WO2016170559A1 - Procédé et système pour une réservation de bande passante pour une ingénierie de trafic à base d'arbre multiple dans des réseaux centrés sur le contenu - Google Patents

Procédé et système pour une réservation de bande passante pour une ingénierie de trafic à base d'arbre multiple dans des réseaux centrés sur le contenu Download PDF

Info

Publication number
WO2016170559A1
WO2016170559A1 PCT/JP2015/002213 JP2015002213W WO2016170559A1 WO 2016170559 A1 WO2016170559 A1 WO 2016170559A1 JP 2015002213 W JP2015002213 W JP 2015002213W WO 2016170559 A1 WO2016170559 A1 WO 2016170559A1
Authority
WO
WIPO (PCT)
Prior art keywords
tree
router
interest
packet
chunk
Prior art date
Application number
PCT/JP2015/002213
Other languages
English (en)
Inventor
Ling Xu
Tomohiko Yagyu
Original Assignee
Nec Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nec Corporation filed Critical Nec Corporation
Priority to PCT/JP2015/002213 priority Critical patent/WO2016170559A1/fr
Publication of WO2016170559A1 publication Critical patent/WO2016170559A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/302—Route determination based on requested QoS
    • H04L45/306—Route determination based on the nature of the carried application
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/63—Routing a service request depending on the request content or context
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/64—Routing or path finding of packets in data switching networks using an overlay routing layer

Definitions

  • the present invention relates to a method and a system for name-based bandwidth reservation for MTTE (Multiple Tree-based Traffic Engineering) in CCN (Content-Centric Networking) networks.
  • MTTE Multiple Tree-based Traffic Engineering
  • CCN Content-Centric Networking
  • the present invention focuses on reducing the energy consumption of CCN networks, and providing superior transmission performance for higher priority traffics in CCN networks.
  • CCN is a network architecture proposed in recent years for solving problems of the IP architecture (see Non-Patent Literature 1).
  • Conventional IP networks focus on retrieving needed contents from certain hosts.
  • CCN networks each content is given an unique name. Users tell the networks the names of the needed contents. The networks independently identify hosts providing the contents, and fetch the contents back to the users.
  • CCN One of CCN’s main features is that routers carry caches to reduce redundant traffic.
  • Each host in the network may provide some contents and this host is called the producer of its contents.
  • the host registers the name of each of its contents on each router. This process is called content publishing.
  • Hosts that request the content is called the consumers of this content.
  • a consumer sends an Interest to its adjacent router.
  • the Interest is then forwarded toward to the producer according to the information the producer registered on routers.
  • the producer receives an Interest, it packs the content into content object packets, denoted, and sends it back to the consumer.
  • Each router along content object’s forwarding path tries to store the content object in its cache. The next time the router receives an Interest for the same content, if the content object is still in its cache, the router replies the content object to the consumer directly.
  • Non-Patent Literature 2 describes Multiple Tree-based Traffic Engineering (MTTE) to reduce CCN’s energy consumption by shutting down as many network interfaces as possible. Previous research has found that interfaces in modern networks are generally underutilized (see Non-Patent Literature 3).
  • MTTE Traffic Engineering
  • MTTE splits traffic on multiple tree-topology overlay networks generated based on the physical network.
  • the trees are generated in such a way that the number of interfaces included in the trees is minimized. Since trees generally contain fewer interfaces than the original physical network, energy can be conserved.
  • Fig. 11 depicts the architecture of MTTE.
  • Fig. 12 depicts an example of MTTE’s work flow.
  • the network contains a central server called the controller (CTR) 61.
  • CTR controller
  • Each circle in Fig. 12 represents a router (RTR) 64.
  • Fig.12 (a) represents the network’s physical topology.
  • the controller (CTR) 61 maintains a database called the tree set. Initially, the controller (CTR) 61 creates one spanning tree (see Fig. 12 (b)) based on the physical topology and adds the tree to the tree set.
  • each network interface of each router contents a queue of a fixed length.
  • the interface also has a fixed capacity.
  • the ratio between the packet (Interest or content) arriving speed to the interface’s capacity is called the interface’s utilization.
  • utilization > 1.0 new incoming packets are added to the queue’s tail and await forwarding.
  • packets’ arriving rate exceeds routers’ forwarding speed routers drop packets and congestions occurs.
  • the controller (CTR) 61 When congestion occurs in the network, the controller (CTR) 61 creates more trees to spread the traffics. Specifically, routers (RTR) 64 periodically report the utilization of their interfaces to the controller (CTR) 61. The controller (CTR) 61 calculates the Congestion Rate (CR) as the maximal value of the received utilization.
  • CTR Congestion Rate
  • the controller (CTR) 61 When CR > ph tcc , where ph tcc is a system parameter, the controller (CTR) 61 creates one more tree and adds the new tree to its tree set. Fig.12 (c) shows an example of the second tree. The controller (CTR) 61 then asks routers to deliver the traffic on trees in the new tree set evenly. This mechanism is termed as Tree-based Congestion Control (TCC).
  • TCC Tree-based Congestion Control
  • each router (RTR) 64 contains a Forwarding Information Base (FIB).
  • An FIB contains a set of entries and each entry is a mapping from one prefix to one or multiple of the router’s interfaces.
  • the router (RTR) 64 finds the interface whose FIB entry’s prefix matches the packet’s name and forwards the packets via this interface.
  • the controller (CTR) 61 sends the tree set to all routers (RTR) 64.
  • H is a collision-proof hash function preloaded on each router (RTR) 64 and N(d) is the name of content d.
  • Routers update their FIBs so that d is forwarded on the H(N(d))%K-th tree, where K is the size of the new tree set. After the update, packets are generally evenly delivered on each tree and the congestion can be mitigated.
  • the controller (CTR) 61 removes off the last tree in the tree set.
  • the controller (CTR) 61 sends the new tree set to all routers (RTR) 64, and asks routers (RTR) 64 to update their FIBs. Routers shut down their adjacent interfaces that are not included in the new tree set. Accordingly, energy consumption can be reduced.
  • Non-Patent Literature 2 In the real world network, traffics that should be delivered with precedence exists. For example, during natural disasters, disaster alarms are of higher priority and entertainment TV programs are of low priority. During congestion, the quality of high priority traffic should be guaranteed.
  • MTTE described in Non-Patent Literature 2 does not consider differentiating traffic of different priorities.
  • a method for bandwidth reservation for MTTE in CCN networks wherein each router maintains multiple sets of FIBs, each packet has a tree-set-id field, each time the tree set changes, a controller gives a new tree set an unique id, dispatches the new tree set and the id as a new tree-set-id to all routers, when a router receives an update of the tree set from the controller, the router creates a new FIB according to the new tree set, and stores the new tree-set-id as the router’s latest-tree-set-id, when a router receives an Interest, the router assigns the latesttree-set-id to the Interest if the tree-set-id field of the Interest is NULL, when a router receives a packet, the router forwards the packet to an interface that packet's prefix of FIB corresponding to the packet's tree-set-id matches, when a producer receives an Interest, the producer forms a chunk with
  • a system for bandwidth reservation for MTTE in CCN networks includes: one or more routers; a controller which controls the each router; a producer which sends a content in chunks; and a consumer which requests a content, wherein each router maintains multiple sets of FIBs, each packet has a tree-set-id field, each time the tree set changes, the controller gives a new tree set an unique id, dispatches the new tree set and the id as a new tree-set-id to all routers, when the router receives an update of the tree set from the controller, the router creates a new FIB according to the new tree set, and stores the new tree-set-id as the router’s latest-tree-set-id, when the router receives an Interest, the router assigns the latesttree-set-id to the Interest if the tree-set-id field of the Interest is NULL, when the router receives a packet, the router forwards the packet to an interface that packet's prefix
  • superior transmission quality can be provided to higher priority traffics on MTTE-enabled CCN networks that deliver traffics of different priorities.
  • FIG. 1 depicts a block diagram illustrating an exemplary embodiment of a network system according to the present invention.
  • Network system according to the present invention is applied to the CCN network based on MTTE.
  • the present invention assumes that in CCN networks, a producer (PDR) 30 splits each of its contents into a set of chunks.
  • a consumer (CSR) 40 requests a content
  • the a consumer (CSR) 40 requests each chunk one by one until receives all chunks.
  • session is usesd to denote the process that the consumer (CSR) 40 receives all chunks.
  • Each Interest belongs to one certain session.
  • sessions transmitting high (low) priority contents are called high (low) priority sessions.
  • high priority sessions transmission quality is improved by reserving bandwidth for them in a best effort manner.
  • each content has a optimal chunk-rate. Chunks of this content should be sent at this rate to ensure satisfiable transmission quality.
  • the consumers (CSR) 40 embed chunk-rates in names of their Interests.
  • Each router (RTR) 10 maintains a reservation table.
  • the router (RTR) 10 When a router (RTR) 10 receives an Interest, the router (RTR) 10 tries to reserve bandwidth for this session according to the chunk-rate embedded in the Interest’s name. The router (RTR) 10 maintains this reservation for a period. If the router (RTR) 10 cannot receive the next Interest of this session before a certain deadline, the router (RTR) 10 releases the reservation.
  • the router (RTR) 10 On each network interface, the router (RTR) 10 has two drop-tail queues that are used for high priority sessions and low priority sessions, respectively. The router (RTR) 10 always forwards packets of high priority sessions first.
  • Tunneling In MTTE, the controller (CTR) 20 adds trees when congestion occurs and removes trees when network is under-utilized, which results in frequent routing path change (route change).
  • CTR controller
  • route change wastes bandwidth.
  • a high priority session that transmits a binary file.
  • the session has successfully delivered all except the last chunk.
  • route change occurs.
  • Routers try to reallocate bandwidth for the session but failed.
  • the last chunks cannot be forwarded to the consumer (CSR) 40 and the whole session fails. Bandwidth reserved for transmitting this sessions is wasted.
  • route change can increase transmission delay and cause route loops.
  • forwarding paths are created along the shortest paths between producers (PDR) 30 and consumers on each tree.
  • Fig. 2 depicts an explanatory diagram illustrating a route change.
  • CSR consumer
  • PDR producer
  • route change occurs and the Interest is henceforth forwarded on a new tree (Fig. 2 (b)).
  • the Interest arrives at the producer (PDR) 30, it indeed traverses a route consisted of parts of both the original tree and the new tree (Fig. 2 (c)).
  • Such a route is not shortest on either the original or the new tree, which means a increase in the transmission delay.
  • the new route can even form loops.
  • routers (RTR) 10 can drop looped Interests, making the session fail.
  • the present invention proposes a transmission protocol called Tunneling to mitigate the negative influence of route change.
  • Each packet carries an ID of the route that it originally uses. Routers forward the packet according to the route ID until the session completes. This enables routers (RTR) 10 to deliver Interests of each session along the same route - the route on which bandwidth for this session is reserved.
  • RTR routers
  • the present invention reduces energy consumption of CCN networks, and provides better transmission quality for high priority traffics in CCN networks.
  • the network system includes one or more routers 10, a controller 20, a producer 30, and a consumer 40 (See Fig. 1).
  • the controller 20 controls the router 10.
  • the consumer 40 request a content using Interests.
  • the network system has a four routers 10. However, the number of routers is not limited to four.
  • Fig. 3 depicts an explanatory diagram illustrating the controller (CTR) 20.
  • the controller (CTR) 20 includes a physical network topology storage unit (PNT) 21, a control unit (CU) 22, and a comnnunication unit (CMU) 23.
  • PNT physical network topology storage unit
  • CU control unit
  • CMU comnnunication unit
  • the physical network topology storage unit (PNT) 21 stores a physical network topology.
  • the control unit (CU) 22 controls the processing of the controller (CTR) 20.
  • the comnnunication unit (CMU) 23 relays communications with other devices.
  • the control unit (CU) 22 may be implemented by a central processing unit (CPU) of a computer which operates according to a program.
  • the control unit (CU) 22 may be implemented by dedicated hardware.
  • the control unit (CU) 22 creates one tree using MTTE’s original initial tree creating algorithm by referring to the physical network topology storage unit (PNT) 21.
  • PNT physical network topology storage unit
  • the control unit (CU) 22 adds the tree’s topology into its tree set TS.
  • the control unit (CU) 22 removes the tree’s topology out of its tree set TS.
  • the control unit (CU) 22 Each time the tree set TS is changed, the control unit (CU) 22 gives the new tree set an unique tree-set-ID tsid. Tsid can be generated as, for example, a 128-bit random number.
  • the control unit (CU) 22 sends the new tree set and the tree-set-id tsid to all routers (RTR) 20 via the communication unit (CMU) 22.
  • the producer (PDR) 30 sends a content in chunks.
  • the session’s transmission quality is mainly decided by the delivering rate of Interests and chunks.
  • the producer (PDR) 30 decides a optimal chunkrate of each content.
  • the producer (PDR) 30 publishes high priority and low priority contents using names in formats of “/HPC/content-id/chunk-rate/chunk-id” and “/LPC/content-id/chunk-rate/chunk-id”, respectively.
  • the /HPC/ part is called the priority-tag field, where HPC (LPC) represents high (low) priority contents.
  • HPC low priority contents.
  • the /HPC/content-id/ (/LPC/content-id/) part is called the prefix of names.
  • Each packet has tree-set-id field in its header.
  • the present invention assumes that Interest (chunks) delivered in the network all have an equal size INTEREST-SIZE (CHUNKSIZE).
  • CSR consumer
  • the consumer (CSR) 40 requests a content
  • the consumer (CSR) 40 first sends an chunk-rate-request message to the producer (PDR) 30 to obtain the content’s chunk-rate.
  • the consumer (CSR) 40 then forms the first Interest of this session.
  • the Interest name’s tree-set-id field is set to NULL.
  • the chunk-rate field is set to be the received chunk-rate.
  • the chunk-id field is set to 0.
  • Fig. 4 depicts an explanatory diagram illustrating the router (RTR) 10.
  • the router (RTR) 10 includes a storage unit 11, a control unit 12, and a network interface (NIF) 13.
  • the control unit 12 controls the processing of the router (RTR) 10.
  • the control unit 12 may be implemented by a central processing unit (CPU) of a computer which operates according to a program.
  • the control unit 12 may be implemented by dedicated hardware.
  • the storage unit 11 stores FIBs (FIB) 111 and a reservation table 112, and includes a session register 113. As described above, in the FIB 111, the destination of packets is defined. In the present exemplary embodiment, the storage unit 11 stores FIBs (FIB) 111 for each tree-set-id tsid. Fig. 5 depicts an explanatory diagram illustrating the FIBs.
  • the router (RTR) 10 maintains multiple sets of FIBs.
  • the control unit 12 creates a new FIB according to the topology of the new tree set.
  • the FIB is created in the same way of MTTE.
  • FIBs[tsid] is used to denote the FIB created for tree-set-id tsid on router RTR
  • FIBs[tsid][p] is used to denote the outgoing interface in FIBs[tsid] for a name prefix p.
  • the reservation table 112 is used for managing a bandwidth reservation. The contents of the reservation table 112 will be described later.
  • the session register 113 stores the received tsid in its local storage as the latesttree-set-id LTSID.
  • the network interface (NIF) 13 sends and receives packets.
  • the router (RTR) 10 has two network interfaces (NIF) 13.
  • the number of the network interface (NIF) 13 is not limited to two.
  • the incoming capacity and outgoing capacity of the network interface (NIF) 13 are equal.
  • the network interface (NIF) 13 includes one or more different priority queues 14.
  • the network interface (NIF) 13 includes two queues 14.
  • Each router (RTR) 10 maintains an HPC queue and a LPC queue on each of its interfaces (NIF) 13. High and low priority packets are piped in the HPC and the LPC queues, respectively. The router (RTR) 10 does not forward packets in the LPC queue before the HPC queue is empty.
  • the router (RTR) 10 maintains a reservaion-table (RT) 112 for managing reservation.
  • the router (RTR) 10 uses names’ prefixes to distinguish sessions.
  • Fig. 6 depicts an explanatory diagram illustrating the reservaion-table.
  • reservaion-table (RT) has an entry that contains three elements: 1) reserved-o: the capacity that this interface has reserved for sessions on the outgoing direction, 2) reserved-i: the capacity that this interface has reserved for sessions on the incoming direction, and 3) A table called session register (SR).
  • Each entry of SR is a mapping between an Interest prefix (i.e., a session) and a three-tuple (request-o, request-i, and expiration-time).
  • Request-o is the bandwidth requirement for forwarding Interests of this prefix
  • request-i is the bandwidth requirement for forwarding returned chunks of this prefix
  • expiration-time represents the deadline at which the router (RTR) 10 tries to release reservation for this prefix.
  • Fig. 7 depicts an explanatory work flow when a router (RTR) 10 receives an Interest.
  • RT[outFace].SR does not contain an entry for i.prefix
  • the control unit 12 tries to make reservation for i.
  • Fig. 8 depicts an explanatory process of reservation.
  • Capacity(outFace) represents the capacity of outFace.
  • RT[outFace].reserved-i + request-i > Capacity(outFace) the reservation fails, too.
  • the control unit 12 still needs to notify its neighbor router (RTR) 10 of the new reservation.
  • RTR1 be the router on the other side of outFace.
  • the control unit 12 informs RTR1 that it has reserved certain bandwidth on the change between them, and sends (request-o, request-i) to RTR1.
  • Let inFace be the interface on which RTR1 receives RTR’s notification.
  • the control unit 12 then puts i into outFace’s HPC queue for forwarding.
  • control unit 12 sends a notification message back to the consumer (CSR) 40.
  • CSR consumer
  • the producer (PDR) 30 When the producer (PDR) 30 receives the Interest, the producer (PDR) 30 extracts the tree-set-id of the Interest. The producer (PDR) 30 forms a chunk with the treeset-id being the extracted tree-set-id. Then, the producer (PDR) 30 issues the chunk via the interface through which the Interest comes.
  • the consumer (CSR) 40 When the consumer (CSR) 40 receives the chunk, the consumer (CSR) 40 finishes the session if the received chunk is the last one of the requested content. Otherwise, the consumer (CSR) 40 forms a new Interest with the treeset-id field be tsid and the chunk-id field be cid.
  • tsid is equal to the tree-set-id of the received chunk
  • cid is the chunk-id of the received chunk plus one.
  • the consumer (CSR) 40 sends out the Interest via the interface via which the chunk is received.
  • Fig. 9 depicts a flowchart illustrating an operation example of the network system according to the present exemplary embodiment. This flowchart shows a name-based bandwidth reservation mechanism for MTTE.
  • Each router maintains multiple sets of FIBs (Step S11). Each packet described in this flow chart has a tree-set-id field. Each time the tree set changes, the controller gives the new tree set an unique id (Step S12), and dispatches the new tree set and the new tree-set-id to all routers (Step S13). When a router receives the update from the controller, the router creates a new FIB according to the new tree set (Step S14), and stores the new tree-set-id as its latest-tree-set-id (Step S15).
  • Step S16 When a router receives an Interest, if the tree-set-id field of the Interest is NULL (Yes in Step S16), the router assigns the latesttree-set-id to the Interest (Step S17). If the tree-set-id field of the Interest is NULL (No in Step S16), processing of step S18 is performed.
  • a producer When a producer receives an Interest, the producer forms a chunk with tree-set-id equal to the Interest’s tree-set-id (Step S19), and returns the chunk to routers (Step S20).
  • Step S22 When the consumer receives a chunk, and if the consumer has not yet received all chunks (No in step 21), the consumer forms an Interest with tree-set-id equal to the chunk’s tree-set-id (Step S22), and sends out the Interest (Step 23). If the consumer has already received all chunks (Yes in step 21), the process is completed.
  • the delay of high priority traffics and energy consumption are suppressed in CCN networks.
  • energy consumption of CCN networks is reduced, and better transmission quality for high priority traffics is provided in CCN networks.
  • Fig. 10 depicts a block diagram illustrating the outline of a system for bandwidth reservation for MTTE in CCN networks according to the present invention.
  • the system according to the present invention includes one or more routers 81 (the router 10, for example), a controller 82 (the controller 20, for example) which controls the each router 81, a producer 83 (the producer 30, for example) which sends a content in chunks, and a consumer 84 (the consumer 40, for example) which requests a content.
  • Each router 81 maintains multiple sets of FIBs (the FIB 111, for example). Each packet has a tree-set-id field. Each time the tree set changes, the controller 82 gives a new tree set an unique id, dispatches the new tree set and the id as a new tree-set-id to all routers 81. When the router 81 receives an update of the tree set from the controller 82, the router 81 creates a new FIB according to the new tree set, and stores the new tree-set-id as the router’s latest-tree-set-id.
  • the router 81 assigns the latesttree-set-id to the Interest if the tree-set-id field of the Interest is NULL.
  • the router 81 forwards the packet to an interface that packet's prefix of FIB corresponding to the packet's tree-set-id matches.
  • the producer 83 receives an Interest, the producer 81 forms a chunk with tree-set-id equal to the Interest’s tree-set-id, and returns the chunk to routers 81.
  • the consumer 84 receives a chunk and has not yet received all chunks, the consumer 84 forms an Interest with tree-set-id equal to the chunk’s tree-set-id, and sends out the Interest.
  • the router 81 may include a plurality of queues for each priority, and forward packets in a lower priority queue after a higher priority queue is empty.
  • the producer may publish different priority contents using names which include a priority-tag representing a priority.
  • the router 81 may add an Interest into the each queue according to the priority indicated by the priority-tag.
  • router 11 strorage unit 12 control unit 13 network interface 14 queue 111 FIB (Forwarding Information Base) 112 reservation table 113 session register 20 controller 21 physical network topology storage unit 22 control unit 23 comnnunication unit 30 producer 40 consumer

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

Le système selon la présente invention porte sur une ingénierie de trafic à base d'arbre multiple dans un réseautage centré sur le contenu (CCN) et comprend un ou plusieurs routeurs, un dispositif de commande qui commande chaque routeur, un producteur qui envoie un contenu en segments et un consommateur qui demande un contenu. Chaque routeur maintient (S11) de multiples ensembles de bases d'informations d'acheminement (FIB). Chaque paquet possède un champ d'identifiant d'ensemble d'arbres (tree-set-id). Chaque instant temporel où l'ensemble d'arbres change, le dispositif de commande donne (S12) à un nouvel ensemble d'arbres un nouvel identifiant unique, expédie (S13) le nouvel ensemble d'arbres et l'identificateur en tant que nouvel tree-set-id à tous les routeurs. Lorsque le routeur reçoit une mise à jour de l'ensemble d'arbres en provenance du dispositif de commande, le routeur crée (S14) une nouvelle FIB selon le nouvel ensemble d'arbres, et stocke (S15) le nouveau tree-set-id en tant que dernier tree-set-id du routeur. Lorsque le routeur reçoit un Intérêt, le routeur affecte (S17) le dernier tree-set-id à l'Intérêt si le champ tree-set-id de l'Intérêt en NUL (S16). Lorsque le routeur reçoit un paquet, le routeur transfère (S18) le paquet vers une interface qui correspond au préfixe de FIB du paquet correspondant au tree-set-id du paquet. Lorsque le producteur reçoit un Intérêt, le producteur forme (S20) un segment avec un tree-set-id égal au tree-set-id de l'Intérêt, et renvoie (S20) le segment à des routeurs. Lorsque le consommateur reçoit un segment et n'a pas encore reçu tous les segments (S21), le consommateur forme (S22) un Intérêt avec un tree-set-id égal au tree-set-id du segment, et envoie (S23) vers l'extérieur l'Intérêt. Chaque paquet transporte ainsi un identifiant (ID) de la route qu'il utilise à l'origine. Des routeurs transfèrent le paquet selon l'ID de route jusqu'à ce que la session se termine. Cela permet à des routeurs d'émettre des Intérêts de chaque session le long de la même route – la route sur laquelle une bande passante pour cette session est réservée. Une session est utilisée pour indiquer le processus selon lequel le consommateur reçoit tous les segments. Chaque Intérêt appartient à une certaine session. Chaque routeur maintient également une table de réservation. Lorsqu'un routeur reçoit un Intérêt, le routeur tente de réserver une bande passante pour cette session selon le taux de segments intégré dans le nom de l'Intérêt. Le routeur maintient cette réservation pendant une période. Si le routeur ne reçoit pas l'Intérêt suivant de cette session avant un certain délai, le routeur libère la réservation.
PCT/JP2015/002213 2015-04-23 2015-04-23 Procédé et système pour une réservation de bande passante pour une ingénierie de trafic à base d'arbre multiple dans des réseaux centrés sur le contenu WO2016170559A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2015/002213 WO2016170559A1 (fr) 2015-04-23 2015-04-23 Procédé et système pour une réservation de bande passante pour une ingénierie de trafic à base d'arbre multiple dans des réseaux centrés sur le contenu

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2015/002213 WO2016170559A1 (fr) 2015-04-23 2015-04-23 Procédé et système pour une réservation de bande passante pour une ingénierie de trafic à base d'arbre multiple dans des réseaux centrés sur le contenu

Publications (1)

Publication Number Publication Date
WO2016170559A1 true WO2016170559A1 (fr) 2016-10-27

Family

ID=53189109

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2015/002213 WO2016170559A1 (fr) 2015-04-23 2015-04-23 Procédé et système pour une réservation de bande passante pour une ingénierie de trafic à base d'arbre multiple dans des réseaux centrés sur le contenu

Country Status (1)

Country Link
WO (1) WO2016170559A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107835129A (zh) * 2017-10-24 2018-03-23 重庆大学 内容中心网络边缘节点势能增强路由方法
WO2020039056A1 (fr) * 2018-08-24 2020-02-27 Koninklijke Kpn N.V. Réseautage centré sur l'information sur des réseaux 5g ou ultérieurs
CN111294394A (zh) * 2020-01-19 2020-06-16 扬州大学 基于复杂网络交汇点的自适应缓存策略
US11558491B2 (en) 2018-08-24 2023-01-17 Koninklijke Kpn N.V. Information-centric networking over 5G or later networks

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140172981A1 (en) * 2012-12-13 2014-06-19 Research & Business Foundation Sungkyunkwan University Peer-to-peer communication method in content centric network environment
WO2015048773A2 (fr) * 2013-09-30 2015-04-02 Northeastern University Système et procédé destinés à l'envoi et à la mise en cache dynamiques conjoints dans le contexte des réseaux de diffusion de contenu

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140172981A1 (en) * 2012-12-13 2014-06-19 Research & Business Foundation Sungkyunkwan University Peer-to-peer communication method in content centric network environment
WO2015048773A2 (fr) * 2013-09-30 2015-04-02 Northeastern University Système et procédé destinés à l'envoi et à la mise en cache dynamiques conjoints dans le contexte des réseaux de diffusion de contenu

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
L. XU; T. YAGYU: "Multiple-tree based online traffic engineering for energy ,efficient content centric networking", IA2013-78, January 2014 (2014-01-01)
L. ZHANG; D. ESTRIN; J. BURKE; V. JACOBSON; J. D. THORNTON; D. K. SMETTERS; B. ZHANG; G. TSUDIK; D. MASSEY; C. PAPADOPOULOS ET AL.: "Named data networking (NDN) project", RELATORIO TECNICO NDN-0001, XEROX PALO ALTO RESEARCH CENTER-PARC, 2010
LING XU ET AL: "Multiple Tree-Based Online Traffic Engineering for Energy Efficient Content-Centric Networking", AFIN 2014: THE SIXTH INTERNATIONAL CONFERENCE ON ADVANCES IN FUTURE INTERNET, 16 November 2014 (2014-11-16), Lisbon, Portugal, pages 79 - 85, XP055240413, ISBN: 978-1-61208-377-3, Retrieved from the Internet <URL:https://www.thinkmind.org/index.php?view=article&articleid=afin_2014_3_30_40038> [retrieved on 20160112] *
M. ZHANG; C. YI; B. LIU; B. ZHANG: "GreenTE: Power-aware traffic engineering", NETWORK PROTOCOLS (ICNP), 2010 18TH IEEE INTERNATIONAL CONFERENCE, 2010, pages 21 - 30

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107835129A (zh) * 2017-10-24 2018-03-23 重庆大学 内容中心网络边缘节点势能增强路由方法
CN107835129B (zh) * 2017-10-24 2020-06-02 重庆大学 内容中心网络边缘节点势能增强路由方法
WO2020039056A1 (fr) * 2018-08-24 2020-02-27 Koninklijke Kpn N.V. Réseautage centré sur l'information sur des réseaux 5g ou ultérieurs
US11503498B2 (en) 2018-08-24 2022-11-15 Koninkluke Kpn N.V. Information-centric networking over 5G or later networks
US11558491B2 (en) 2018-08-24 2023-01-17 Koninklijke Kpn N.V. Information-centric networking over 5G or later networks
CN111294394A (zh) * 2020-01-19 2020-06-16 扬州大学 基于复杂网络交汇点的自适应缓存策略

Similar Documents

Publication Publication Date Title
EP1883181B1 (fr) Procede et appareil pour calculer un chemin d&#39;acces dans un domaine de reseau
CN101496348B (zh) 用于标签交换数据流量的多径转发的技术
CN100596102C (zh) 最小化路径抢占代价的标签交换路径建立方法
US7710869B1 (en) Packet routing to reduce susceptibility to disturbances
US9270598B1 (en) Congestion control using congestion prefix information in a named data networking environment
CN100512209C (zh) 分布式路径计算中解决资源冲突的方法
US7602702B1 (en) Fast reroute of traffic associated with a point to multi-point network tunnel
JP4357537B2 (ja) 配信経路制御装置
CN108924054B (zh) 一种多优先级的跨域资源预约集成服务保障方法
JP2004048146A (ja) リクエストルーティングネットワーク、リクエストルータ装置、ルータ装置及びネットワークにおけるパス設定方法
JPH11163854A (ja) データ通信方法
WO2016170559A1 (fr) Procédé et système pour une réservation de bande passante pour une ingénierie de trafic à base d&#39;arbre multiple dans des réseaux centrés sur le contenu
WO2017096564A1 (fr) Mccn d&#39;architecture de routage centralisé basé sur un contenu
WO2017193569A1 (fr) Procédé d&#39;établissement de trajet et contrôleur
CN105357124B (zh) 一种MapReduce带宽优化方法
CN102355421A (zh) 一种lsp网络拥塞处理的方法、装置及***
CN101272347B (zh) 负载分担的方法和设备
US8743685B2 (en) Limiting transmission rate of data
CN101515892B (zh) 一种对等网络中实现多资源处理的方法、设备及***
US20230291679A1 (en) Dynamic use of a packet recovery mechanism to avoid congestion along a network path
Chaitanya et al. Load distribution using multipath-routing in wired packet networks: A comparative study
EP4250669A1 (fr) Procédé de retransmission de paquet, dispositif électronique et support de stockage
US8798050B1 (en) Re-optimization of loosely routed P2MP-TE sub-trees
JP2002077257A (ja) ストリーム配信ネットワークサービス方法およびシステム
US11882340B2 (en) Content distribution system, unicast multicast converter, content distribution method and content distribution program

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15723312

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15723312

Country of ref document: EP

Kind code of ref document: A1