CN106533733B - The CCN collaboration caching method and device routed based on network cluster dividing and Hash - Google Patents
The CCN collaboration caching method and device routed based on network cluster dividing and Hash Download PDFInfo
- Publication number
- CN106533733B CN106533733B CN201610787121.4A CN201610787121A CN106533733B CN 106533733 B CN106533733 B CN 106533733B CN 201610787121 A CN201610787121 A CN 201610787121A CN 106533733 B CN106533733 B CN 106533733B
- Authority
- CN
- China
- Prior art keywords
- node
- cluster
- content
- network
- cache
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- 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
- H04L45/02—Topology update or discovery
-
- 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
- H04L45/46—Cluster building
-
- 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
- H04L45/74—Address processing for routing
- H04L45/742—Route cache; Operation thereof
Abstract
The present invention provides a kind of CCN collaboration caching method and device routed based on network cluster dividing and Hash, comprising: obtains the rate of discharge of each nodal information and Egress node in topological structure;Network cluster dividing model is generated according to the rate of discharge of nodal information and Egress node;It is calculated according to network cluster dividing model and obtains node cluster information;Judge to determine corresponding node cluster after Ingress node receives content requests;Cache node in corresponding cluster is obtained according to node cluster information and content requests;Content requests are forwarded to cache node in cluster, if there being the content of request in cluster in cache node, content is returned.A kind of CCN collaboration caching method and device routed based on network cluster dividing and Hash provided by the invention, by establishing network cluster dividing model to the node in topological structure, and node cluster is obtained using simulated annealing, cache node in cluster is searched out using hash algorithm again, in content requests repeating process, it realizes that routing and caching joint consider, guarantees efficient content distribution.
Description
Technical field
The present invention relates to network technique field more particularly to a kind of CCN collaboration routed based on network cluster dividing and Hash are slow
Deposit method and device.
Background technique
Growing network size and user demand brings lot of challenges to internet, wherein with multimedia content
Growth, current Generation Internet exposes it in the inefficiencies of content distribution.User usually more concerned with their demands what
Content, rather than the position of these contents, existing communication pattern end to end therewith and mismatch.How from network structure,
These challenges of protocol architecture and service mode etc. reply, will become an important directions of future network research.In content
Heart network (CCN, Content Centric Network) is more significant a solution in many researchs, it is one
The completely new network architecture based on content of kind, main difference is the name based on content, the road based on content name
By with embedded caching (In-network Caching) etc..
In CCN, the built-in store function of router, for caching the data packet (Data) of process, to accelerate other use
The response time of family access cache data packet, while the flow in network can be reduced.When a router receives user's request
Interest packet (Interest) carries out longest prefix match lookup using name, and there are three key data structures to complete forwarding: CS
(Content Store)、PIT(Pending Interest Table)、FIB(Forwarding Interest Base)。CS
Can the packet of Buffer forwarding for a long time as far as possible, for other consumers use.PIT records the interest packet forwarded, is
Allow the data packet of response to reach its requestor.FIB of the FIB similar to IP.When an interest packet arrival, first matching CS, such as
Fruit has, then responds and abandon interest packet, secondly matches PIT, finally matches FIB, abandon if not matching.When data packet arrives
It reaches, then needs to be decided whether according to cache policy to be buffered in router CS.
There are two problems demands to solve in CCN design: how router judges whether data cached packet;If router does not have
Have and meets how interest packet to neighbor node forwards the interest packet.The cache decision of CCN and forwarding decision are broadly divided into two at present
Class: along road (On-path) and it is detached from path (Off-path).Along road, caching means that data packet is stored on request path.
The cache decision LCE (Leave Copy Everywhere) of CCN default is exactly along road cache policy, when the content of request returns
When, all nodes on the way all cache the copy of content, it is clear that this mode causes to cache redundancy, reduces spatial cache and utilizes
Rate.Even if researchers propose it is some can reduce caching redundancy along road buffering scheme, for example select on path a certain
A node or node set are cached, or the priority of cache contents are determined according to content popularit, but these schemes pair
The raising of network performance is than relatively limited.In addition, interest packet forwards generally in a manner of flooding to neighbor node, request, which is hit, is
" by chance ";If joint considers the forwarding strategy that caching is placed, the retransmission protocol of some complexity is needed to interact caching section
Storage information between point, these can all reduce network performance.
On the contrary, Off-path cache decision is then according to predefined rule by content caching to specified node, interest
Packet is forwarded according to identical rule to specified node, it is clear that Off-path caching can make full use of spatial cache, significantly mention
High network hit rate.It is that wherein a solution the most outstanding, this motion are also main that Hash, which routes (Hash-routing),
Hash routing is paid close attention to, i.e., Hash route technology is introduced into CCN, content designator is mapped to by caching section by hash function
Point, and cache decision and request forwarding are united, optimize cyberspace service efficiency, maximizes cache hit rate.
Although CCN can greatly improve cache hit rate after introducing Hash route technology, this is to be surrounded by need with interest
More multi-hop to be forwarded is just hit as cost, and then leads to download link load in time delay and domain and increase.As shown in figure 4,
Assuming that couple in router R8Receive the request to content A (interest packet), usual R8It can will request along shortest path R7-R4-R9To
Egress router R10Forwarding, the content of request is obtained from respective server.However, R8Router is known by calculating hash function
R11(specified node) is responsible for cache contents A, and then request path is R7-R11.If router R11Really content A has been cached,
So content A is along path R7-R11User is returned to, otherwise, continues to request along path R4-R9To egress router R10Turn
Hair, this is referred to as path and extends (Path stretch) than jumping along shortest path forwarding is more by one.How path to be avoided to prolong
Exhibition is also to introduce Hash to route the ultimate challenge faced below.
Lorenzo Saino et al. proposes a kind of caching mechanism based on Hash routing, which depends on Hash letter
The cache node position and egress router position that number is calculated, dynamic select content return path.If cache node
That on shortest paths, it is assumed that cache node is R11, content A will be along shortest path R10-R9-R4-R7-
R8 most returns to user fastly, while also can be along reverse request path R10-R9-R15-R11 in specified cache node cache contents
Copy.
Sen Wang et al., which is proposed, divides the collaboration caching mechanism routed with Hash based on content space, passes through intelligence and draws
Divide content space, which being capable of operative constraint path extension caused by Hash routing.Therefore, it is needed when carrying out division of teaching contents
Meet worst path extension (worst path stretch, wps) and all requests of covering as much as possible.As shown in figure 4,
If wps is set as a jump, the content that R14 is produced from outlet is allocated to node R 3, R4, R5, R7, R14 (cluster-
R14), the content that R10 is produced from outlet is allocated to node R 4, R9, R10, R13 (cluster-R10)
The content return path that Lorenzo Saino et al. is proposed is asymmetric with request path, and not applicable CCN is emerging
The natural symmetric path environment of interest packet and data packet.
The constraint worst path extension that Sen Wang et al. is proposed is realized based on each path, is lacked the overall situation and is examined
Amount, resulting even in some node free time is not arranged cache contents.Also some nodes are had while belonging to multiple divisions, this
Content is frequently replaced on shared node a bit, and hit rate is caused to decline.In addition, not accounting for difference when dividing content space
The flow difference that egress router can service leads to the frequent replacement of multiplexer node spatial cache, to reduce internetworking
Energy.
Summary of the invention
The present invention provides a kind of CCN collaboration caching method and device routed based on network cluster dividing and Hash, for solving
In the prior art in network topology structure route distribution content inefficiency problem.
In a first aspect, the present invention provides a kind of CCN collaboration caching method routed based on network cluster dividing and Hash, comprising:
Obtain the topological structure of a network;
The rate of discharge of each nodal information and Egress node is obtained according to the topological structure;
Network cluster dividing model is generated according to the rate of discharge of the nodal information and the Egress node;
It is calculated according to network cluster dividing model and obtains node cluster information;
After Ingress node receives content requests, corresponding node cluster is determined according to content requests judgement;
Cache node in corresponding cluster is obtained according to node cluster information and content requests;
Content requests are forwarded to cache node in cluster, if having the content of request in cluster in cache node, content
It returns.
Preferably, the network cluster dividing model, comprising:
nj=ntj, j=1,2 ..., | ER | (5)
Wherein, ER is the Egress node set of topological structure, and IR is the Ingress node set of topological structure, and V is i-th of section
Point, ERjFor j-th of Egress node, IRkFor k-th of Ingress node, | ER | it is the size of egress router set, njIt is j-th
The size of cluster, tjFor the rate of discharge of j-th of Egress node.
Preferably, described calculated according to network cluster dividing model obtains node cluster information, comprising: is adopted according to network cluster dividing model
With simulated annealing calculate obtain node cluster information, the node cluster information include node cluster number and each node
The nodal information of cluster.
Preferably, described that cache node in corresponding cluster is obtained according to node cluster information and content requests, comprising: according to section
Point cluster information and content requests are using cache node in the corresponding cluster of hash function acquisition.
Preferably, further includes: content requests are forwarded to cache node in cluster, if there not being request in cluster in cache node
Content, then continue to Egress node forward.
Second aspect, the present invention provide a kind of CCN collaboration buffer storage routed based on network cluster dividing and Hash, comprising:
First obtains module, for obtaining the topological structure of a network;
Second obtains module, for obtaining the outlet stream of each nodal information and Egress node according to the topological structure
Amount;
Generation module, for generating network cluster dividing mould according to the rate of discharge of the nodal information and the Egress node
Type;
Computing module obtains node cluster information for calculating according to network cluster dividing model;
First choice module, for judging to determine according to the content requests after Ingress node receives content requests
Corresponding node cluster;
Second selecting module, for obtaining cache node in corresponding cluster according to node cluster information and content requests;
Forwarding module, for content requests to be forwarded to cache node in cluster, if there being request in cluster in cache node
Content then returns to content.
Preferably, the network cluster dividing model, comprising:
nj=ntj, j=1,2 ..., | ER | (5)
Wherein, ER is the Egress node set of topological structure, and IR is the Ingress node set of topological structure, and V is i-th of section
Point, ERjFor j-th of Egress node, IRkFor k-th of Ingress node, | ER | it is the size of egress router set, njIt is j-th
The size of cluster, tjFor the rate of discharge of j-th of Egress node.
Preferably, the computing module is specifically used for: being calculated and is obtained using simulated annealing according to network cluster dividing model
Node cluster information, the node cluster information include the number of node cluster and the nodal information of each node cluster.
Preferably, second selecting module is specifically used for: using hash function according to node cluster information and content requests
Obtain cache node in corresponding cluster.
Preferably, the forwarding module is also used to: content requests being forwarded to cache node in cluster, if cache node in cluster
In do not have the content of request, then continue to forward to Egress node.
As shown from the above technical solution, a kind of CCN collaboration routed based on network cluster dividing and Hash provided by the invention is slow
Method and device is deposited, by establishing network cluster dividing model to the node in topological structure, and is saved using simulated annealing
Point cluster, then cache node in cluster is searched out using hash algorithm, in content requests repeating process, realize routing and caching joint
Consider, guarantees efficient content distribution.
Detailed description of the invention
Fig. 1 is the process that the CCN routed based on network cluster dividing and Hash that the embodiment of the present invention 1 provides cooperates with caching method
Schematic diagram;
Fig. 2 is the schematic diagram of content caching provided in an embodiment of the present invention and request path;
Fig. 3 is the structure that the CCN routed based on network cluster dividing and Hash that the embodiment of the present invention 2 provides cooperates with buffer storage
Schematic diagram;
Fig. 4 is the schematic diagram of content caching and request path in the prior art.
Specific embodiment
With reference to the accompanying drawings and examples, specific embodiments of the present invention will be described in further detail.Implement below
Example is not intended to limit the scope of the invention for illustrating the present invention.
Fig. 1 is shown the embodiment of the present invention 1 and provides a kind of CCN collaboration caching side routed based on network cluster dividing and Hash
Method, comprising:
S11, the topological structure for obtaining a network.
In this step, it should be noted that the topological structure for obtaining the domain AS carries out net for the topological structure to the domain AS
Network node (router) sub-clustering.
S12, the rate of discharge that each nodal information and Egress node are obtained according to the topological structure.
In this step, it should be noted that node clustering is carried out to topological structure and needs to establish network cluster dividing model, is built
Formwork erection type first has to obtain information needed for establishment process.For a topological structure, existing each node letter
Breath can obtain.Nodal information may include the number and positional relationship of Ingress node, Egress node and intermediate node.
In addition to this, the size of cluster intuitively reflects the service ability of a cluster, and how many request flow is directly sent out
Giving some outlet means that this cluster needs that how many flow carried.
Therefore this method, which carries out sub-clustering, to realize reasonable employment network by the size according to rate of discharge dynamic adjustment cluster
The target of cache resources, therefore the rate of discharge of each Egress node need to be obtained by calculating.
S13, network cluster dividing model is generated according to the rate of discharge of the nodal information and the Egress node.
In this step, it should be noted that first have to establish network cluster dividing model, can be defined as follows the variable of table 1:
Table 1 is expressed as node variable
Variable | Definition |
G | Topological structure |
ER | Egress node set |
IR | Ingress node set |
Vi | I-th of node |
ERj | J-th of Egress node |
IRk | K-th of Ingress node |
|ER| | The size of Egress node set |
nj | The size of j-th of cluster |
tj | The rate of discharge of j-th of Egress node |
Assuming that any AS caches topological structure in CCN, indicated by non-directed graph G=(V, E), which has n vertex and m
Side.Network is divided into corresponding cluster according to outlet, i.e. decision node ViWhether belong to cluster j, be specifically expressed as follows:
Wherein node is node, and cluster is cluster.
Given one group of Egress node and Ingress node, the target of the optimization problem are will to scheme G to be divided into size limited | ER
| a cluster, at the same minimize specified node to corresponding outlet distance and specified node to all entrances distance with.That is network
Clustering Model are as follows:
nj=ntj, j=1,2 ..., | ER | (5)
Wherein, first constraint means that all nodes all have to be divided into some cluster, does not omit, will not go out in this way
Existing idle node;Second constraint then shows that a node can only belong to a cluster, otherwise on the node shared by multiple clusters
The content of caching can be replaced frequently, and cache hit rate is caused to decline;The size of each cluster of third constraint explanation is limited, formula (5)
It then further illustrates that the size of cluster is proportional to rate of discharge, and the size of cluster can be adjusted according to rate of discharge dynamic.
S14, acquisition node cluster information is calculated according to network cluster dividing model.
In this step, it should be noted that calculated according to network cluster dividing model using simulated annealing and obtain node
Cluster information, the node cluster information include the number of node cluster and the nodal information of each node cluster.
In this method, optimal dividing be defined as divide after between cache node and Ingress node and Egress node always away from
From minimum.Therefore, this motion uses simulated annealing, which is a kind of greedy algorithm, may not be able to find the overall situation
Optimal solution, but can be than finding the approximate optimal solution of problem faster, if parameter setting is proper, simulated annealing search effect
Rate is higher than the method for exhaustion.Detailed process is as follows for algorithm:
Algorithm is first originated with the sub-clustering state of one in search space given cluster size (being proportional to rate of discharge): every
Once belong to two nodes of two different clusters by exchange to select " neighbours " sub-clustering state, if new sub-clustering state
There is smaller total distance, then receives current solution;Otherwise poor solution, and this are solved than current to receive one with certain probability
A probability gradually decreases (gradually decreasing could tend to stablize) as time goes by.Therefore simulated annealing is a kind of random calculation
Method, it is possible to which the optimal solution that can jump out this part reaches global optimal solution.
It is understood that the embodiment of the present invention, which is not limited to above-mentioned algorithm, obtains global optimal solution.For can be with
Other algorithms for obtaining globally optimal solution can be used.Such as ant group algorithm, genetic algorithm.
Network cluster dividing model is solved by above-mentioned algorithm, the node cluster in topological structure, each node can be obtained
It include multiple nodes in cluster.
S15, after Ingress node receives content requests, corresponding node cluster is determined according to content requests judgement.
In this step, it should be noted that after sub-clustering, one and only one Egress node in each cluster.Therefore, from this
The content that Egress node is forwarded to content source only caches in corresponding cluster.When Ingress node is received to certain content requests interest packet
When, it will be produced first from which outlet according to the judgement of the content designator of request, and then find corresponding cluster, each section in cluster
Point has a cluster interior label.
S16, cache node in corresponding cluster is obtained according to node cluster information and content requests.
In this step, it should be noted that determine that this is interior after step S15 determines node cluster, then through hash function
Hold the specific cache location in cluster, and carries out Hash routing in cluster.Determine that the content has in cluster by hash function
The cache location of body is more mature technology, and details are not described herein.
S17, content requests are forwarded to cache node in cluster, if there being the content of request in cluster in cache node,
Content returns.
In this step, it should be noted that by Hash calculation, content designator is mapped to cache node in cluster,
Then the label of the cache node is added in interest packet prefix, directly interest packet is routed on specified node, saved along road
Point does not need to be inquired.If the cache node has the content of request, content is returned to user;Otherwise, continue to
Outlet forwarding, to obtain content from content source.
When request content enters in the domain AS from Egress node, user is returned to along reverse path, and in specified node
Upper cache contents copy.Therefore, some content caches on an at most node in cluster, if produced from different Egress nodes
Content is isomery, then some content also at most caches on one node in entire AS, so as to avoid redundancy is cached,
Spatial cache is able to more efficiently utilize.In addition, being guaranteed by Hash routing, it can be achieved that routing and caching joint consider
Efficient content distribution.
Specific explanations explanation is carried out to the above method with specific embodiment below:
By hash function, content designator is mapped to value xi:i∈[1,nj], wherein njIt is router number in cluster j.
Any hash function may be used to this programme.In order to facilitate statement, mould Hash (key mould n) is now used.Mould Hash is widely applied
In routing forwarding and cache decision.In the method, key is content designator, and n is router number in cluster.
Topology is as illustrated in fig. 2, it is assumed that Egress node R14The total flow that flow is 70%, Egress node R10Flow is remaining
30%.It can be obtained according to cluster algorithm: node R6,R9,R10,R12And R13Belong to cluster (cluster)-R10, remaining 11 nodes
Belong to cluster (cluster)-R14.Therefore, value (0,1,2,3,4) is randomly assigned to cluster-R10, value (0,1,2 ..., 10) with
Machine distributes to cluster-R14。cluster-R14Allocation result it is as shown in the table.The requiring of each Ingress node safeguards this
The allocation map table of sample.
As shown in Fig. 2, when Ingress node R8 receives the request interest packet to content A, before inquiring the request content title
Sewing, which will appreciate that, which Egress node to leave AS information from, that is, request will be gone out if being not met in AS by this
Mouthful and AS link is transmitted to corresponding contents source between verifying corresponding field.It is R14, the then cache node specified by inquiry Egress node
It should belong to cluster cluster-R14.By obtaining value 10 after Hash calculation, and 10 corresponding routers are R16 (specified to be used to delay
Deposit the node of content A), then, interest packet is directly transmitted to specified cache node R16.If content A is buffered in node
In R16, then user is directly returned to by entrance R8.Otherwise, continue to outlet R14 forwarding, to obtain content from content source.
A kind of CCN routed based on network cluster dividing and Hash that the embodiment of the present invention 1 provides cooperates with caching method, by right
Node in topological structure establishes network cluster dividing model, and obtains node cluster using simulated annealing, then use hash algorithm
Cache node in cluster is searched out, in content requests repeating process, realizes that routing and caching joint consider, guarantees efficient content
Distribution.
The CCN routed based on network cluster dividing and Hash that Fig. 3 shows the offer of the embodiment of the present invention 2 cooperates with buffer storage,
Module 21, second, which is obtained, including first obtains module 22, generation module 23, computing module 24, the choosing of first choice module 25, second
Select module 26 and forwarding module 27, in which:
First obtains module 21, for obtaining the topological structure of a network;
Second obtains module 22, for obtaining the outlet stream of each nodal information and Egress node according to the topological structure
Amount;
Generation module 23, for generating network cluster dividing mould according to the rate of discharge of the nodal information and the Egress node
Type;
Computing module 24 obtains node cluster information for calculating according to network cluster dividing model;
First choice module 25, for being judged according to the content requests true after Ingress node receives content requests
Fixed corresponding node cluster;
Second selecting module 26, for obtaining cache node in corresponding cluster according to node cluster information and content requests;
Forwarding module 27, for content requests to be forwarded to cache node in cluster, if there being request in cluster in cache node
Content, then content is returned.
In the process of implementation, first the topological structure that module 21 obtains a network is obtained.Second obtains the acquisition of module 22 should
The rate of discharge of topological structure each nodal information and Egress node.Generation module 23 goes out according to nodal information and Egress node
Mouth flow establishes network cluster dividing model.Computing module 24 calculate to network cluster dividing model and be saved using simulated annealing
Point cluster information.After Ingress node receives content requests, first choice module 25 is corresponding according to content requests judgement determination
Node cluster.Second selecting module 26 is obtained to cache in corresponding cluster and be saved according to node cluster information and content requests using hash algorithm
Point.Content requests are forwarded to cache node in cluster by forwarding module 27, if there being the content of request in cluster in cache node,
Content is returned.If there not being the content of request in cluster in cache node, continue to forward to Egress node.
Since 4 described device of the embodiment of the present invention is identical as the principle of above-described embodiment the method, in further detail
Explanation content details are not described herein.
It should be noted that can be by hardware processor (hardware processor) come real in the embodiment of the present invention
Existing related function module.
A kind of CCN routed based on network cluster dividing and Hash provided by the invention cooperates with buffer storage, by tying to topology
Node in structure establishes network cluster dividing model, and obtains node cluster using simulated annealing, then search out using hash algorithm
Cache node in cluster realizes that routing and caching joint consider, guarantees efficient content distribution in content requests repeating process.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
Meaning one of can in any combination mode come using.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability
Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch
To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame
Claim.
Those of ordinary skill in the art will appreciate that: the above embodiments are only used to illustrate the technical solution of the present invention., and
It is non-that it is limited;Although present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art
It is understood that it is still possible to modify the technical solutions described in the foregoing embodiments, either to part of or
All technical features are equivalently replaced;And these are modified or replaceed, it does not separate the essence of the corresponding technical solution this hair
Bright claim limited range.
Claims (8)
1. a kind of CCN routed based on network cluster dividing and Hash cooperates with caching method characterized by comprising
Obtain the topological structure of a network;
The rate of discharge of each nodal information and Egress node is obtained according to the topological structure;
Network cluster dividing model is generated according to the rate of discharge of the nodal information and the Egress node;
It is calculated according to network cluster dividing model and obtains node cluster information;
After Ingress node receives content requests, corresponding node cluster is determined according to content requests judgement;
Cache node in corresponding cluster is obtained according to node cluster information and content requests;
Content requests are forwarded to cache node in cluster, if there being the content of request in cluster in cache node, content is returned;
Wherein, the network cluster dividing model, comprising:
nj=ntj, j=1,2 ..., | ER | (5)
Wherein, ER is the Egress node set of topological structure, and IR is the Ingress node set of topological structure, and V is i-th of node,
ERjFor j-th of Egress node, IRkFor k-th of Ingress node, | ER | it is the size of egress router set, njFor j-th cluster
Size, tjFor the rate of discharge of j-th of Egress node.
2. the method according to claim 1, wherein described calculated according to network cluster dividing model obtains node cluster letter
Breath, comprising: calculated according to network cluster dividing model using simulated annealing and obtain node cluster information, the node cluster information includes
The nodal information of the number of node cluster and each node cluster.
3. according to the method described in claim 2, it is characterized in that, described corresponded to according to node cluster information and content requests
Cluster in cache node, comprising: obtain to cache in corresponding cluster using hash function according to node cluster information and content requests and save
Point.
4. the method according to claim 1, wherein further include: content requests are forwarded to cache node in cluster,
If there not being the content of request in cluster in cache node, continue to forward to Egress node.
5. a kind of CCN routed based on network cluster dividing and Hash cooperates with buffer storage characterized by comprising
First obtains module, for obtaining the topological structure of a network;
Second obtains module, for obtaining the rate of discharge of each nodal information and Egress node according to the topological structure;
Generation module, for generating network cluster dividing model according to the rate of discharge of the nodal information and the Egress node;
Computing module obtains node cluster information for calculating according to network cluster dividing model;
First choice module is corresponded to for judging to determine according to the content requests after Ingress node receives content requests
Node cluster;
Second selecting module, for obtaining cache node in corresponding cluster according to node cluster information and content requests;
Forwarding module, for content requests to be forwarded to cache node in cluster, if there being the content of request in cluster in cache node,
Then content is returned;
Wherein, the network cluster dividing model, comprising:
nj=ntj, j=1,2 ..., | ER | (5)
Wherein, ER is the Egress node set of topological structure, and IR is the Ingress node set of topological structure, and V is i-th of node,
ERjFor j-th of Egress node, IRkFor k-th of Ingress node, | ER | it is the size of egress router set, njFor j-th cluster
Size, tjFor the rate of discharge of j-th of Egress node.
6. device according to claim 5, which is characterized in that the computing module is specifically used for: according to network cluster dividing mould
Type is calculated using simulated annealing and obtains node cluster information, and the node cluster information includes the number of node cluster, and each
The nodal information of node cluster.
7. device according to claim 6, which is characterized in that second selecting module is specifically used for: according to node cluster
Information and content requests are using cache node in the corresponding cluster of hash function acquisition.
8. device according to claim 5, which is characterized in that the forwarding module is also used to: content requests are forwarded to
Cache node in cluster continues to forward to Egress node if there not being the content of request in cluster in cache node.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610787121.4A CN106533733B (en) | 2016-08-30 | 2016-08-30 | The CCN collaboration caching method and device routed based on network cluster dividing and Hash |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610787121.4A CN106533733B (en) | 2016-08-30 | 2016-08-30 | The CCN collaboration caching method and device routed based on network cluster dividing and Hash |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106533733A CN106533733A (en) | 2017-03-22 |
CN106533733B true CN106533733B (en) | 2019-09-27 |
Family
ID=58344862
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610787121.4A Active CN106533733B (en) | 2016-08-30 | 2016-08-30 | The CCN collaboration caching method and device routed based on network cluster dividing and Hash |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106533733B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108449277B (en) * | 2016-12-12 | 2020-07-24 | 华为技术有限公司 | Message sending method and device |
CN110611688B (en) * | 2018-06-15 | 2022-04-12 | 伊姆西Ip控股有限责任公司 | Method, electronic device and computer program product for searching nodes |
CN108900599B (en) * | 2018-06-26 | 2020-05-22 | 西安交通大学 | Software-defined content-centric network device and cluster cache decision method thereof |
CN111404972B (en) * | 2019-01-02 | 2023-04-07 | 中国移动通信有限公司研究院 | Data processing method, device, equipment and storage medium |
CN111510382B (en) * | 2020-07-01 | 2020-09-29 | 南京航空航天大学 | Real-time dynamic clustering method and system suitable for EXAta simulator |
CN112052198B (en) * | 2020-08-11 | 2022-09-20 | 大连理工大学 | Hash route cooperative caching method based on node betweenness popularity under energy consumption monitoring platform |
CN113965588B (en) * | 2021-10-18 | 2023-08-22 | 北京航空航天大学 | Content-centric-network-oriented autonomous domain collaborative caching method |
CN114189468B (en) * | 2021-11-02 | 2024-04-12 | 云端领航(北京)通信科技股份有限公司 | Multi-identification network system routing method based on identification clustering |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0793639B2 (en) * | 1989-10-30 | 1995-10-09 | インターナシヨナル・ビジネス・マシーンズ・コーポレーシヨン | Method and system for mitigating processing and recording requirements at a network node |
CN101472166B (en) * | 2007-12-26 | 2011-11-16 | 华为技术有限公司 | Method for caching and enquiring content as well as point-to-point medium transmission system |
CN105898802A (en) * | 2016-03-25 | 2016-08-24 | 西安电子科技大学 | Cluster-based caching cooperation and sharing method in heterogeneous wireless network |
-
2016
- 2016-08-30 CN CN201610787121.4A patent/CN106533733B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN106533733A (en) | 2017-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106533733B (en) | The CCN collaboration caching method and device routed based on network cluster dividing and Hash | |
Zhang et al. | A survey of caching mechanisms in information-centric networking | |
CN104753797B (en) | A kind of content center network dynamic routing method based on selectivity caching | |
US9166817B2 (en) | Load balancing | |
CN105262681B (en) | Distributed routing framework | |
US9929954B2 (en) | Hash-based overlay routing architecture for information centric networks | |
CN111245722B (en) | SDN data center network flow forwarding method based on genetic algorithm | |
Wu et al. | MBP: A max-benefit probability-based caching strategy in information-centric networking | |
CN105681438A (en) | Centralized caching decision strategy in content-centric networking | |
CN112399485A (en) | CCN-based new node value and content popularity caching method in 6G | |
Banerjee et al. | Characteristic time routing in information centric networks | |
CN103888517B (en) | Caching method for publishing and subscription system to achieve historical event subscription | |
CN103457855B (en) | Classless inter-domain routing table is established and the method and apparatus of message forwarding | |
CN106790421B (en) | ICN two-step caching method based on community | |
CN107302571B (en) | The routing of information centre's network and buffer memory management method based on drosophila algorithm | |
Alduayji et al. | PF-EdgeCache: Popularity and freshness aware edge caching scheme for NDN/IoT networks | |
CN102739550A (en) | Multi-memory flow routing architecture based on random duplication allocation | |
Jiang et al. | Sequence-preserving parallel IP lookup using multiple SRAM-based pipelines | |
Yao et al. | A pre-caching mechanism of video stream based on hidden Markov model in vehicular content centric network | |
Xu et al. | Minimizing bandwidth cost of CCN: a coordinated in-network caching approach | |
Iqbal | Cache-MCDM: A hybrid caching scheme in Mobile Named Data Networks based on multi-criteria decision making | |
Jung et al. | A multiple hash routing scheme for fast data retrieval in ICN | |
CN102739551A (en) | Multi-memory flow routing architecture | |
CN114745440B (en) | CCN cache replacement method and device | |
CN110380972B (en) | SDN joint routing selection and rule caching method based on user stream transmission cost optimization |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |