CN102291447B - Content distribution network load scheduling method and system - Google Patents
Content distribution network load scheduling method and system Download PDFInfo
- Publication number
- CN102291447B CN102291447B CN201110223153.9A CN201110223153A CN102291447B CN 102291447 B CN102291447 B CN 102291447B CN 201110223153 A CN201110223153 A CN 201110223153A CN 102291447 B CN102291447 B CN 102291447B
- Authority
- CN
- China
- Prior art keywords
- unit
- network information
- gslb
- network
- application layer
- 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
Abstract
The present invention discloses a kind of content distribution network load scheduling method and system.Wherein in content distribution network load scheduling method, content distributing network receives the content requests that user terminal sends; The network information that GSLB unit utilizes application layer traffic Optimized Service device to provide selects local load balancing unit according to the first selection strategy; Local load balancing unit selects edge cache equipment according to the second selection strategy; Described content requests is redirected to described edge cache equipment by content distributing network, and gives described edge cache equipment by contents distribution; Described edge cache equipment according to described content requests by described contents distribution to user terminal.The network information provided due to application layer traffic Optimized Service device reflects the information such as network topology, network traffics more accurately, thus the request of user can be directed to nearest node or region by GSLB unit, be convenient to content distributing network and serve nearby for user provides better.
Description
Technical field
The present invention relates to field of network transmission, particularly relate to content distributing network load allocation method and system.
Background technology
(ContentDeliveryNetwork is called for short: CDN) technology is a kind of fringe node content being distributed to close user from central server, thus realizes the distributed method of accessing of content nearby content distributing network.CDN technology effective improves the service quality of access to content, and can support large-scale Concurrency Access.
The key technology of CDN mainly contains contents construction technology, content distribution techniques etc.
1. contents construction technology
CDN SiteServer LBS realizes the content routing function of CDN.Its effect is the optimal node content requests of user led in whole CDN.The selected of optimal node can carry out according to multiple strategy, and such as nearest, node load is the lightest, link bandwidth is high and link cost is low etc.SiteServer LBS is the core of whole CDN, and the accuracy and efficiency of load balancing directly determines efficiency and the performance of whole CDN.
Existing CDN framework as shown in Figure 1.Usual load balancing is divided into two levels: GSLB (GlobalServiceLoadBalance, be called for short: GSLB) and local load balancing (ServiceLoadBalance, abbreviation: SLB).The main object of GSLB is, in whole network range, the request of user is directed to nearest node or region.Therefore, close judgement is the major function of GSLB.SLB is generally confined in certain regional extent, its target in specific regional extent, finds an optimal node provide service, therefore, the running status such as media formats of the health of CDN node, loading condition, support is the Main Basis that local load balancing carries out decision-making.
Load balancing by accomplished in many ways, main method comprise domain name system (DomainNameSystem, be called for short: DNS), application layer be redirected, transport layer be redirected etc.
For GSLB, judging to perform close, usually adopting in two ways, a kind of is static setting, such as according to Internet protocol (InternetProtocol, the abbreviation: IP) address arranges and shows to carry out the mapping of IP address to CDN node of static state.Another kind of mode is dynamic detection, CDN node is such as allowed to detect the distance of Target IP in real time, the linear module of distance can adopt relative retention time (RelativeRetentionTime, be called for short: RRT) or jump (Hops), then comparison detection result carries out load balancing.Certainly, the mode of Static and dynamic also can integrate use.
For SLB, in order to perform effective decision-making, need the running status obtaining buffer memory (Cache) equipment in real time.The method obtained generally has two kinds, and one is active probe, and one is protocol interaction.Active probe does not have the situation of protocol interaction interface for SLB equipment and buffer memory device, initiatively initiates detection, according to returning results analysis state by orders such as ping.Another kind is protocol interaction, i.e. SLB and the buffer memory device agreement real-time exchange running state information good according to predefined, to carry out load balancing.Comparatively speaking, protocol interaction accurately and reliably, but will still not have the agreement of standard at present than detection mode, and the realization of each producer is generally only proprietary protocol, and intercommunication is more difficult.
2. content distribution techniques
Contents distribution comprises the process from content source to the buffer memory device at CDN edge.From realization, there is the content distribution techniques of two kinds of main flows: push PUSH and pull PULL.
PUSH is the technology of a kind of active distribution.Usually, PUSH is initiated by Content Management System, content is distributed to the cache node at each edge from source or center media resources bank.The agreement of distribution can adopt HTML (Hypertext Markup Language) (HyperTextTransferProtocol, abbreviation: (FileTransferProtocol is called for short: FTP) etc. HTTP)/file transfer protocol (FTP).
PULL is a kind of passive distribution technology, and PULL distribution is asked to drive by user usually.When the content that user asks does not exist or be miss on the edge cache equipment of this locality, buffer memory device starts PULL method from content source or other CDN node Real-time Obtaining contents.Under PULL mode, the distribution of content is as required.
Existing CDN is when carrying out load and distributing, usually the positional information of current network topology and node is not considered, even if or have consideration, but information source is often for oneself is collected, the network condition of reaction is inaccurate, upgrade not in time, therefore cannot serve nearby for user provides well, can not Internet resources be effectively utilized.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of content distributing network load allocation method and system, thus more can effectively utilize Internet resources, promotes the service quality of CDN.
According to an aspect of the present invention, a kind of content distribution network load scheduling method is provided, comprises:
GSLB unit in content distributing network receives the content requests that user terminal sends;
The network information that described GSLB unit utilizes application layer traffic Optimized Service device to provide, according to the local load balancing unit in the first selection strategy chosen content distributing network;
Described local load balancing unit is according to the edge cache equipment in the second selection strategy chosen content distributing network;
Described content requests is redirected to described edge cache equipment by described GSLB unit;
Contents distribution is given described edge cache equipment by content distributing network;
Described edge cache equipment according to described content requests by described contents distribution to user terminal.
According to a further aspect in the invention, provide a kind of GSLB unit, this GSLB unit comprises receiving element, selected cell and scheduling unit, wherein:
Receiving element, for receiving the content requests that user terminal sends;
Selected cell, the network information provided for utilizing application layer traffic Optimized Service device selects the local load balancing unit in content distributing network according to the first selection strategy; And indicate described local load balancing unit to select edge cache equipment in content distributing network according to the second selection strategy;
Scheduling unit, for being redirected to described edge cache equipment by described content requests.
According to a further aspect in the invention, provide a kind of application layer traffic Optimized Service device, this server comprises information receiving unit, query unit, information transmitting unit, wherein:
Information receiving unit, for receiving the query messages for inquiring about current network information that GSLB unit sends;
Query unit, for the described query messages requester network information received according to described information receiving unit;
Information transmitting unit, for sending the network information of query unit inquiry to described GSLB unit.
According to a further aspect of the invention, provide a kind of content distributing network load dispatch system, it comprises GSLB unit, application layer traffic Optimized Service device, Dispatching Unit, multiple local load balancing unit, multiple edge cache equipment, wherein:
GSLB unit, for receiving the content requests that user terminal sends, utilize the network information that application layer traffic Optimized Service device provides, the local load balancing unit in content distributing network is selected according to the first selection strategy, and indicate described local load balancing unit to select edge cache equipment in content distributing network according to the second selection strategy, described content requests is redirected to described edge cache equipment;
Dispatching Unit, for giving described edge cache equipment by contents distribution;
Application layer traffic Optimized Service device, provides the network information for the query messages inquiring about current network information to GSLB list for what send according to described GSLB cell arrangement;
Local load balancing unit, for selecting the edge cache equipment in content distributing network according to described second selection strategy;
Edge cache equipment, for the content of storage distribution unit distribution, and distributes described content according to described content requests to user terminal.
The network information that the present invention utilizes ALTO server to provide by the GSLB unit in CDN selects the SLB unit in CDN according to selection strategy, the network information provided due to ALTO server reflects the information such as network topology, network traffics more accurately, thus the request of user can be directed to nearest node or region by GSLB unit in the scope of CDN, thus CDN serves for user provides better, and efficiently utilize Internet resources nearby.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of CDN framework in prior art.
Fig. 2 is the schematic diagram of a CDN load dispatching method of the present invention embodiment.
Fig. 3 is the schematic diagram that CDN of the present invention registers an embodiment in ALTO server.
Fig. 4 is the schematic diagram of ALTO server of the present invention to a CDN informing network information change embodiment.
Fig. 5 is the schematic diagram of another embodiment of CDN load dispatching method of the present invention.
Fig. 6 is the schematic diagram of a GSLB unit of the present invention embodiment.
Fig. 7 is the schematic diagram of another embodiment of GSLB unit of the present invention.
Fig. 8 is the schematic diagram of another embodiment of GSLB unit of the present invention.
Fig. 9 is the schematic diagram of an ALTO server of the present invention embodiment.
Figure 10 is the schematic diagram of another embodiment of ALTO server of the present invention.
Figure 11 is the schematic diagram of another embodiment of ALTO server of the present invention.
Figure 12 is the schematic diagram of a CDN load dispatch system of the present invention embodiment.
Embodiment
With reference to the accompanying drawings the present invention is described more fully, exemplary embodiment of the present invention is wherein described.
Fig. 2 is the schematic diagram of a CDN load dispatching method of the present invention embodiment.As shown in Figure 2, the CDN load dispatching method of this embodiment is as follows:
Step 101, the GSLB unit in CDN receives the content requests that user terminal sends.
Step 102, GSLB unit utilizes application layer traffic optimization, and (ApplicationLayerTrafficOptimization is called for short: the network information that ALTO) server provides selects the SLB unit in CDN according to the first selection strategy.
Step 103, SLB selects the edge cache equipment in CDN according to the second selection strategy.
Step 104, content requests is redirected to edge cache equipment by GSLB unit.
Step 105, CDN by contents distribution to edge cache equipment.
Step 106, edge memory device according to content requests by contents distribution to user terminal.
Based on the content distributing network load allocation method that the above embodiment of the present invention provides, the network information utilizing ALTO server to provide by the GSLB unit in CDN selects the SLB unit in CDN according to selection strategy, the network information provided due to ALTO server reflects the information such as network topology, network traffics more accurately, thus the request of user can be directed to nearest node or region by GSLB unit in the scope of CDN, thus CDN serves for user provides better, and efficiently utilize Internet resources nearby.
ALTO technology provides a kind of based on client/server (Client/Server, be called for short: the C/S) network information service of framework, be mainly equity (peertopeer, be called for short: P2P) class application provides service, P2P application has a lot of resource providers, the selection of resource provider have a lot alternative, it needs some mechanism to help it to carry out the selection of resource provider, by obtaining network topology accurately, the selection that the information such as network traffics carry out resource provider is a good idea, current application has a lot of method can detection network topology and the information such as network traffics, but these information obtained based on the method for detection prediction are not very accurate, and the expense of these work is also larger, affect the performance of overall applicability.The object of ALTO allows effectively to cooperate between application and network provider exactly, more effective and Fair utilization Internet resources under the high-performance condition improved or keep P2P application.
Due to original ALTO information service-oriented to as if P2P business, not high to the requirement of real-time of information updating, but by ALTO Information application to CDN field, due to CDN carrying is all important service, usually the flow of its carrying is also very large, and the change of any network all should notify in CDN timely.
For ALTO server, under P2P environment, the number of ALTO client computer is very many, the difficulty and the expense that design a kind of reliable informing mechanism are very large, but in the applied environment of CDN, the number of CDN provider is relatively less, can be carried out the notice of ALTO business information by the mode of registration notification.Under this change, ALTO business is no longer a communication service based on client/server (C/S) mode, and become a kind of can initiatively and the service of ALTO client communication.
Consider the Virtual network operator often disposing ALTO business in real network, and often there is close contact in CDN provider with Virtual network operator, ALTO business may be a kind of charge operation, therefore in one particular embodiment of the present invention, the login mechanism of CDN can be off-line, CDN producer and ALTO service supplier signature service contract, switched communication address after signing a contract, the information such as port, thus the notice realizing ALTO business upgrades.In another specific embodiment of the present invention, the login mechanism of CDN can be online, and as shown in Figure 3, CDN is registered in ALTO server by registration interface.
In another specific embodiment of content distribution network load scheduling method of the present invention, the network information utilizing ALTO server to provide at GSLB unit selects the local load balancing unit in content distributing network according to the first selection strategy before, mailing address, port information are supplied to application layer traffic Optimized Service device to register by CDN.As shown in Figure 3, CDN sends registration request RegisterReq to ALTO server, and ALTO server sends registration reply RegisterRes according to registration request to CDN.
After registration process completes, ALTO server will know be hopeful to obtain mailing address, the port information of the CDN of notice in time, when the network information changes, ALTO server can notify as shown in Figure 4 one by one, as shown in Figure 4, when the network information changes, ALTO server to send a notification message NewPolicyNotify to CDN.Thus the situation of change of the network information can be supplied to CDN by ALTO server.
In another specific embodiment of content distribution network load scheduling method of the present invention, the first selection strategy and the second selection strategy comprise with nearest, the edge cache apparatus of load of user terminal is the lightest, link bandwidth is high, link cost low in strategy set at least one strategy.
Fig. 5 is the schematic diagram of another embodiment of CDN load dispatching method of the present invention.As shown in Figure 5, the CDN load dispatching method of this embodiment is as follows:
Step 201, the GSLB unit in CDN receives the content requests that user sends.
Step 202, whether GSLB unit judges data buffering stores the network information relevant to content requests in storehouse, if when data buffering storehouse is the network information relevant to content requests, perform step 204, if when data buffering storehouse has the network information relevant to content requests, perform step 203.
Step 203, whether GSLB unit judges data buffering storehouse has the network information relevant to content requests expired, if do not have expired, performs step 207, if expired, performs step 204.
Step 204, GSLB unit sends the query messages for inquiring about current network information to ALTO server.
According to another specific embodiment of the present invention, GSLB unit comprises the information such as endpoint attribute business GetEpProp of the new network map GetMap of inquiry, network cost GetCost or ALTO to the query messages that ALTO server sends.
Step 205, ALTO server provides the network information according to query messages to GSLB unit.
Step 206, the network information that GSLB unit utilizes ALTO server to provide selects the SLB unit in CDN according to the first selection strategy.Afterwards, step 208 is performed.
Step 207, GSLB unit utilizes the network information stored in data buffering storehouse to select SLB unit in CDN according to the first selection strategy.
CDN is after obtaining the network information that provides of ALTO server, by using the mapping service (mapservice) of ALTO business, can accurately awareness network topology information, network cost information and the access way of terminal and the information on services of access capability, and these information are placed in the load-balancing algorithm of GSLB, just realize the access ability nearby of user, and prioritizing selection high bandwidth, low-cost link.
Such as, the network information that ALTO server provides comprises network topological diagram and is connected cost figure with node, GSLB can calculate network cost between user and different SLB according to these information, can obtain the SLB of the Least-cost providing service for this user according to this process.Such as this user is that the SLB unit in A, CDN is respectively SLB1, SLB2 and SLB3, and the network cost between user A and SLB1 is 4, i.e. cost (A, SLB1)=4, the network cost between user A and SLB2 is 5, i.e. cost (A, SLB2)=5, network cost between user A and SLB3 is 6, i.e. cost (A, SLB3)=6, according to Least-cost principle, GSLB can select SLB1 for user A provides service.
In addition, CDN, by obtaining the endpoint attribute business (EndpointPropertyService) of ALTO, can understand the access module of user, thus can provide in the selection of fringe node and serve more targetedly.
According to another specific embodiment of the present invention, the network information that ALTO server provides by CDN is as the override information of GSLB Unit selection SLB unit.Namely GSLB unit preferentially use ALTO server to provide the network information to select suitable SLB unit.
According to the another specific embodiment of the present invention, CDN also preferentially can use oneself loading algorithm, namely first find out the lower CDN region of load in CDN by GSLB unit, then the network information applied ALTO server provide in this CDN region selects suitable SLB unit.
Step 208, SLB selects the edge cache equipment in CDN according to the second selection strategy.
Step 209, content requests is redirected to the edge cache equipment selected by GSLB.
Step 210, CDN by contents distribution to edge cache equipment.
Step 211, edge memory device according to content requests by contents distribution to user terminal.
According to another specific embodiment of the present invention, edge memory device can adopt push PUSH mode or pull PULL mode according to content requests by contents distribution to user terminal.
According to another specific embodiment of the present invention, the first selection strategy and the second selection strategy comprise with nearest, the edge cache apparatus of load of user terminal is the lightest, link bandwidth is high, link cost low in strategy set at least one strategy.
According to another specific embodiment of the present invention, the network information utilizing ALTO server to provide at GSLB unit selects the SLB unit in CDN according to selection strategy before, mailing address, port information are supplied to ALTO server to register by GSLB unit.Thus the network information just can be sent to GSLB unit according to this mailing address, port information by ALTO server.
According to another specific embodiment of the present invention, registration can adopt offline mode or online mode.
In the above-described embodiments, the network information that GSLB unit can utilize ALTO server to provide updates stored in the network information in data buffering storehouse.
According to another specific embodiment of the present invention, when CDN initialization, arrive predetermined updated time or GSLB unit receive the NewPolicyNotify that ALTO server sends time, GSLB unit can send to ALTO server the query messages obtaining new internet message, such as, inquire about the endpoint attribute business GetEpProp message etc. of new network map GetMap, network cost GetCost or ALTO.
ALTO server provides the network information according to query messages to GSLB unit, and GSLB unit utilizes this network information to upgrade the network information stored in data buffering storehouse.
Fig. 6 is the schematic diagram of a GSLB unit of the present invention embodiment.As shown in Figure 6, this GSLB unit comprises receiving element 601, selected cell 602, scheduling unit 603, wherein:
Receiving element 601, for receiving the content requests that user terminal sends;
Selected cell 602, the network information provided for utilizing ALTO server selects the SLB unit in CDN according to the first selection strategy; And indicate described SLB unit to select edge cache equipment in CDN according to the second selection strategy;
Scheduling unit 603, for being redirected to described edge cache equipment by described content requests.
Based on the GSLB unit that the above embodiment of the present invention provides, the network information provided by utilizing ALTO server selects the SLB unit in CDN according to selection strategy, the network information provided due to ALTO server reflects the information such as network topology, network traffics more accurately, thus the request of user can be directed to nearest node or region by GSLB unit in the scope of CDN, thus CDN serves nearby for user provides better.
According to another specific embodiment of the present invention, the first selection strategy and the second selection strategy comprise with nearest, the edge cache apparatus of load of user terminal is the lightest, link bandwidth is high, link cost low in strategy set at least one strategy.
After the network information that CDN acquisition ALTO server provides, by using the mapping service (mapservice) of ALTO business, whole topology of networks can be understood, and these information are placed in the load-balancing algorithm of GSLB, realize the access ability nearby of user, and prioritizing selection high bandwidth, low-cost link.
In addition, CDN, by obtaining the endpoint attribute business (EndpointPropertyService) of ALTO, can understand the access module of user, thus can provide in the selection of fringe node and serve more targetedly.
According to another specific embodiment of the present invention, the network information that ALTO server provides by selected cell 602 selects the override information of SLB as GSLB.I.e. selected cell 602 network information that preferentially uses ALTO server to provide select suitable SLB unit according to the first selection strategy.
According to the another specific embodiment of the present invention, selected cell 602 also preferentially can use oneself loading algorithm, first find out the CDN region that load in CDN is lower, then the network information applied ALTO server provide in this CDN region selects suitable SLB unit according to the first selection strategy.
Fig. 7 is the schematic diagram of another embodiment of GSLB unit of the present invention.In the figure 7, receiving element 601, selected cell 602 and scheduling unit 603 with embodiment illustrated in fig. 6 in receiving element 601, selected cell 602 identical with scheduling unit 603.In the present embodiment, GSLB unit also comprises transmitting element 701, judging unit 702, data buffering storehouse 703, wherein,
Transmitting element 701, for sending the query messages for inquiring about current network information to ALTO server.
Data buffering storehouse 703, for storing the network information that ALTO server provides.
Judging unit 702, for receive at receiving element 601 user terminal send content requests after, judge whether store the network information relevant to content requests in data buffering storehouse 703, when data buffering storehouse 703 have the network information relevant to content requests simultaneously this network information do not have expired, judging unit 702 indicates selected cell 602 directly to utilize the network information stored in data buffering storehouse 703 to select SLB unit in CDN according to the first selection strategy.
If when the network information not relevant to content requests in data buffering storehouse 703, or data buffering storehouse 703 there is the network information relevant to content requests but this network information is expired time, instruction transmitting element 701 sends query messages to ALTO server, and the network information provided by ALTO server indicating selected cell 602 to utilize receiving element 601 to receive selects the SLB unit in CDN according to the first selection strategy.
According to another specific embodiment of the present invention, mailing address, port information are supplied to ALTO server to register by transmitting element 701.After registration process completes, ALTO server will know be hopeful to obtain the Notify Address of the CDN of notice in time, when the network information changes, ALTO server can notify as shown in Figure 4 one by one, as shown in Figure 4, when the network information changes, ALTO server to send a notification message NewPolicyNotify to CDN, thus the situation of change of the network information can be supplied to CDN by ALTO server.
According to another specific embodiment of the present invention, transmitting element 701 comprises the endpoint attribute business GetEpProp message etc. of the new network map GetMap of inquiry, network cost GetCost or ALTO to the query messages that ALTO server sends.
Fig. 8 is the schematic diagram of another embodiment of GSLB unit of the present invention.In fig. 8, receiving element 601, selected cell 602, scheduling unit 603, transmitting element 701, judging unit 702, data buffering storehouse 703 with embodiment illustrated in fig. 7 in receiving element 601, selected cell 602, scheduling unit 603, transmitting element 701, judging unit 702, data buffering storehouse 703 identical.In the present embodiment, GSLB unit also comprises updating block 801, and the network information provided for utilizing ALTO server updates stored in the network information in data buffering storehouse 702.
According to another specific embodiment of the present invention, when CDN carries out initialization or reach predetermined updated time, or when receiving element 601 receives the NewPolicyNotify of ALTO server transmission, updating block 801 indicates transmitting element 701 to send query messages to ALTO server, and the network information provided by ALTO server utilizing receiving element 601 to receive updates stored in the network information in data buffering storehouse 703.
Fig. 9 is the schematic diagram of an ALTO server of the present invention embodiment.As shown in Figure 9, the ALTO server of this embodiment comprises information receiving unit 901, query unit 902 and information transmitting unit 903, wherein:
Information receiving unit 901, for receiving the query messages that GSLB unit sends.
Query unit 902, for the query messages requester network information received according to information receiving unit 901;
Information transmitting unit 903, for sending the network information that query unit 902 is inquired about to GSLB unit.
Based on the ALTO server that the above embodiment of the present invention provides, the network information utilizing ALTO server to provide by the GSLB unit in CDN selects the SLB unit in CDN according to selection strategy, the network information provided due to ALTO server reflects the information such as network topology, network traffics more accurately, thus the request of user can be directed to nearest node or region by GSLB unit in the scope of CDN, thus CDN serves for user provides better, and efficiently utilize Internet resources nearby.
Figure 10 is the schematic diagram of another embodiment of ALTO server of the present invention.As shown in Figure 10, information receiving unit 901, query unit 902 and information transmitting unit 903 and the information receiving unit 901 shown in Fig. 9, query unit 902 are identical with information transmitting unit 903.In the present embodiment, ALTO server also comprises detecting unit 1001, whether change for Sampling network information, and indication information transmitting element 903 to send a notification message NewPolicyNotify, so that the situation of change of CDN awareness network information to CDN when the network information sends change.
Figure 11 is the schematic diagram of another embodiment of ALTO server of the present invention.As shown in figure 11, information receiving unit 901, query unit 902, information transmitting unit 903 and detecting unit 1001 and the information receiving unit 901 shown in Figure 10, query unit 902, information transmitting unit 903 are identical with detecting unit 1001.In the present embodiment, ALTO server also comprises registering unit 1101, for being received log-on message that GSLB unit sends by information receiving unit 901 to register GSLB unit, and mailing address, port address that indication information transmitting element 903 can utilize log-on message to comprise send the network information or communication information to GSLB unit.
Figure 12 is the schematic diagram of CDN load dispatch system of the present invention, and as shown in figure 12, this system comprises GSLB unit 1201, ALTO server 1202, Dispatching Unit 1203, multiple SLB unit 1204, multiple edge cache equipment 1205, wherein:
GSLB unit 1201, for receiving the content requests that user terminal sends, the network information utilizing ALTO server 1202 to provide selects the SLB unit 1204 in content distributing network according to the first selection strategy, and the SLB unit that instruction is selected selects the edge cache equipment in CDN according to the second selection strategy, described content requests is redirected to described edge cache equipment.
ALTO server 1202, provides the network information for the query messages sent according to GSLB unit 1201 to GSLB unit.
Dispatching Unit 1203, for giving described edge cache equipment by contents distribution.
SLB unit 1204, for selecting the edge cache equipment in CDN according to the second selection strategy.
Edge cache equipment 1205, for the content that storage distribution unit 1203 is distributed, and distributes described content according to content requests to user terminal.
Based on the CDN load dispatch system that the above embodiment of the present invention provides, the network information utilizing ALTO server to provide by the GSLB unit in CDN selects the SLB unit in CDN according to selection strategy, the network information provided due to ALTO server reflects the information such as network topology, network traffics more accurately, thus the request of user can be directed to nearest node or region by GSLB unit in the scope of CDN, thus CDN serves for user provides better, and efficiently utilize Internet resources nearby.
According to another specific embodiment of the present invention, GSLB unit 1201 specifically can be realized by the GSLB unit of specific embodiment arbitrary shown in Fig. 6 to Fig. 8.
According to another specific embodiment of the present invention, ALTO server 1202 specifically can be realized by the ALTO server of specific embodiment arbitrary shown in Fig. 9 to Figure 11.
According to another specific embodiment of the present invention, Dispatching Unit 1203 specifically adopts the mode that pushes PUSH or pull PULL by contents distribution to user terminal.
The present invention obtains the corresponding network information by CDN from ALTO server, the maintenance and the use that achieve the network information are separated, reduce the development difficulty of contents construction algorithm in CDN, for CDN provides reliable network aware service and client perception service, CDN can be provided better serve nearby, effectively can avoid CDN alternate route in the transmission simultaneously, avoid congested section, contribute to CDN and carry out traffic balancing and system deployment.
Description of the invention provides in order to example with for the purpose of describing, and is not exhaustively or limit the invention to disclosed form.Many modifications and variations are obvious for the ordinary skill in the art.Selecting and describing embodiment is in order to principle of the present invention and practical application are better described, and enables those of ordinary skill in the art understand the present invention thus design the various embodiments with various amendment being suitable for special-purpose.
Claims (22)
1. a content distribution network load scheduling method, is characterized in that, comprising:
GSLB unit in content distributing network receives the content requests that user terminal sends;
The network information that described GSLB unit utilizes application layer traffic Optimized Service device to provide, according to the local load balancing unit in the first selection strategy chosen content distributing network;
Described local load balancing unit is according to the edge cache equipment in the second selection strategy chosen content distributing network;
Described content requests is redirected to described edge cache equipment by described GSLB unit;
Contents distribution is given described edge cache equipment by content distributing network;
Described edge cache equipment according to described content requests by described contents distribution to user terminal;
Wherein, the endpoint attribute business information that the network information that application layer traffic Optimized Service device provides comprises is sent to local load balancing unit by described GSLB unit, and local load balancing unit selects the edge cache equipment in content distributing network according to the second selection strategy and in conjunction with described endpoint attribute business information;
Wherein, after described GSLB unit receives content requests, judge whether store the network information relevant to described content requests in the data buffering storehouse in content distributing network, if described data buffering storehouse has the network information relevant to described content requests, this network information is not expired simultaneously, GSLB unit directly utilizes the network information stored in described data buffering storehouse, the local load balancing unit in content distributing network is selected according to described first selection strategy, the network information that the network information wherein stored in data buffering storehouse provides for application layer traffic Optimized Service device,
If when described data buffering storehouse is the network information not relevant to described content requests, or described data buffering storehouse there is the network information relevant to described content requests but this network information is expired time, GSLB unit sends the query messages for inquiring about current network information to application layer traffic Optimized Service device, application layer traffic Optimized Service device provides the network information according to query messages to GSLB unit, the network information that GSLB unit utilizes application layer traffic Optimized Service device to provide, the local load balancing unit in content distributing network is selected according to described first selection strategy.
2. method according to claim 1, is characterized in that,
Described first selection strategy and the second selection strategy adopt at least one strategy in strategy set, wherein strategy set comprise with user terminal nearest, edge cache apparatus of load is the lightest, link bandwidth is high, link cost is low.
3. method according to claim 1, is characterized in that,
The network information that GSLB unit utilizes application layer traffic Optimized Service device to provide updates stored in the network information in described data buffering storehouse.
4. method according to claim 3, is characterized in that,
When content distributing network initialization or arrive predetermined updated time, GSLB unit sends the query messages for inquiring about current network information to application layer traffic Optimized Service device, application layer traffic Optimized Service device provides the network information according to query messages to GSLB unit, and the network information that GSLB unit utilizes application layer traffic Optimized Service device to provide updates stored in the network information in data buffering storehouse.
5. method according to claim 3, is characterized in that,
When the notification message that the expression network information that GSLB unit receives the transmission of application layer traffic Optimized Service device changes, GSLB unit sends query messages for inquiring about current network information according to described notification message to application layer traffic Optimized Service device, application layer traffic Optimized Service device provides the network information according to query messages to GSLB unit, and the network information that GSLB unit utilizes application layer traffic Optimized Service device to provide updates stored in the network information in data buffering storehouse.
6. method according to claim 1, is characterized in that,
In the network information that GSLB unit utilizes application layer traffic Optimized Service device to provide, select the local load balancing unit in content distributing network according to the first selection strategy before, mailing address, port information are supplied to application layer traffic Optimized Service device to register by GSLB unit, and application layer traffic Optimized Service device utilizes described mailing address, port information to provide the network information to GSLB unit.
7. method according to claim 6, is characterized in that, described registration adopts offline mode or online mode.
8. method according to claim 2, is characterized in that,
The network information that described GSLB unit utilizes application layer traffic Optimized Service device to provide, comprise according to the local load balancing unit that the first selection strategy is selected in content distributing network: first GSLB unit calculates the region that in content distributing network, load is low, the network information that recycling application layer traffic Optimized Service device provides, selects the local load balancing unit in content distributing network in the region that described load is low according to described first selection strategy.
9. method according to claim 2, is characterized in that,
The network information that described GSLB unit utilizes application layer traffic Optimized Service device to provide, comprise according to the local load balancing unit that the first selection strategy is selected in content distributing network: the network information that GSLB unit utilizes application layer traffic Optimized Service device to provide, select the local load balancing unit in content distributing network according to described first selection strategy.
10. method according to claim 1, is characterized in that,
Content distributing network adopts and pushes PUSH mode or pull PULL mode by contents distribution to described edge cache equipment.
11. 1 kinds of GSLB unit, is characterized in that, this GSLB unit comprises receiving element, data buffering storehouse, judging unit, selected cell, transmitting element and scheduling unit, wherein:
Receiving element, for receiving the content requests that user terminal sends;
Data buffering storehouse, for storing the network information that application layer traffic Optimized Service device provides;
Judging unit, for after described receiving element receives described content requests, judge whether store the network information relevant to described content requests in the data buffering storehouse in content distributing network, if described data buffering storehouse has the network information relevant to described content requests, this network information is not expired simultaneously, indicate described selected cell directly to utilize the network information stored in described data buffering storehouse, select the local load balancing unit in content distributing network according to the first selection strategy; If when described data buffering storehouse is the network information not relevant to described content requests, or described data buffering storehouse there is the network information relevant to described content requests but this network information is expired time, described transmitting element is indicated to send the query messages for inquiring about current network information to application layer traffic Optimized Service device, and the network information provided by application layer traffic Optimized Service device indicating described selected cell to utilize described receiving element to receive, the local load balancing unit in content distributing network is selected according to described first selection strategy;
Selected cell, for the instruction according to judging unit, directly utilizes the network information stored in described data buffering storehouse, selects the local load balancing unit in content distributing network according to described first selection strategy; Also for the instruction according to judging unit, the network information utilizing application layer traffic Optimized Service device to provide selects the local load balancing unit in content distributing network according to the first selection strategy; And indicate described local load balancing unit to select edge cache equipment in content distributing network according to the second selection strategy;
Transmitting element, for the instruction according to judging unit, sends query messages to application layer traffic Optimized Service device, the network information that requesting query is current; The endpoint attribute business information also comprised for the network information provided by application layer traffic Optimized Service device sends to local load balancing unit, so that local load balancing unit selects the edge cache equipment in content distributing network according to the second selection strategy and in conjunction with described endpoint attribute business information;
Scheduling unit, for being redirected to described edge cache equipment by described content requests.
12. GSLB unit according to claim 11, it is characterized in that, described first selection strategy and the second selection strategy adopt at least one strategy in strategy set, wherein strategy set comprise with user terminal nearest, edge cache apparatus of load is the lightest, link bandwidth is high, link cost is low.
13. GSLB unit according to claim 11, it is characterized in that, described GSLB unit also has updating block, and the network information for utilizing application layer traffic Optimized Service device to provide updates stored in the network information in described data buffering storehouse.
14. GSLB unit according to claim 13, it is characterized in that, described updating block is also when content distributing network initialization or arrive predetermined updated time, indicate described transmitting element to send the query messages for inquiring about current network information to application layer traffic Optimized Service device, and the network information provided by application layer traffic Optimized Service device utilizing described receiving element to receive update stored in the network information in described data buffering storehouse.
15. GSLB unit according to claim 13, it is characterized in that, during the notification message that the expression network information that described updating block also receives the transmission of application layer traffic Optimized Service device at described receiving element changes, indicate described transmitting element according to described notification message to the query messages of application layer traffic Optimized Service device transmission for inquiring about current network information, and the network information provided by application layer traffic Optimized Service device utilizing described receiving element to receive updates stored in the network information in data buffering storehouse.
16. GSLB unit according to claim 11, it is characterized in that, the network information that described transmitting element utilizes application layer traffic Optimized Service device to provide at described selected cell, select the local load balancing unit in content distributing network according to the first selection strategy before, mailing address, port information are supplied to application layer traffic Optimized Service device to register.
17. GSLB unit according to claim 12, it is characterized in that, first described selected cell calculates the region that in content distributing network, load is low, and the network information that recycling application layer traffic Optimized Service device provides selects the local load balancing unit in content distributing network in the region that described load is low according to the first selection strategy.
18. GSLB unit according to claim 12, it is characterized in that, the network information that described selected cell preferentially utilizes application layer traffic Optimized Service device to provide, selects the local load balancing unit in content distributing network according to the first selection strategy.
19. 1 kinds of content distributing network load dispatch systems, is characterized in that, comprise GSLB unit, application layer traffic Optimized Service device, Dispatching Unit, multiple local load balancing unit, multiple edge cache equipment, wherein:
GSLB unit, the GSLB unit related to any one of claim 11-18;
Dispatching Unit, for giving described edge cache equipment by contents distribution;
Application layer traffic Optimized Service device, provides the network information for the query messages inquiring about current network information to GSLB unit for what send according to described GSLB cell arrangement;
Local load balancing unit, for going out the edge cache equipment in content distributing network according to described second selection strategy and in conjunction with end points attribute service Information Selection; Wherein the endpoint attribute business information that the network information that application layer traffic Optimized Service device provides comprises is sent to local load balancing unit by GSLB unit;
Edge cache equipment, for the content of storage distribution unit distribution, and distributes described content according to described content requests to user terminal.
20. systems according to claim 19, is characterized in that, Dispatching Unit specifically adopts and pushes PUSH mode or pull PULL mode by contents distribution to described edge cache equipment.
21. systems according to claim 19, is characterized in that,
Application layer traffic Optimized Service device is also for sending a notification message to GSLB unit when the network information changes, so that GSLB unit sends query messages for inquiring about current network information according to notification message to application layer traffic Optimized Service device.
22. systems according to claim 19, is characterized in that,
Application layer traffic Optimized Service device also for send according to GSLB unit log-on message time GSLB unit is registered, and the mailing address utilizing log-on message to comprise, port information provide the network information to GSLB unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110223153.9A CN102291447B (en) | 2011-08-05 | 2011-08-05 | Content distribution network load scheduling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110223153.9A CN102291447B (en) | 2011-08-05 | 2011-08-05 | Content distribution network load scheduling method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102291447A CN102291447A (en) | 2011-12-21 |
| CN102291447B true CN102291447B (en) | 2016-04-06 |
Family
ID=45337548
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110223153.9A Active CN102291447B (en) | 2011-08-05 | 2011-08-05 | Content distribution network load scheduling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102291447B (en) |
Families Citing this family (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102546774A (en) * | 2011-12-27 | 2012-07-04 | 厦门市美亚柏科信息股份有限公司 | CDN (content distribution network) routing method for secondary redirection and system |
| CN102546820B (en) * | 2012-02-16 | 2015-12-16 | 华为技术有限公司 | The storage means of transmission optimization method, map information, Apparatus and system |
| US9350671B2 (en) | 2012-03-22 | 2016-05-24 | Futurewei Technologies, Inc. | Supporting software defined networking with application layer traffic optimization |
| CN103475626B (en) * | 2012-06-07 | 2017-03-08 | 华为技术有限公司 | A kind of method for asking resource, equipment and system |
| CN104170342B (en) * | 2012-06-20 | 2018-04-20 | 华为技术有限公司 | The method and network side equipment of contents processing |
| CN102769669A (en) * | 2012-07-03 | 2012-11-07 | Ut斯达康通讯有限公司 | Intelligent content distribution control device and system |
| US20150172762A1 (en) * | 2012-07-10 | 2015-06-18 | Sharp Kabushiki Kaisha | Content transmission device, content playback device, content delivery system, control method for content transmission device, control method for content playback device, data structure, control program, and recording medium |
| CN103546516B (en) * | 2012-07-16 | 2016-12-21 | 华为技术有限公司 | Generate polymer layer network and the method and device of polymer layer expense figure |
| CN102811256B (en) * | 2012-07-19 | 2017-11-07 | 中兴通讯股份有限公司 | A kind of method and system for realizing application layer transmission optimization |
| CN102780775B (en) * | 2012-07-19 | 2017-11-07 | 中兴通讯股份有限公司 | A kind of method and system for realizing application layer transmission optimization |
| CN103595644B (en) * | 2012-08-15 | 2017-02-08 | 上海云途信息技术有限公司 | Method of realizing data permutation based on data exchange network, server and system thereof |
| CN102891804B (en) * | 2012-10-16 | 2018-08-10 | 南京中兴新软件有限责任公司 | The method of adjustment and system of control strategy |
| CN102917287A (en) * | 2012-11-21 | 2013-02-06 | 北京邮电大学 | Intelligent optical network exchange device and edge cashing method facing content center |
| CN105227517B (en) * | 2014-05-29 | 2018-09-14 | 中国电信股份有限公司 | Utilize the content distribution network system and its service implementation method of edge cache equipment |
| CN105704176B (en) * | 2014-11-26 | 2018-10-30 | 中国联合网络通信集团有限公司 | A kind of method, decision device and network architecture ensureing bandwidth and service quality |
| CN104717231B (en) * | 2014-12-18 | 2019-01-25 | 北京蓝汛通信技术有限责任公司 | The pre- distribution processing method and device of content distributing network |
| CN104618506B (en) * | 2015-02-24 | 2019-09-27 | 深圳梨享计算有限公司 | A kind of content distribution network system of crowdsourcing, method and apparatus |
| CN106302216A (en) * | 2015-05-20 | 2017-01-04 | 天脉聚源（北京）科技有限公司 | A kind of method and system of Network Load Balance |
| CN106686034B (en) * | 2015-11-10 | 2019-12-06 | 中国电信股份有限公司 | CDN scheduling enhancement method, device and system |
| CN105847353A (en) * | 2016-03-22 | 2016-08-10 | 中国科学院信息工程研究所 | Mobile CDN (content delivery network) content scheduling method and system for mobile communication network |
| CN105959354A (en) * | 2016-04-25 | 2016-09-21 | 乐视控股（北京）有限公司 | Load scheduling method and load scheduling system of streaming media service |
| CN105791150B (en) * | 2016-05-09 | 2018-11-30 | 中国联合网络通信集团有限公司 | CDN network dispositions method and system |
| CN105763386A (en) * | 2016-05-13 | 2016-07-13 | 中国工商银行股份有限公司 | Service processing system and method |
| CN105872093A (en) * | 2016-05-31 | 2016-08-17 | 乐视控股（北京）有限公司 | CDN acceleration method and system |
| CN107623580B (en) * | 2016-07-15 | 2021-06-29 | 阿里巴巴集团控股有限公司 | Task processing method, device and system in content distribution network |
| CN106412043B (en) * | 2016-09-20 | 2019-09-13 | 网宿科技股份有限公司 | CDN network flow bootstrap technique and device |
| CN106254550A (en) * | 2016-09-30 | 2016-12-21 | 杭州索浪信息技术有限公司 | A kind of method and system of content distribution |
| CN107995531B (en) * | 2016-10-26 | 2020-10-09 | 中国电信股份有限公司 | Method, device and system for transmitting IPTV video of mobile terminal |
| CN106790482B (en) * | 2016-12-13 | 2020-05-22 | 网宿科技股份有限公司 | Resource scheduling method and resource scheduling system |
| CN108270835B (en) * | 2017-01-03 | 2019-10-22 | 中国移动通信有限公司研究院 | A kind of dispatching method of business tine, device, strategic server and system |
| CN110178356B (en) * | 2017-04-26 | 2020-07-28 | 华为技术有限公司 | Redirection method, control plane network element, aggregated user plane network element and terminal equipment |
| WO2020062048A1 (en) * | 2018-09-28 | 2020-04-02 | 西门子股份公司 | Method for transmitting update data, content delivery network node and data update system |
| CN109660598B (en) * | 2018-11-17 | 2020-05-19 | 华中科技大学 | Cache replacement method and system for transient data of Internet of things |
| CN110418367B (en) * | 2019-06-14 | 2021-05-14 | 电子科技大学 | 5G forwarding network hybrid edge cache low-delay method |
| CN110336885B (en) * | 2019-07-10 | 2022-04-01 | 深圳市网心科技有限公司 | Edge node distribution method, device, scheduling server and storage medium |
| CN110830564B (en) * | 2019-10-30 | 2022-11-01 | 北京金山云网络技术有限公司 | CDN scheduling method, device, system and computer readable storage medium |
| CN113132251A (en) * | 2020-01-15 | 2021-07-16 | 中国电信股份有限公司 | Service scheduling method, device and storage medium |
| CN111432231B (en) * | 2020-04-26 | 2023-04-07 | 中移（杭州）信息技术有限公司 | Content scheduling method of edge network, home gateway, system and server |
| CN114071068A (en) * | 2020-08-04 | 2022-02-18 | 重庆艾艺荷马教育科技股份有限公司 | Non-compressed high-definition network real-time audio and video transmission equipment based on 5G |
| CN113259413B (en) * | 2021-04-16 | 2023-05-09 | 卓望数码技术（深圳）有限公司 | CDN scheduling method, scheduling system and storage medium |
| CN114598654A (en) * | 2022-01-30 | 2022-06-07 | 阿里巴巴（中国）有限公司 | Content delivery network CDN-based flow equalization processing method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101222424A (en) * | 2007-12-24 | 2008-07-16 | 中国电信股份有限公司 | Content distribution network and scheduling method based on content in the network |
| CN101242422A (en) * | 2007-02-06 | 2008-08-13 | 中国电信股份有限公司 | Content distribution network system and method applicable to multi-service fusion |
| CN101431532A (en) * | 2008-12-15 | 2009-05-13 | 中国电信股份有限公司 | Content routing method, load balancing equipment and resource management equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102012931A (en) * | 2010-12-01 | 2011-04-13 | 北京瑞信在线系统技术有限公司 | Filter cache method and device, and cache system |
-
2011
- 2011-08-05 CN CN201110223153.9A patent/CN102291447B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101242422A (en) * | 2007-02-06 | 2008-08-13 | 中国电信股份有限公司 | Content distribution network system and method applicable to multi-service fusion |
| CN101222424A (en) * | 2007-12-24 | 2008-07-16 | 中国电信股份有限公司 | Content distribution network and scheduling method based on content in the network |
| CN101431532A (en) * | 2008-12-15 | 2009-05-13 | 中国电信股份有限公司 | Content routing method, load balancing equipment and resource management equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102291447A (en) | 2011-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102291447B (en) | Content distribution network load scheduling method and system | |
| CN103023768B (en) | Edge routing node and its method from multi-source prefetching content | |
| CN106657287B (en) | Data access method and system | |
| CN102216923B (en) | Request routing and updating routing information utilizing client location information | |
| US10015243B2 (en) | Optimized content distribution based on metrics derived from the end user | |
| US9172632B2 (en) | Optimized content distribution based on metrics derived from the end user | |
| CN107517228B (en) | Dynamic acceleration method and device in content distribution network | |
| CN105162900B (en) | A kind of domain name mapping of multi-node collaboration and caching method and system | |
| EP2701363A1 (en) | Content processing method, device and system | |
| CN102047243A (en) | Request routing based on class | |
| CN101989951A (en) | Method, system and equipment for selecting service node | |
| WO2015019224A1 (en) | Dynamic edge server allocation | |
| CN101287011A (en) | Method, system and device for responding service request from user in content distributing network | |
| CN102077189A (en) | Request routing using network computing components | |
| CN101779437A (en) | Method, apparatus and system for mobility management and efficient information retrieval in a communications network | |
| CN105229993B (en) | For executing method, system and the computer-readable medium of the service routing of enhancing | |
| CN101014046B (en) | Method for integrating service location with service quality routing in service loading network | |
| Gregori et al. | Sensing the Internet through crowdsourcing | |
| US20140310375A1 (en) | Network node apparatus for information-centric networking and operating method of the network node apparatus | |
| CN105681413A (en) | Method and device for cooperative processing of data between CDN (Content Delivery Network) and ISP (Internet Service Provider) | |
| CN105024938A (en) | Mobile load balancing method and system | |
| Naeem et al. | Caching content on the network layer: A performance analysis of caching schemes in icn-based internet of things | |
| CN108769097A (en) | Support the content distribution network system of network-control | |
| Jazaeri et al. | Toward caching techniques in edge computing over SDN-IoT architecture: A review of challenges, solutions, and open issues | |
| CN103685344A (en) | Synergetic method and system for multiple P2P (point-to-point) cache peers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |