data packets to return along the path of the interest packet.

Accordingly, in the present disclosure, a transit packet that occupies most of the traffic of the core network in a node located in the core network of the content-oriented network, that is, the network to which the node belongs is not the network in which the packet was generated nor the destination network. As an auxiliary table that can be used instead of PIT, we propose a tag table. According to the present disclosure, an existing PIT can be used for packets that are generated in a network to which a node of a content-oriented network belongs or a network to which a node of a content-oriented network belongs is a destination network, so that a PIT and an auxiliary table (tag Table, Tag Table) to implement core network nodes of several Tbps or more.

1 is a diagram illustrating a Map-n-Encap packet forwarding method in a content-oriented network to which the present disclosure can be applied.

As an embodiment, all networks may have a network name, and nodes constituting each network may know the network name in which desired data is located through a Name Resolution Server (NRS). As mentioned above, in the present disclosure, the node may include a provider edge (PE) node and a non-PE node, and the PE node is at the edge of an ISP network (Internet Service Provider Network) constituting the core network. As a located node, it may be a node that is connected to another operator network or a customer network. Nodes in the customer network and internal nodes in the operator network can be classified into non-PE nodes.

As an embodiment, in the present disclosure, the operator network may mean a network in which the source and destination of traffic support the routing of transit traffic rather than its own network, and the customer network is a network that does not support transit traffic. Can mean

FIG. 2 is a diagram for describing an interest packet forwarding method based on a tag table in a content-oriented network according to an embodiment of the present disclosure.

As an embodiment, there may be a customer network CN-1 and CN-2, which are content-oriented networks, and a service provider network PN-1, and CN-1 and CN-2 may be connected to the service provider network PN-1. It is assumed that a data requester of CN-1 transmits an interest packet requesting data of a data owner located in CN-2. When transmitting an interest packet, it may be necessary to convert the header of the interest packet. The data requestor may transmit the data owner's network name by including NN (network name) = CN-2 in the interest packet. A tag may be included in this interest packet, but may not be included. If not included, the tag T1 can be added at the PE1 node, the Ingress Edge node of the PN-1. Here, the tag T1 is newly allocated and may be expressed as an output tag (out_tag), and may be null because the current tag, that is, the original tag (in_tag) of the interest packet did not exist. PE1 can add (or update) {in_tag, in_if, out_tag} to the tag table after receiving the interest packet, where in_if refers to the interface that received the interest packet from PE1, that is, the input interface. Can be.

As an embodiment, in order to forward the interest packet to the network (CN-2) to which the data owner belongs after storing {null, in_if, out_tag=T1} in the tag table, a non-PE node located in PN-1 Can forward to R2. In this case, the current tag (in_tag) of the interest packet is T1, and R2 may change (swap) the current tag (in_tag) of the interest packet to T2. As a PE node, which is an egress node, PE2 to which an interest packet is forwarded may change (swap) the current tag (in_tag) to T3 again. After that, PE2 can forward interest packets to R3, which is the edge node and non-PE node of the CN-2 network. R3 can recognize that the destination network is the network to which it belongs, since the network name to which the data requester of the input interest packet belongs is CN-2. Accordingly, R3 can forward the tag to the data owner (producer, producer) after removing the tag and storing that the tag is T3 (tag=T3) in the PIT.

Meanwhile, R1, R2, R3, PE1, and PE2, which are nodes constituting the content-centric network mentioned above, will be described in more detail with reference to FIGS. 6 to 8 and 19. A method of processing the rest packet and the data packet will be described in more detail with reference to FIGS. 13 to 19. In addition, the PIT, tag table, and packet mentioned above may be the same as those described in FIGS. 9 to 12.

3 is a diagram for describing a data packet forwarding method based on a tag table in a content-oriented network according to an embodiment of the present disclosure.

As an embodiment, there are customer networks CN-1 and CN-2, which are content-oriented networks, and a service provider network PN-1, and CN-1 and CN-2 may be connected to the service provider network PN-1.

As an embodiment, FIG. 3 may show a process in which the data owner transmits the data packet as a response to the data requester after the interest packet transmitted from the data requester is received as shown in FIG. 2. In the following, it is assumed that the process of FIG. 2 has been given priority for clarity of description. In other words, it may be a process in which the data owner (producer, producer) of CN-2 sends a data packet to the data requester (consumer, consumer) of CN-1.

As an embodiment, only the data name may be included in a header of a data packet transmitted by a data owner located in CN-2 to a data requestor located in CN-1. That is, the tag may not be included in the data packet. In the PIT of R3, which is an edge node located in CN-2, an output tag (tag=T3) may be stored for the corresponding data name. In this case, the PIT may be searched based on the data name of the header, and an output tag (tag=T3) matching the data name in the PIT may be added to the data packet. Thereafter, the data packet may be forwarded to the associated interface (interrest packet interface) of the matching tag and transmitted to PE2 of the PN-1.

In addition, in the PE2 node, the tag T3 may be searched for the output tag field ({out_tag} field) of the tag table. In relation to this, the tag of the data packet does not mean an output tag by itself, but may be used as information for searching an output tag field of a tag table. For example, if a tag is searched and the matching entry is found from the tag table, the {in_tag, in_if} of the corresponding entry is obtained and the tag (T3) of the data packet is changed to the matched interest group tag (in_tag) T2. (swap) and forwarding to the matching interface (in_if). Similarly, in the R2 node to which the interest packet is forwarded by PE2, the current tag of the data tag, that is, the tag T2, can be searched for the output tag field ({out_tag} field) of the tag table. When a matching entry is found by searching for a tag, the tag (T2) of the data packet is changed (swap) to the matched interest group tag (in_tag) T1 by acquiring the {in_tag, in_if} of the corresponding entry. It can be forwarded to the interface (in_if). Here, based on the tag table of the PE1 node to which the data packet was forwarded, if an entry matching the tag T1 is searched, it can be seen that the interest packet did not contain a tag (tag=null). have. In other words, if you search the out_tag field of the tag table with T1, {null, in_if} can be returned. Since in_tag = null, the data packet is forwarded to the interface (in_if) where the interest packet was input after removing the tag of the data packet. can do.

Meanwhile, R1, R2, R3, PE1, and PE2, which are nodes constituting the content-centric network mentioned above, will be described in more detail with reference to FIGS. 6 to 8 and 19. A method of processing the rest packet and the data packet will be described in more detail with reference to FIGS. 13 to 19. In addition, the PIT, tag table, and packet mentioned above may be the same as those described in FIGS. 9 to 12.

4 is a diagram for explaining an example of using a tag table in a packet forwarding method based on a plurality of core networks in a content-oriented network according to an embodiment of the present disclosure. In particular, FIG. 4 shows an example of using a tag table according to a transmission path of an interest packet and a data packet according to FIGS. 2 and 3 described above.

Referring to FIG. 4, for example, there may be a content-oriented network such as a service provider network NetB and NetC, and a customer network NetA and NetD. In this case, FIG. 4 shows a case where a data owner and a data requestor exist in a customer network (NetD, NetA).

Specifically, FIG. 4 is a diagram illustrating a packet exchange process when a data requestor belonging to a customer network NetA requests and receives data of a data owner belonging to a customer network NetD. In this case, the tag of the interest packet may be changed to 20, 30, 40, 50 through NetB and NetC. Conversely, the tag of the data packet can be changed to 50, 40, 30, 20 through NetB and NetC. This process may be the same as the process with reference to FIGS. 2 and 3, and may be described in more detail below with reference to other drawings.

5 is a diagram illustrating a packet exchange method between adjacent core networks in a content-oriented network according to an embodiment of the present disclosure.

Referring to FIG. 5, for example, there may be content-oriented networks, such as operator networks PN-1 and PN-2, and customer networks CN-1 and CN-2. In this case, FIG. 5 shows a case in which a data owner (producer) and a data requestor (consumer) exist in neighboring operator networks (PN-1 and PN-2), respectively.

Specifically, FIG. 5 is a diagram for explaining a packet exchange process when a data requester belonging to an operator network PN-1 requests and receives data of a data owner belonging to an adjacent operator network PN-2 . In this case, for the interest packet, the PN-1 network is the source network and the PN-2 is the destination network, so this interest packet can be classified as a local interest packet in both PN-1 and PN-2. Therefore, it is possible not to include the tag in the packet.

6 is a diagram illustrating an example of a node constituting a content-oriented network applicable to the present disclosure.

As an embodiment, the node of FIG. 6 may be a provider edge (PE) node and a non-PE node, and a processor 601, a content store (CS) 602, and a pending interest table (PIT) 603 ), Forwarding Information Base (FIB) 604, tag table 605, PIT/Tag Table Selection Logic 606, and one or more interfaces 607 may be included. . However, the configuration of FIG. 6 is merely illustrated as an example to describe the present disclosure. Accordingly, it is obvious that some or all of the configurations of FIG. 6 may be integrated and implemented in one hardware or software. For example, the PIT/tag table selection logic 606 may be implemented as a component in the processor 601. This can be equally applied to FIGS. 7 to 8 to be described later.

As an embodiment, the processor 601 may perform overall control of other components 602, 603, 604, 605, 606, 607 of the node, and generate interest packets and data packets to be transmitted and received through the interface. I can. As an embodiment, the processor may be interlocked with the PIT/tag table selection logic, and may perform a series of processes for packet forwarding. As an embodiment, the processor is involved in the initialization of each table, and the packet forwarding process may be performed in each table implementation logic. The packet forwarding process may include FIGS. 2 to 5 and FIGS. 13 to 18 below. When a tag is included in the received packet, based on the PIT/tag table selection logic, the processor searches the tag table for an entry matching the tag of the received packet and the interface of the received packet. I can. In addition, when the search is successful, the processor may change a tag (in_tag) of a packet to an output tag (out_tag) or change an output tag (out_tag) to a tag (in_tag) based on the tag table. On the other hand, when the search fails in the tag table, the processor may change the tag of the received packet to the output tag by allocating an output tag. In addition, when a tag is not included in the received packet, the processor may assign a tag to the received packet based on the interface on which the packet was received. The tag assigned to the packet and the received interface may be added to the tag table. When the received packet is received from an external network, the processor may forward the packet through the Network Name (FIB) Forwarding Information Base (FIB) lookup. When a received packet is received from an internal network, the processor may perform packet forwarding through the Data Name Prefix FIB lookup. When the network name of the received packet is the same as the router region, the processor may forward the packet through the Data Name Prefix FIB lookup. When a tag is included in the received packet, the processor may update the tag and the interface on which the packet is received in the PIT (Pending Interest Table).

As an embodiment, when a data packet is received through the interface, and the data packet includes an output tag, the processor may search for an entry matching the output tag in the tag table. The processor may change the output tag to the interest packet tag of the matched entry, and forward the received packet to the interface of the matched entry. In the case of a data packet that does not include an output tag in the received packet, the processor transmits the packet through an interface and a tag of an interest packet obtained based on a data name prefix from a PIT (Pending Interest Table). Can be forwarded.

Meanwhile, the processor may be expressed as, for example, a packet forwarding unit. This will be described in more detail in FIG. 19.

As an embodiment, the content store 602 may caching a data packet passing through a corresponding node.

As an embodiment, the PIT (Pending Interest Table) 603 may be a table that stores information on an interest packet that has not received a data packet. This will be described in more detail in FIG. 11.

As an embodiment, the Forwarding Information Base (FIB) 604 may be based on a data name prefix and a network name, and a node may look up an FIB to forward an interlist packet.

As an embodiment, the tag table 605 may include tags allocated to packets for each interface that can be used for transit traffic to other networks and related information of interest packets in the form of a table. have. In this case, the tag may include a general tag (input tag, in_tag) and an output tag (out_tag). For example, the tag table includes a tag of the received packet, an output tag for forwarding the received packet, information on the best reception time of the received packet, and a flag for aging management of the received packet. It may include at least any one. This may be the same as mentioned above, and will be described in more detail in FIGS. 7 to 19 below.

As an embodiment, the PIT/tag table selection logic 606 may perform overall control of the tag table. In addition, depending on whether the interface to which the interest packet is input is connected to an external network and whether the network name of the interest matches the region of the node, that is, the router region, a PIT or tag table can be appropriately selected. . As an embodiment, for a data packet, when a tag is included, a tag table may be selected, and when a tag is not included, a PIT may be selected. More specifically, according to the PIT/tag table selection logic, when the interest packet is received from an external network, a tag table is selected when the network name is different from the node region, that is, the router region, and when received from the internal network. You can select a tag table only when the interest packet contains a tag. In other cases, PIT can be selected. In this case, the node may recognize as a preset value whether the network connected for each interface is internal or external, or may be recognized by an appropriate inter-router control protocol. In addition, as an embodiment, the processor may interlock with or include the PIT/tag table selection logic, and the PIT/tag table selection logic directly changes the tag (in_tag) of the packet to the output tag (out_tag) based on the tag table. Alternatively, it is possible to change the output tag (out_tag) to a tag (in_tag).

As an embodiment, one or more interfaces 607 may transmit/receive an interest packet and a data packet to and from a plurality of remote nodes.

Meanwhile, since the node of FIG. 6 and the following node correspond to an embodiment, some components may be excluded or other components may be further included. In addition, one component may be implemented by a plurality of hardware or software, and a plurality of components may be implemented by a single hardware or software. As an embodiment, the processor 601 and/or the PIT/tag table selection logic 606 may be expressed as one processor or one packet forwarding unit. This will be described in more detail in FIG. 19.

As an embodiment, the description of FIG. 6 may be applied to the nodes shown in FIGS. 7 and 8 as long as the description of FIG. 6 is not arranged with the contents of FIGS. 7 to 8 below.

7 is a diagram for describing another example of a node constituting a content-oriented network applicable to the present disclosure. In more detail, it is a diagram for explaining in detail the configuration of PIT, tag table, and PIT/tag table selection logic and FIB configuration that may be included in a node constituting a content-oriented network.

As an embodiment, although not shown, the node of FIG. 7 may include a PIT, a tag table, and PIT/tag table selection logic, a processor, a content store, and one or more interfaces in addition to the FIB.

As an embodiment, the FIB of the node may include a data name prefix FIB and a network name FIB. In processing the local interest packet 701, which is an interest packet existing in the network of the data requester or data owner, the data name prefix FIB 703 is looked up through CS and PIT lookup. I can.

On the other hand, in the case of a transit interest packet 702, which is an interest packet existing in other networks, a process of looking up the network name FIB 704 may be performed through a tag table. In this case, the lookup of the network name FIB may be performed based on a Forwarding Hint field that stores the network name together with the data name included in the interest packet.

In this case, the lookup of the data name prefix FIB or the network name FIB may be performed by the LPM (Longest Prefix Match) lookup method.

As an embodiment, even in the case of a transit interest packet, the network name of the transit interest packet is generally much shorter than the data name prefix and may have a limited length. Therefore, separating the network name FIB from the data name prefix FIB, the network name FIB It is possible to implement a high-speed lookup with hardware such as TCAM (ternary content-addressable memory), and thus provide excellent temporal performance when searching for FIB. Since most of the traffic in the Internet core network is transit traffic, according to this, by separating the network name FIB from which the transit traffic looks up, it can have an advantage that it is easy to increase the capacity of the core router of the content-oriented network.

8 is a diagram for explaining another example of a node constituting a content-oriented network applicable to the present disclosure. In more detail, it is a diagram for explaining in detail the configuration of PIT, tag table, and PIT/tag table selection logic and FIB configuration that may be included in a node constituting a content-oriented network.

As an embodiment, the interest packet of FIG. 8 may include an application name instead of a data name. In addition, although not shown, the node of FIG. 8 may include a PIT, a tag table, and a PIT/tag table selection logic, a processor in addition to the FIB, and one or more interfaces. In this case, the content store CS for caching data packets passing through the corresponding node may not be included. When the content store is included in the node, it may be implemented as hardware or software interlocked with the PIT.

As an embodiment, the interest packet may include an application name instead of a data name, and may include a network name. The FIB of the node may include an application name prefix FIB 801 and a network name FIB. In this case, the network name FIB may be the same as described in FIG. 7. In processing a local interest packet, which is an interest packet existing in a network of a data requester or data owner, the application name prefix FIB 801 may be looked up through CS and PIT lookup.

On the other hand, in the case of a transit interest packet, which is an interest packet existing in other networks, a process of looking up a network name FIB may be performed through a tag table as mentioned above. In this case, the lookup of the network name FIB may be performed based on a Forwarding Hint field that stores the network name together with the data name included in the interest packet.

In this case, the lookup of the data name prefix FIB or the network name FIB may be performed by the LPM (Longest Prefix Match) lookup method.

9 is a diagram illustrating an example of a tag table in a content-oriented network applicable to the present disclosure.

As an embodiment, the tag table may include a tag (input tag, in_tag) 901, an interface (in_interface, in_if) 902, an output tag (out_tag) 903, and the best reception time (last_visit_time) 904. have. The order of each field in the tag table is given arbitrarily and is not limited to FIG. 9.

As an embodiment, the tag (in_tag) 901 may mean a current tag (or an input tag) of an interest packet. Tags are not immutable and may change during packet transit. In the case of interest packet forwarding, the tag of the packet may be an in_tag value, and for example, it may be changed to an output tag (out_tag) value each time it passes through a node that processes transit traffic. In the case of data packet forwarding, the tag of the packet may be an out_tag value, and for example, it may be changed to a tag value corresponding to in_tag each time it passes through a node that processes transit traffic. This is as mentioned above, and may be described in more detail below with reference to other drawings.

As an embodiment, the interface (in_interface, in_if) 902 may be an interface to which an interest packet is input. As mentioned in FIG. 6, one or more interfaces may exist, and may be used to transmit/receive interest packets and data packets to a remote node.

As an embodiment, the output tag (out_tag) 903 is a tag value obtained as a tag is changed when an interest packet is forwarded, and may be determined from an entry that matches a tag (in_tag) of the tag table. . That is, it may be a value that is matched by searching the tag table with {in_tag, in_if} based on the interest packet including in_tag. The output tag, which is a matched value, may be a tag value of an interest packet during packet forwarding. A data packet that is a response to an interest packet may include an output tag (out_tag). The output tag Out_tag may be an entry of a tag table within a node, that is, assigned as a unique value (unique value) for each interface. Meanwhile, the output tag may be a tag included in the data packet, and the tag of the data packet may be changed to a value corresponding to in_tag as the search succeeds in searching for an entry matching the output tag in the tag table. This is as mentioned above, and may be described in more detail below with reference to other drawings.

As an embodiment, the best reception time (last_visit_time) 904 may indicate a reception time of the most recent interest packet matched to a corresponding entry, that is, input to a corresponding interface. As an embodiment, a tag table aging time is set for each node, and a value obtained by subtracting the best reception time (last_visit_time) from the current time (current_time) at an appropriate time interval is less than the aging time. Large entries can be deleted. The best reception time (Last_visit_time) does not update for each matched interest packet, and samples the interest packet at a certain rate, e.g. 1/1000, or at a certain time interval, e.g., the interest packet sampled once every second. You can also update about it.

10 is a diagram illustrating another example of a tag table in a content-oriented network applicable to the present disclosure.

As an embodiment, InFace (IF) of FIG. 10 may be the same as 902 of FIG. 9, InTag (ITag) may be the same as 901 of FIG. 9, and OutTag (Otag) may be the same as 903 of FIG. 9.

As an embodiment, the tag table may simply manage interest packet information by using the flag 1001 for aging management, which is an active flag, instead of the best reception time (last_visit_time). As an embodiment, the active flag may be reset to '0' at the start of each period with a certain aging time period in the tag table. Thereafter, when the interest packet accesses the entry of a specific tag, that is, accesses the entry matching the interface of the interest packet, the active flag may be set to '1'.

At the end of the aging time period, the tag table aging management can be simplified by removing entries whose active flag is 0 until then. This method can greatly reduce the number of times the entry is modified when the tag table is looked up for matching interest packets after the entry is created in the tag table, so performance can be improved when looking up the tag table. .

In the tag table of the present disclosure, including FIGS. 9 and 10, a newly selected entry is searched in the tag table, that is, when the pair of {in_tag, in_if} matching the tag (in_tag) of the interest packet does not exist in the tag table, an entry is newly created. Can be generated. At this time, while creating an entry, a value of an output tag (out_tag) that will become a new tag of the interest packet may be assigned a unique value within the node. At this time, if there is no tag in the received interest packet, it may be stored as in_tag=null. When looking up the tag table for an interest packet, you can search the {in_tag, in_if} field, and when looking up the tag table for a data packet with a tag, you can search the {out_tag} field. The tag table may be implemented as a ternary content-addressable memory (TCAM).

Meanwhile, a single tag table may simultaneously support a search for an interest packet and a search for a data packet, or it may be possible to divide the tag table into a tag table for interest packet search and a tag table for data search.

11 is a diagram for describing an example of a PIT (Pending Interest Table) supporting a tag in a content-oriented network applicable to the present disclosure. As an embodiment, the PIT of FIG. 11 may be compatible with the tag table described above. In other words, tag table and PIT can be selected and used according to packets to be forwarded. As another embodiment, when a node is not located in a core network supporting transit traffic, it is possible to implement only PIT without a tag table and use it for packet forwarding.

As an embodiment, the modified PIT of FIG. 11 may include a data name 1101, a list of interface and tag pairs 1102, and other fields 1103.

As an embodiment, while a general PIT stores an input interface list for a data name, the modified PIT supporting a tag is {interface (in_interface), tag (in_tag)} (or {tag, interface}) list ( 1102). When an interest packet without a tag is received, null may be recorded in the in_tag field of the modified PIT. As an embodiment, the data name 1101 /x/y/z/a of FIG. 11 is a case where the data requestor's internal network and the external network are simultaneously requested, and the eth_3 interface of the node is a tag-free interest in the internal network. A packet has been received, and an interest packet with tag=150 may be received from an external network through the eth_1 interface.

Meanwhile, in the case of a data packet not including a tag, the modified PIT may be used to search for matching entries based on the data name. If the search for a matching entry is successful, the input interface and the tag value of the interest packet can be acquired, and the data packet can be forwarded through the acquired input interface with the acquired tag value included.

As an embodiment, a nonce value or other information may be recorded in the other field 1103.

12 is a diagram illustrating a packet format in a content-oriented network according to an embodiment of the present disclosure. In more detail, a diagram for describing various examples of packet formats in a content-oriented network applicable to the present disclosure.

As an embodiment, the content-oriented network packet may include an interest packet and a data packet. In a content-oriented network network, an additional packet format may exist in addition to the interest packet and data packet format of the basic structure as shown in a), d) and e) of FIG. 12.

12A is an example of a basic interest packet that does not include a network name. In this case, the node constituting the content-oriented network treats the network name of the interest packet as the same as the region of the content-oriented network node, that is, the region (router region) of the node that received the current interest packet, and forwarding based on the data name. You can do it.

b) is an embodiment of a general interest packet to which a network name is added, and c) is an interest packet to which a network name and a tag are added, and is an example of a format of a transit interest packet. d) is an example of a general data packet format, and includes only a data payload and a data name, and e) is an example of a data packet format including a tag.

Meanwhile, the above-mentioned interest packet, PIT, and tag table may be described with reference to FIGS. 9 to 12 above.

A flow chart of a forwarding procedure of an interest packet and a data packet in a content-oriented network node proposed in the present disclosure is illustrated in FIGS. 13 to 18.

13 is a diagram for describing an interest packet forwarding method performed by a node constituting a content-oriented network according to an embodiment of the present disclosure.

As an embodiment, the interest packet forwarding method of FIG. 13 may be based on the packet forwarding procedure of FIGS. 2 to 5 and 15 to 18.

As an embodiment, the node performing the interest packet forwarding method of FIG. 13 may exist in the same network as the node in which the data requestor is located, or in the same network as the node in which the data owner is located. It can be located within the network. In addition, the node may be a PE node of FIGS. 6 to 8 and 19 or a non-PE node, but is not limited thereto.

As an embodiment, the node may first receive 1701 an interest packet of a content-oriented network. The received interest packet may be one of the interest packets of FIG. 12. The interest packet may be received through the node's interface.

In this case, it may be checked whether the received interest packet is a local interest packet or a transit interest packet (1702). If the network name of the interest packet received from the interface connected to the external network is not the same as the router region of the node, or if the tag is included in the interest packet received from the interface connected to the internal network, it is classified as a transit interest packet. do. All other cases can be classified as local interest packets. That is, when the network name of the interest received from the external network is the same as the router region of the node, or when the interest received from the internal network does not include a tag, it may be classified as a local interest packet.

When classified as a transit interest packet, after tag table processing is performed (1306), packet processing is performed through FIB lookup based on the network name. In the case of a non-transit-interest packet, that is, a local-interest packet, after PIT processing is performed (1304), packet processing is performed through FIB lookup based on the data name.

Meanwhile, 1304 and 1306 described above will be described in more detail with reference to FIGS. 15 to 17 below.

This will be described in more detail below with reference to FIGS. 15 to 18.

14 is a diagram illustrating a data packet forwarding method performed by a node constituting a content-oriented network according to an embodiment of the present disclosure.

As an embodiment, the forwarding method of the interest packet for the data packet of FIG. 14 may be based on the packet forwarding procedure of FIGS. 13, 15, and 16, and the data packet forwarding method of FIG. 14 may be based on FIG. 18. I can.

As an embodiment, the node performing the interest packet forwarding method of FIG. 14 may exist in the same network as the node in which the data requestor is located, or in the same network as the node in which the data owner is located. It can be located within the network. In addition, the node may be a PE node of FIGS. 6 to 8 and 19 or a non-PE node, but is not limited thereto.

As an embodiment, the node may first receive (1801) a data packet of a content-oriented network. The received data packet may be one of the data packets of FIG. 12. The data packet can be received through the interface of the node.

Thereafter, the node may review (1802) whether a tag is included in the received packet (data packet). If it is determined that the tag is included, the tag will correspond to the output tag of the tag table, so it is possible to search whether there is an entry matching the output tag of the tag table. When the data packet does not include a tag, a PIT search may be performed based on a data name included in the data packet (1603). This will be described in more detail in FIG. 18.

Thereafter, the node can forward the received packet. In this case, when a tag is included in the received data packet, the data packet may be forwarded 1803 to another node based on the tag. When the tag table is based, since the data packet includes a tag value corresponding to the (out_tag) field of the tag table, packet forwarding may be performed while being changed to the tag (in_tag) of the tag table again. In addition, since the data packet may be forwarded through the input interface of the interest packet, interest packet forwarding and symmetric forwarding may be performed. This may be the same as described above with reference to other drawings.

15 is a diagram for explaining an interest packet forwarding method in a PE node in a content-centric network according to an embodiment of the present disclosure, and FIG. 16 is a non-PE node in a content-centric network according to an embodiment of the present disclosure A diagram for explaining an interest packet forwarding scheme in FIG.

In more detail, the packet forwarding scheme may include a case where the node does not exist in the destination network when an interest packet is forwarded from an external network to a network in which a node performing packet forwarding exists. That is, the node may be a node performing transit traffic. At this time, the node may be a PE node or a non-PE node. In the case of a PE node, the packet forwarding unit (and/or processor, PIT/tag table selection logic) selects and inputs the packet forwarding to the tag table. Tags that are uniquely assigned for each interface can be included in the interest packet. In this case, the external network may be a customer network or another business network. The tag may be used for the purpose of forwarding an interest packet through an input interface when a data packet returns, that is, forwarding the interest packet in the reverse direction along the path as it was forwarded to the data owner.

Here, FIG. 15 is a diagram for explaining an interest packet forwarding procedure in a PE node, and FIG. 16 is a diagram for explaining an interest packet forwarding procedure in a non-PE node.

As an embodiment, in the packet forwarding procedure, after the PE/Non-PE node receives the interest packet, appropriately parses (1301, 1401), and performs necessary processing based on the content store (CS) ( operation) (1302, 1402).

Thereafter, it may be examined whether the network name of the interest packet is the same as the router region, that is, the region of the node (1303, 1403). In FIGS. 13 and 14, a case in which the network name of the interest packet and the router region are different will be described as an embodiment, and a case 1304 in which the network name and the router region are the same will be described with reference to FIG. 15.

As an embodiment, when packet forwarding is performed by the PE node, it may be checked whether or not the received interest packet is received from an external network (1305). That is, it can be checked whether the interest packet is included in the transit traffic.

At this time, if the interest packet is received from an external network, it is reviewed whether or not a tag is included in the interest packet (1306), and if the tag is included, an entry including the tag in the tag table, that is, You can search for matching entries. In this case, a tag table may be searched 1308 using a pair {tag, in_if} of a tag of a current interest packet and an input interface (in_if) from which the interest packet is received.

As an embodiment, if the received interest packet is not received from an external network, it is checked whether or not a tag is included in the interest packet (1309), and if the tag is included, the packet is received from the external network. It can be treated in the same way as the case of If the tag is not included, after the PIT process is generally performed (1311), the FIB lookup may be performed. Here, the FIB lookup may be performed based on the data name prefix FIB.

The above process may be performed regardless of whether an interest packet is received from an external network in the case of a non-PE node. That is, the non-PE node may examine whether a tag is included in the interest packet (1405) after the necessary CS processing is performed (1402). When a tag is included, a tag table may be searched 1407 by a pair {tag, in_if} of a tag of a current interest packet and an input interface (in_if) in which the interest packet is received.

As an embodiment, if the matching entry exists in the tag table (1310, 1408), the tag of the interest packet is changed to the output tag (out_tag) of the matching entry (swap) (1314, 1410), and through FIB lookup. Packet forwarding (1315, 1411) may be performed. In this case, the FIB lookup may be performed based on the network name prefix FIB. As another embodiment, if there is no matching entry in the tag table (1310, 1408), the {in_tag, in_if, out_tag} entry is added to the tag table by newly allocating an unused output (out_tag) value in the node. The tag (in_tag) of the rest packet may be changed to an output tag (out_tag) (1313, 1409). Thereafter, as mentioned above, packet forwarding 1315 and 1411 may be performed through FIB lookup. In this case, the FIB lookup may be performed based on the data name prefix FIB. Meanwhile, even if the interest packet does not include a tag, in the case of a PE node, a tag may be included in the interest packet by allocating (1307) a new tag to the packet based on the input interface (in_if). In this case, it may be expressed in the tag table that the tag is not included in the interest packet and that the tag is newly included. For example, a {null, in_if, tag} entry may be added to the tag table. Here, a tag, which is one of the added tag table entry fields, may be a newly allocated tag.

Through the above process, an interest packet including a tag may be forwarded to enter a destination network in which the data owner exists. The process after entering the destination network will be described in more detail with reference to FIG. 17.

Meanwhile, the above-mentioned interest packet, PIT, and tag table may be described with reference to FIGS. 9 to 12 above.

17 is a diagram for explaining an interest packet forwarding method when a network name in a content-oriented network is the same as a router region according to an embodiment of the present disclosure. In more detail, a diagram for explaining a method of performing interest packet forwarding by a PE/Non-PE node of a destination network to a data owner.

As an embodiment, the interest packet may have passed through the processes of FIGS. 13 and 15 to 16 until it is forwarded to the destination network, and the interest packet may be expressed as a local packet.

As an embodiment, when the edge router determines that the destination of the interest packet is the corresponding network (1304), the PE/Non-PE node may review 1501 whether or not a tag is included in the interest packet. have.

When a tag is included in the interest packet (1502), the tag may be stored in the PIT while removing the tag of the interest packet. In this case, the input interface and the tag of the packet may be stored in the PIT as a pair ({in_if, tag} or {tag, in_if}), and the current tag of the interest packet may be stored in the in_tag field.

On the other hand, when the tag is not included in the interest packet (1503), general PIT processing may be performed.

Thereafter, the data name prefix FIB lookup may be performed (1504) regardless of whether or not the tag of the interest packet is used. Through this, forwarding 1504 to the owner of the data in the corresponding network may be performed.

Meanwhile, the above-mentioned interest packet, PIT, and tag table may be described with reference to FIGS. 9 to 12 above.

18 is a diagram for describing a data packet forwarding method in a content-oriented network according to an embodiment of the present disclosure. In more detail, to explain a method of performing symmetric forwarding in which the PE/Non-PE node forwards the data packet that is the response of the interest packet, but the forwarding path of the data packet follows the interest packet forwarding path. It is a drawing.

As an embodiment, the data packet, PIT, and tag table mentioned in FIG. 18 may be those described above with reference to FIGS. 9 to 12.

As an embodiment, a data packet may be received from another node, and appropriate data packet parsing may be performed (1601 ). The data packet may be forwarded from a network to which the current node belongs or a node located in an external network.

The node that has performed necessary data parsing may review (1602) whether a tag is included in the received data packet.

When a data packet includes a tag, data packet forwarding may be performed based on a tag table. As an embodiment, a tag table may be searched based on a tag included in a data packet. In this case, the tag included in the data packet is searched for the {out_tag} field of the tag table (1605, 1606), and if an entry matching the tag table exists, that is, if the search is successful, the tag included in the data packet is The tag may be changed to a tag of a matching entry (1607). That is, since the tag included in the data packet may correspond to the out_tag of the tag table, it may be changed to the tag value in the {in_tag} field of the matching entry. Meanwhile, when there is no matching entry in the tag table, that is, when the search fails, the data packet may be forwarded through a general PIT process 1608.

Meanwhile, when the data packet does not include a tag, a PIT search may be performed (1603) based on a data name included in the data packet. That is, based on the data name, a search may be performed to see if there is an entry matching the PIT, and of the matching entry, the input interface of the interest packet and the tag pair {in_if, in_tag} or {in_tag, in_if} may be obtained. have. The acquired interface and tag pair may be forwarded through the interface (in_if) by including the acquired tag value (in_tag) in the data packet.

In this case, when the tag value (in_tag) obtained from the PIT is null, forwarding through the interface (in_if) may be performed in a state in which the tag is not included in the data packet (1604). That is, “” of 1604 may mean a data packet.

Meanwhile, the above-mentioned interest packet, PIT, and tag table may be described with reference to FIGS. 9 to 12 above.

19 is a diagram illustrating a node constituting a content-oriented network according to an embodiment of the present disclosure. In more detail, it may be a node constituting a content-centric network capable of executing the packet forwarding method described above, and may be a computing system or a packet processing device that executes the packet processing method in the content-centric network.

As an embodiment, a node constituting a content-oriented network may include a PIT and tag table 1901, one or more interfaces 1902, and a packet forwarding unit 1902. The PIT and tag table 1901 may be used to forward the interest packet to the data owner, as mentioned above. One or more interfaces 1902 may be configured to receive and transmit interest packets and data packets for a plurality of remote nodes, as mentioned above. As mentioned above, the packet forwarding unit 1902 may perform packet forwarding using the tag table and/or PIT.

As an embodiment, a node that performs packet forwarding constituting a content-oriented network may have a packet received through the interface as the interest packet, and a network name of the received packet may be different from a router region. In this case, the forwarding of the received packet includes an entry matching the tag of the received packet and the interface of the received packet in the tag table when the tag is included in the received packet. It can be done by searching.

In addition, when the search is successful in the tag table, the packet forwarding unit may change the tag of the received packet to an output tag of the entry. If the search fails in the tag table, an output tag may be assigned to the received packet, and the packet forwarding may change the tag of the received packet to the output tag. When a tag is not included in the received packet, the packet forwarding unit may assign a tag to the received packet based on an interface on which the packet was received.

In addition, when the received packet is received from an external network, forwarding of the received packet may be performed by the packet forwarding unit through a network name FIB lookup. On the other hand, when the received packet is received from an internal network, forwarding of the received packet by the packet forwarding unit may be performed through a Data Name Prefix FIB lookup.

As an embodiment, a packet received through the interface may be an interest packet, and a network name of the received packet may be the same as a router region. In addition, forwarding of the received packet by the packet forwarding unit may be performed through a Data Name Prefix FIB lookup. When a tag is included in the received packet, the tag and the interface on which the packet is received may be updated in a PIT (Pending Interest Table).

In one embodiment, the tag table includes a tag of the received packet, an output tag for forwarding the received packet, information on the best reception time of the received packet, and a flag for aging management of the received packet. It may include at least any one of. In addition, when the received packet is a data packet including an output tag, forwarding of the received packet may be performed by the packet forwarding unit searching the tag table for an interface of an interest packet matching the output tag. . In this case, in the packet forwarding unit, the output tag may be changed to a tag of the interest packet associated with the matched interface.

As an embodiment, in the case of a data packet that does not include an output tag in the received packet, forwarding of the received packet is performed by the packet forwarding unit based on a data name prefix in a PIT (Pending Interest Table). It can be performed through the tag and interface of the acquired interest packet.

Meanwhile, although omitted in FIG. 19, the node and/or device of FIG. 19 may include a memory including a random access memory (RAM) and a read only memory (ROM), a processor, a storage, a bus, a PIT, an FIB, and a CS. have.

Various embodiments of the present disclosure are not listed in all possible combinations, but are intended to describe representative aspects of the present disclosure, and matters described in the various embodiments may be applied independently or may be applied in combination of two or more.

In addition, various embodiments of the present disclosure may be implemented by hardware, firmware, software, or a combination thereof. For implementation by hardware, one or more ASICs (Application Specific Integrated Circuits), DSPs (Digital Signal Processors), DSPDs (Digital Signal Processing Devices), PLDs (Programmable Logic Devices), FPGAs (Field Programmable Gate Arrays), general purpose It may be implemented by a processor (general processor), a controller, a microcontroller, a microprocessor, or the like. For example, it is obvious that it can be implemented in the form of a program stored in a non-transitory computer readable medium that can be used at the end or edge, or a program stored in a non-transitory computer readable medium that can be used at the edge or the cloud. Do. In addition, it can be implemented by a combination of various hardware and software.

In the object autonomous control program through imitation learning according to an embodiment of the present disclosure, obtaining an output value of an object operated according to an operation command based on a policy in a computer, determining a compensation for the policy, and performing the compensation. It may be a program stored in a non-transitory computer-readable medium that executes the step of updating the policy in the direction of optimizing.

The scope of the present disclosure is software or machine-executable instructions (for example, operating systems, applications, firmware, programs, etc.) that allow an operation according to a method of various embodiments to be executed on a device or a computer, and such software or It includes a non-transitory computer-readable medium (non-transitory computer-readable medium) which stores instructions and the like and is executable on a device or a computer.

The present disclosure described above is capable of various substitutions, modifications, and changes within the scope of the technical spirit of the present disclosure to those of ordinary skill in the technical field to which the present disclosure belongs, so the scope of the present disclosure is described above. It is not limited by one embodiment and the accompanying drawings.

1901: PIT and tag table
1902: multiple interfaces
1903: packet forwarding unit

Claims (20)

In the nodes constituting the content-oriented network,
A plurality of interfaces configured to receive and transmit interest packets and data packets to a plurality of remote nodes;
A PIT (Pending Interest Table) for storing information of a local interest packet corresponding to a network in which a source network or a destination network is located among interest packets that have not been answered;
A tag table for storing information of a transit interest packet, which is not a network in which a corresponding node is located, from the source network and the destination network among the unanswered interest packets in a group unit associated with the source network;
Forwarding Information Base (FIB) for forwarding an interest packet to a destination node;
Packet forwarding, including a processor that selects the PIT or the tag table according to whether the packet is a local packet or a transit packet during packet forwarding, and performs packet forwarding based on the selected table. The node that performs.
The method of claim 1
When the destination network name of the interest packet received from the interface connected to the external network is not the same as the network name in which the node is located, or a tag is included in the interest packet received by the interface connected to the internal network,
The processor is
A node that performs packet forwarding, dividing the received interest packet into a transit interest packet.
The method of claim 1,
When an interest packet received from an interface connected to the internal network does not contain a tag, or an interest packet having the same destination network name as the network name in which the node is located is received from an interface connected to the external network.
The processor is
A node that performs packet forwarding, dividing the received interest packet into a local interest packet.
The method of claim 2,
When a tag is included in the received transit interest packet,
The processor is
A node performing packet forwarding, searching the tag table for an entry matching the tag of the received packet and interface related information of the received packet.
According to claim 4
If the tag table search is successful,
The processor is
A node that performs packet forwarding, changing the tag of the received packet to the output tag of the entry.
According to claim 4
If the search fails in the tag table above,
The processor is
A node that performs packet forwarding to change the tag of the received packet to the output tag by allocating an output tag.
The method of claim 2
When a tag is not included in the received transit interest packet
The processor is
A node performing packet forwarding, which assigns a tag to the received packet based on the interface-related information on which the packet is received.
The method of claim 7,
The tag assigned to the packet and the received interface-related information are added to the tag table.
The method of claim 3
When the received local interest packet includes a tag
A node that removes the tag from a header, stores the tag and reception interface information in a PIT entry of an interest, and performs packet forwarding.
The method of claim 1
Each entry of the PIT includes a tag of a received interest packet and interface information of the received interest packet.
The method of claim 3,
The FIB is a node that performs packet forwarding, including a network name FIB (network name FIB) and a data name prefix FIB (data name prefix FIB).
The method of claim 11,
When the received interest packet is a transit interest packet,
The processor is
A node performing packet forwarding for forwarding a packet through the network name (FIB) Forwarding Information Base (FIB) lookup.
The method of claim 11,
If the received interest packet is a local interest packet,
The processor is
A node that performs packet forwarding, performing packet forwarding through the Data Name Prefix FIB lookup.
The method of claim 1,
The tag table is
Includes a tag of the received packet and an output tag for forwarding of the received packet,
The node further comprising at least one of latest reception time information of the received packet or a flag for aging management of the received packet.
The method of claim 1,
Receiving a data packet through the interface,
If the data packet contains a tag,
The processor is
A node performing packet forwarding for searching the tag table for an entry having an output tag matching the tag.
The method of claim 15,
The processor,
The tag is changed to the interest packet tag of the matched entry,
A node performing packet forwarding for forwarding the received packet to the interface of the matched entry.
The method of claim 1,
When the received packet is a data packet without an output tag,
The processor is
Search for an interest packet matching the data name of the data packet in the Pending Interest Table (PIT),
If the search is successful, the node performs packet forwarding, adding a tag of the interest packet to a data packet, and forwarding the packet based on interface information receiving the interest packet.
In the packet forwarding method performed by a node constituting a content-oriented network,
Receiving an interest packet in a content-oriented network;
Including; examining whether a tag is included in the interest packet, and forwarding the interest packet; including,
The interest packet was received by an interface connected to an external network, and the destination network name of the interest packet is not the same as the network name in which the node is located, or
When the interest packet is received by an interface connected to the internal network, and a tag is included in the interest packet,
The received interest packet is divided into a transit interest packet.
The method of claim 18,
The interest packet was received on an interface connected to the internal network and did not contain a tag, or
When the destination network name is the same as the network name where the node is located on the interface connected to the external network,
The packet forwarding method, wherein the received interest packet is divided into a local interest packet.
In the packet forwarding method performed by a node constituting a content-oriented network,
Receiving a data packet of a content-oriented network;
Including; examining whether a tag is included in the received data packet, and forwarding the data packet; Including,
When a tag is included in the received data packet,
Searching for an entry including an output tag matching the tag in the tag table,
A packet forwarding method for performing forwarding based on interface information of an interest packet included in the entry.

Images