CN103491112A - Large-scale cluster task scheduling method, large-scale cluster task scheduling system and CDN node - Google Patents
Large-scale cluster task scheduling method, large-scale cluster task scheduling system and CDN node Download PDFInfo
- Publication number
- CN103491112A CN103491112A CN201210189888.9A CN201210189888A CN103491112A CN 103491112 A CN103491112 A CN 103491112A CN 201210189888 A CN201210189888 A CN 201210189888A CN 103491112 A CN103491112 A CN 103491112A
- Authority
- CN
- China
- Prior art keywords
- client
- cdn node
- node
- cdn
- address
- 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.)
- Pending
Links
Images
Abstract
The invention provides a large-scale cluster task scheduling method, a large-scale cluster task scheduling system and a CDN node. The method comprises the following steps: a client sends a (domain name system) DNS query request to a DNS; the DNS performs scheduling according to local DNS configuration of the client and returns specified content delivery network (CDN) node information to the client; the client is accessed to a specified CDN node according to the specified CDN node information; the specified CDN node judges whether the access request of the client belongs to the service of the node according to an address of the client; and the specified CDN node schedules the client to a CDN node to which the client belongs through temporary jumping according to the address of the client and a preset scheduling strategy if the access request of the client does not belong to the service of the node through judgment. According to the method provided by the embodiment of the invention, precise scheduling can be achieved, a judging step is added at the CDN node, no central scheduling machine needs to be added, and cost brought by a central scheduling machine can be saved.
Description
Technical field
The present invention relates to Internet technical field, relate in particular to a kind of large-scale cluster method for scheduling task, system and CDN node.
Background technology
As shown in Figure 1, the CDN of prior art (Content Delivery Network, content distributing network) adopt DNS (Domain Name System or Domain Name Service when GSLB, the computer domain name system) scheduling, pass through the IP address ownership of user's local DNS, be assigned to specific CDN node.The problem that prior art exists is, due to DNS can only obtain the user local DNS the IP address and can't obtain user's IP address, can not guarantee that user configured local DNS holds water, therefore can't precisely dispatch in user configured local DNS mistake simultaneously.
Can't carry out the precisely problem of scheduling in order to solve above-mentioned DNS, as shown in Figure 2, now technology is solved by HTTP 302 scheduling modes, be that DNS does not carry out the intelligent domain name analysis, but all requests all are resolved on a central dispatching machine, this central dispatching machine operation HTTP server, obtain IP address, then obtained IP address is judged, obtain the user location, re-use 302 functions of HTTP, the user is navigated to specified node, owing to being dispatched by IP address, more accurate than the DNS scheduling.Prior art also has problems, and the central dispatching machine adopts more than once 302 redirects of 302 scheduling, and the processing time is elongated, has increased user's delay, and simultaneously owing to having increased the central dispatching machine, corresponding cost also increases, such as needs maintenance centre dispatcher etc.
Summary of the invention
The present invention is intended at least one of solve the problems of the technologies described above.
For this reason, first purpose of the present invention is to propose a kind ofly can realize precisely scheduling and cost-saving large-scale cluster method for scheduling task.
Second purpose of the present invention is to propose a kind of large-scale cluster task scheduling system.
The 3rd purpose of the present invention is to propose a kind of CDN node.
To achieve these goals, the large-scale cluster method for scheduling task of embodiment comprises the following steps according to a first aspect of the invention: client sends the DNS inquiry request to domain name system DNS; Described DNS is dispatched according to the local dns configuration of described client, and returns to given content distributing network CDN nodal information to described client; Described client is according to the described appointment CDN node of described appointment CDN nodal information access; Described appointment CDN node judges whether to belong to the service of this node according to the address of described client; And if judgement do not belong to the service of this node, described appointment CDN node is dispatched to CDN node described client under according to the address of described client and default scheduling strategy by described client by interim redirect.
Large-scale cluster method for scheduling task according to the embodiment of the present invention, local dns in client configures in inaccurate situation, the appointment CDN node returned by domain name system DNS can be dispatched to the CDN node under client by the user by interim redirect, can realize accurate scheduling thus, increase determining step at the CDN node simultaneously, without increasing the central dispatching machine, can save the cost that the central dispatching machine brings.
To achieve these goals, the large-scale cluster task scheduling system of embodiment comprises according to a second aspect of the invention: client, domain name system DNS, the one CDN node and the 2nd CDN node, wherein, described client, for to described DNS, sending the DNS inquiry request, and the described CDN nodal information returned according to described DNS accesses a described CDN node, and when a described CDN node does not belong to the service of this node according to the judgement of the address of described client, access described the 2nd CDN node that a described CDN node is dispatched according to address and the default scheduling strategy of described client by interim redirect, described DNS, dispatched for the configuration of the local dns according to described client, and returned to a described CDN nodal information to described client, a described CDN node, for according to the address of described client, judging whether to belong to the service of this node, and, when judgement does not belong to the service of this node, by interim redirect, according to the address of described client and default scheduling strategy, described client is dispatched to described the 2nd CDN node.
Large-scale cluster task scheduling system according to the embodiment of the present invention, local dns in client configures in inaccurate situation, the CDN node returned by domain name system DNS can be dispatched to the 2nd CDN node by the user by interim redirect, it is the CDN node under client, can realize accurate scheduling thus, increase determining step at a CDN node simultaneously, without increasing the central dispatching machine, can save the cost that the central dispatching machine brings.
To achieve these goals, the CDN node of embodiment comprises according to a third aspect of the invention we: receiver module, for receiving the access request of client; Judge module, judge for the address according to described client whether the access request of described client belongs to the service of this node; Scheduler module, for do not belong to this node when service in judgement, be dispatched to CDN node described client under according to the address of described client and default scheduling strategy by described client by interim redirect.
According to the CDN node of the embodiment of the present invention, at the local dns of client, configure in inaccurate situation, can the user be dispatched to the CDN node under client by interim redirect by judge module and scheduler module, can realize accurate scheduling thus.
Additional aspect of the present invention and advantage part in the following description provide, and part will become obviously from the following description, or recognize by practice of the present invention.
The accompanying drawing explanation
Above-mentioned and/or additional aspect of the present invention and advantage are from obviously and easily understanding becoming the description of embodiment in conjunction with following accompanying drawing, wherein:
Fig. 1 is the schematic diagram that the CDN of prior art adopts the DNS scheduling;
Fig. 2 is the schematic diagram that the CDN of prior art adopts HTTP 302 scheduling;
Fig. 3 is the schematic diagram of large-scale cluster method for scheduling task according to an embodiment of the invention;
Fig. 4 is the flow chart of large-scale cluster method for scheduling task according to an embodiment of the invention;
Fig. 5 is the flow chart of large-scale cluster method for scheduling task according to an embodiment of the invention;
Fig. 6 is the structured flowchart of large-scale cluster task scheduling system according to an embodiment of the invention;
Fig. 7 is the structured flowchart of CDN node according to an embodiment of the invention;
Fig. 8 is the structured flowchart of CDN node according to an embodiment of the invention; And
Fig. 9 is the structured flowchart of CDN node according to an embodiment of the invention.
Embodiment
Below describe embodiments of the invention in detail, the example of described embodiment is shown in the drawings, and wherein same or similar label means same or similar element or the element with identical or similar functions from start to finish.Be exemplary below by the embodiment be described with reference to the drawings, only for explaining the present invention, and can not be interpreted as limitation of the present invention.
In description of the invention, it will be appreciated that, term " first ", " second " etc. are only for describing purpose, and can not be interpreted as indication or hint relative importance.In description of the invention, it should be noted that, unless otherwise clearly defined and limited, term " is connected ", " connection " should do broad understanding, for example, can be to be fixedly connected with, and can be also to removably connect, or connects integratedly; Can be mechanical connection, can be also to be electrically connected to; Can be directly to be connected, also can indirectly be connected by intermediary.For the ordinary skill in the art, can concrete condition understand above-mentioned term concrete meaning in the present invention.In addition, in description of the invention, except as otherwise noted, the implication of " a plurality of " is two or more.
With reference to following description and accompanying drawing, these and other aspects of embodiments of the invention will be known.These describe and accompanying drawing in, specifically disclose some specific implementations in embodiments of the invention, mean to implement some modes of the principle of embodiments of the invention, still should be appreciated that the scope of embodiments of the invention is not limited.On the contrary, embodiments of the invention comprise spirit and interior all changes, modification and the equivalent of intension scope that falls into additional claims.
The logic and/or the step that mean or otherwise describe at this in flow chart, for example, can be considered to the sequencing list of the executable instruction for realizing logic function, may be embodied in any computer-readable medium, use for instruction execution system, device or equipment (as the computer based system, comprise that the system of processor or other can and carry out the system of instruction from instruction execution system, device or equipment instruction fetch), or use in conjunction with these instruction execution systems, device or equipment.With regard to this specification, " computer-readable medium " can be anyly can comprise, storage, communication, propagation or transmission procedure be for instruction execution system, device or equipment or the device that uses in conjunction with these instruction execution systems, device or equipment.
Below with reference to Figure of description, the large-scale cluster method for scheduling task according to the embodiment of the present invention is described.
A kind of large-scale cluster method for scheduling task comprises the following steps: client sends the DNS inquiry request to domain name system DNS; DNS is dispatched according to the local dns configuration of client, and returns to given content distributing network CDN nodal information to client; Client is according to specifying the access of CDN nodal information to specify the CDN node; Specify the CDN node to judge whether to belong to the service of this node according to the address of client; And if judgement do not belong to the service of this node, specify the CDN node according to the address of client and default scheduling strategy, client to be dispatched to the CDN node under client by interim redirect.
The schematic diagram of the large-scale cluster method for scheduling task that Fig. 3 is one embodiment of the invention, the flow chart of the large-scale cluster method for scheduling task that Fig. 4 is one embodiment of the invention.
As shown in Figure 3 and Figure 4, according to the large-scale cluster method for scheduling task of the embodiment of the present invention, comprise the steps.
Step S101, client sends the DNS inquiry request to domain name system DNS.
Particularly, at first the user is in the client network address that for example input will be accessed in browser, and it is the DNS inquiry request that the network address that then client is inputted according to the user sends the domain name mapping request to domain name system DNS.
Step S102, DNS is dispatched according to the local dns configuration of client, and returns to given content distributing network CDN nodal information to client.
Particularly, the DNS inquiry request that DNS sends according to client can be obtained the local dns configuration of client, then specify corresponding CDN node according to the local dns configuration of client, the nearest CDN node of chosen distance client for example, and return to the Query Result of the IP address of this appointment CDN node as this request to client.
Step S103, client is according to specifying the access of CDN nodal information to specify the CDN node.
Particularly, the IP address of the appointment CDN node that client acquisition DNS returns, then initiate access request according to the IP address of specifying the CDN node to appointment CDN node, accesses this appointment CDN node.
Step S104, specify the CDN node to judge whether to belong to the service of this node according to the address of client.
Particularly, when in client, according to appointment CDN nodal information, for example the CDN node is specified in the access of IP address, this appointment CDN node can obtain the IP address of client, and this appointment CDN node judges according to the IP address of obtained client whether client belongs to the service of this appointment CDN node.
Step S105, judgement does not belong to the service of this node, specifies the CDN node according to the address of client and default scheduling strategy, client to be dispatched to the CDN node under client by interim redirect.
In one embodiment of the invention, specify the affiliated CDN node of CDN node and client to there is identical scheduling strategy.
In one embodiment of the invention, add mark in interim redirect and show that with the CDN node under client client accesses by interim redirect.
In one embodiment of the invention, the IP address that the affiliated CDN node of client is used in interim redirect.
Particularly, interim redirect can be 302 redirects of HTTP.When specifying the CDN node not belong to the service of appointment CDN node according to client ip judgement client, specify the CDN node according to the address of client and default scheduling strategy, client to be dispatched to the CDN node under client by 302 redirects, and in 302 redirects, the IP address of the CDN node under client is served as a mark, to the CDN node under client, show that client passes through 302 redirects accesses.
In specific implementation process of the present invention, there will be the phenomenon of endless loop while carrying out 302 redirect.Particularly, specify the affiliated CDN node of CDN node and client to there is identical scheduling strategy, for example the user in Hebei is dispatched to the Shijiazhuang node, but when carrying out each CDN node updates scheduling strategy, because numerous CDN nodes can't precisely upgrade simultaneously, there will be in the inconsistent situation of scheduling strategy of two CDN nodes sometime, for example the scheduling strategy of Shijiazhuang node is updated to the user in Hebei is dispatched to the Zhengzhou node, but the scheduling strategy of Zhengzhou node remains the user in Hebei is dispatched to the Shijiazhuang node, can be between Shijiazhuang and Zhengzhou node when the user in Hebei asks unlimited carry out 302 redirects, until the Zhengzhou node has also upgraded scheduling strategy.The IP address of the CDN node under use client in 302 redirects can prevent that as added mark the phenomenon of endless loop from appearring in 302 redirects, more specifically, the IP of the CDN node when in 302 redirect under the use client, and do not use domain name, when the user asks for the second time, the host domain name of asking (Host territory) is exactly the IP address, each CDN node is after the request for the second time that receives the user, whether the host domain name (Host territory) that at first judgement is asked is the IP address, if be, do not carry out redirect, avoid thus the endless loop phenomenon of 302 redirects.In addition, for normal directly IP access, also can think that the user wishes to access this node, and not wish to be dispatched by DNS, be also to belong to rational so do not carry out redirect.
Large-scale cluster method for scheduling task according to the embodiment of the present invention, local dns in client configures in inaccurate situation, the appointment CDN node returned by domain name system DNS can be dispatched to the CDN node under client by the user by interim redirect, can realize accurate scheduling thus, increase determining step at the CDN node simultaneously, without increasing the central dispatching machine, can save the cost that the central dispatching machine brings.
The flow chart of the large-scale cluster method for scheduling task that Fig. 5 is one embodiment of the invention.
As shown in Figure 5, according to the large-scale cluster method for scheduling task of the embodiment of the present invention, comprise the steps.
Step S201, client sends the DNS inquiry request to domain name system DNS.
Particularly, at first the user is in the client network address that for example input will be accessed in browser, and it is the DNS inquiry request that the network address that then client is inputted according to the user sends the domain name mapping request to domain name system DNS.
Step S202, DNS is dispatched according to the local dns configuration of client, and returns to given content distributing network CDN nodal information to client.
Particularly, the DNS inquiry request that DNS sends according to client can be obtained the local dns configuration of client, then specify corresponding CDN node according to the local dns configuration of client, the nearest CDN node of chosen distance client for example, and return to the Query Result of the IP address of this appointment CDN node as this request to client.
Step S203, client is according to specifying the access of CDN nodal information to specify the CDN node.
Particularly, the IP address of the appointment CDN node that client acquisition DNS returns, then initiate access request according to the IP address of specifying the CDN node to appointment CDN node, accesses this appointment CDN node.
Step S204, specify the CDN node to judge whether to belong to the service of this node according to the address of client.
Particularly, when in client, according to appointment CDN nodal information, for example the CDN node is specified in the access of IP address, this appointment CDN node can obtain the IP address of client, and this appointment CDN node judges according to the IP address of obtained client whether client belongs to the service of this appointment CDN node.
Step S205, judgement does not belong to the service of this node, specifies the CDN node according to the address of client and default scheduling strategy, client to be dispatched to the CDN node under client by interim redirect.
In one embodiment of the invention, specify the affiliated CDN node of CDN node and client to there is identical scheduling strategy.
In one embodiment of the invention, add mark in interim redirect and show that with the CDN node under client client accesses by interim redirect.
In one embodiment of the invention, the IP address that the affiliated CDN node of client is used in interim redirect.
Particularly, interim redirect can be 302 redirects of HTTP.When specifying the CDN node not belong to the service of appointment CDN node according to client ip judgement client, specify the CDN node according to the address of client and default scheduling strategy, client to be dispatched to the CDN node under client by 302 redirects, and in 302 redirects, the IP address of the CDN node under client is served as a mark, to the CDN node under client, show that client passes through 302 redirects accesses.
In specific implementation process of the present invention, there will be the phenomenon of endless loop while carrying out 302 redirect.Particularly, specify the affiliated CDN node of CDN node and client to there is identical scheduling strategy, for example the user in Hebei is dispatched to the Shijiazhuang node, but when carrying out each CDN node updates scheduling strategy, because numerous CDN nodes can't precisely upgrade simultaneously, there will be in the inconsistent situation of scheduling strategy of two CDN nodes sometime, for example the scheduling strategy of Shijiazhuang node is updated to the user in Hebei is dispatched to the Zhengzhou node, but the scheduling strategy of Zhengzhou node remains the user in Hebei is dispatched to the Shijiazhuang node, can be between Shijiazhuang and Zhengzhou node when the user in Hebei asks unlimited carry out 302 redirects, until the Zhengzhou node has also upgraded scheduling strategy.The IP address of the CDN node under use client in 302 redirects can prevent that as added mark the phenomenon of endless loop from appearring in 302 redirects, more specifically, the IP of the CDN node when in 302 redirect under the use client, and do not use domain name, when the user asks for the second time, the host domain name of asking (Host territory) is exactly the IP address, each CDN node is after the request for the second time that receives the user, whether the host domain name (Host territory) that at first judgement is asked is the IP address, if be, do not carry out redirect, avoid thus the endless loop phenomenon of 302 redirects.In addition, for normal directly IP access, also can think that the user wishes to access this node, and not wish to be dispatched by DNS, be also to belong to rational so do not carry out redirect.
Step S206, if judgement belongs to the service of this node, specify the CDN node to provide service for client.
Particularly, if judgement belongs to the service of this node, illustrate that this locality configuration of client is correct, specify the CDN node to provide service for client.
Large-scale cluster method for scheduling task according to the embodiment of the present invention, local dns in client configures in inaccurate situation on the one hand, the appointment CDN node returned by domain name system DNS can be dispatched to the CDN node under client by the user by interim redirect, can realize accurate scheduling thus.On the other hand, under the local dns of client configures accurately situation, the CDN node directly provides service, without additionally having more once useless interim redirect, saves time.Increase determining step at the CDN node simultaneously, without increasing the central dispatching machine, can save the cost that the central dispatching machine brings.
Below with reference to Figure of description, the large-scale cluster task scheduling system according to the embodiment of the present invention is described.
A kind of large-scale cluster task scheduling system, comprise: client, domain name system DNS, a CDN node and the 2nd CDN node, wherein, client, for to DNS, sending the DNS inquiry request, and CDN nodal information access the one CDN node returned according to DNS, and, when a CDN node does not belong to the service of this node according to the address judgement of client, access the 2nd CDN node that a CDN node is dispatched according to address and the default scheduling strategy of client by interim redirect; DNS, dispatched for the configuration of the local dns according to client, and returned to a CDN nodal information to client; The one CDN node, judge whether to belong to the service of this node for the address according to client, and, when judgement does not belong to the service of this node, by interim redirect, according to the address of client and default scheduling strategy, client is dispatched to the 2nd CDN node.
The structured flowchart of the large-scale cluster task scheduling system that Fig. 6 is one embodiment of the invention.
As shown in Figure 6, comprise client 110, domain name system DNS 120, a CDN node 130 and the 2nd CDN node 140 according to the large-scale cluster task scheduling system of the embodiment of the present invention.
Particularly, client 110 is for sending the DNS inquiry request to DNS 120, and CDN node 130 information access the one CDN node 130 returned according to DNS 120, and, when a CDN node 130 does not belong to the service of this node according to the judgement of the address of client 110, access the 2nd CDN node 140 that a CDN node 130 is dispatched according to address and the default scheduling strategy of client 110 by interim redirect.DNS 120 is dispatched for the configuration of the local dns according to client 110, and returns to CDN node 130 information to client 110.The one CDN node 130 judges whether to belong to the service of this node for the address according to client 110, and, when judgement does not belong to the service of this node, by interim redirect, according to the address of client 110 and default scheduling strategy, client is dispatched to the 2nd CDN node 140.Wherein, interim redirect can be 302 redirects of HTTP.
In one embodiment of the invention, a CDN node 130 and the 2nd CDN node 140 have identical scheduling strategy.
In one embodiment of the invention, add mark to show that to the 2nd CDN node 140 client 110 accesses by interim redirect in interim redirect.
In one embodiment of the invention, the IP address that the 2nd CDN node 140 is used in interim redirect.
More specifically, at first the user is in client 110 network address that for example input will be accessed in browser, and it is the DNS inquiry request that the network address that client 110 is inputted according to the user sends the domain name mapping request to DNS 120.Then, the DNS inquiry request that DNS 120 sends according to client 110 can be obtained the local dns configuration of client 110, specify corresponding CDN node according to the local dns configuration of client 110, the nearest CDN node 130 of chosen distance client 110 for example, and return to the Query Result of the IP address of a CDN node 130 as this request to client 110.The IP address of the CDN node 130 that client 110 acquisition DNS 120 return, and initiate access request to access a CDN node 130 according to the IP address of a CDN node 130 to a CDN node 130.When client 110 access the one CDN node 130, the one CDN node 130 can obtain the IP address of client 110, the one CDN node 130 judges according to the IP address of obtained client 110 whether client 110 belongs to the service of this node, when judgement client 110 does not belong to the service of this node, the one CDN node 130 is dispatched to two CDN node 140 according to the address of client 110 and default scheduling strategy by client 110 by 302 redirects, wherein the 2nd CDN node 140 is the CDN node under client, and in 302 redirects, the IP address of the 2nd CDN node 140 is served as a mark, show that to the 2nd CDN node 140 client 110 is by 302 redirect accesses.
In specific implementation process of the present invention, system there will be the phenomenon of endless loop in carrying out 302 redirects.Particularly, the one CDN node 130 and the 2nd CDN node 140 have identical scheduling strategy, for example the user in Hebei is dispatched to the Shijiazhuang node, but when carrying out each CDN node updates scheduling strategy, because numerous CDN nodes can't precisely upgrade simultaneously, there will be in the inconsistent situation of scheduling strategy of two CDN nodes sometime, for example the scheduling strategy of Shijiazhuang node is updated to the user in Hebei is dispatched to the Zhengzhou node, but the scheduling strategy of Zhengzhou node remains the user in Hebei is dispatched to the Shijiazhuang node, can be between Shijiazhuang and Zhengzhou node when the user in Hebei asks unlimited carry out 302 redirects, until the Zhengzhou node has also upgraded scheduling strategy.When using the IP address of the 2nd CDN node 140 can prevent that as added mark the phenomenon of endless loop from appearring in 302 redirects in 302 redirects, more specifically, use the IP of the 2nd CDN node 140 when in 302 redirect, and do not use domain name, when the user asks for the second time, the host domain name of asking (Host territory) is exactly the IP address, each CDN node is after the request for the second time that receives the user, whether the host domain name (Host territory) that at first judgement is asked is the IP address, if be, do not carry out redirect, avoid thus the endless loop phenomenon of 302 redirects.In addition, for normal directly IP access, also can think that the user wishes to access this node, and not wish to be dispatched by DNS, be also to belong to rational so do not carry out redirect.
Large-scale cluster task scheduling system according to the embodiment of the present invention, local dns in client configures in inaccurate situation, the CDN node returned by domain name system DNS can be dispatched to the 2nd CDN node by the user by interim redirect, it is the CDN node under client, can realize accurate scheduling thus, increase determining step at a CDN node simultaneously, without increasing the central dispatching machine, can save the cost that the central dispatching machine brings.
In one embodiment of the invention, a CDN node 130 also provides service for belong to this node when service in judgement for client.Particularly, if judgement belongs to the service of this node, illustrate that this locality configuration of client 110 is correct, a CDN node 130 provides service for client 110.
According to the large-scale cluster task scheduling system of the embodiment of the present invention, under the local dns of client configures accurately situation, a CDN node directly provides service, without additionally having more once useless interim redirect, saves time.
Below with reference to Figure of description, the CDN node according to the embodiment of the present invention is described.
A kind of CDN node comprises: receiver module, for receiving the access request of client; Judge module, judge for the address according to client whether the access request of client belongs to the service of this node; Scheduler module, for do not belong to this node when service in judgement, be dispatched to CDN node client under according to the address of client and default scheduling strategy by client by interim redirect.
The structured flowchart of the CDN node that Fig. 7 is one embodiment of the invention.
As shown in Figure 7, comprise receiver module 210, judge module 220 and scheduler module 230 according to the CDN node of the embodiment of the present invention.
Particularly, receiver module 210 is for receiving the access request of client.Wherein, the received access request of receiver module 210 is the access request that CDN node IP that client is returned according to domain name system DNS is initiated.
Judge module 220 judges for the address according to client whether the access request of client belongs to the service of this node.When receiver module 210 receives the access request of client, obtain accordingly the IP address of client, then judge module 220 judges according to the IP address of client whether the access request of client belongs to the service of this node.
Scheduler module 230 is for do not belong to this node when service in judgement, by interim redirect, according to the address of client and default scheduling strategy, client is dispatched to the CDN node under client.Wherein, interim redirect can be HTTP 302 redirects.
According to the CDN node of the embodiment of the present invention, at the local dns of client, configure in inaccurate situation, can the user be dispatched to the CDN node under client by interim redirect by judge module and scheduler module, can realize accurate scheduling thus.
The structured flowchart of the CDN node that Fig. 8 is one embodiment of the invention.
As shown in Figure 8, comprise receiver module 210, judge module 220, scheduler module 230 and mark module 240 according to the CDN node of the embodiment of the present invention.
Particularly, receiver module 210 is for receiving the access request of client.Judge module 220 judges for the address according to client whether the access request of client belongs to the service of this node.Scheduler module 230 is for do not belong to this node when service in judgement, by interim redirect, according to the address of client and default scheduling strategy, client is dispatched to the CDN node under client.Mark module 240 shows that with the CDN node under client client accesses by interim redirect in interim redirect, adding mark.Wherein, historical redirect is HTTP 302 redirects.
In one embodiment of the invention, the IP address that the affiliated CDN node of client is used in interim redirect.
In one embodiment of the invention, the CDN node under CDN node and client has identical scheduling strategy.
In specific implementation process of the present invention, there will be the phenomenon of endless loop in carrying out 302 redirects.Particularly, node under CDN node and client has identical scheduling strategy, for example the user in Hebei is dispatched to the Shijiazhuang node, but when carrying out each CDN node updates scheduling strategy, because numerous CDN nodes can't precisely upgrade simultaneously, there will be in the inconsistent situation of scheduling strategy of two CDN nodes sometime, for example the scheduling strategy of Shijiazhuang node is updated to the user in Hebei is dispatched to the Zhengzhou node, but the scheduling strategy of Zhengzhou node remains the user in Hebei is dispatched to the Shijiazhuang node, can be between Shijiazhuang and Zhengzhou node when the user in Hebei asks unlimited carry out 302 redirects, until the Zhengzhou node has also upgraded scheduling strategy.When the IP address by the CDN node under mark module 240 use clients in 302 redirects can prevent that as added mark the phenomenon of endless loop from appearring in 302 redirects, more specifically, the IP of the CDN node when in 302 redirect under the use client, and do not use domain name, when the user asks for the second time, the host domain name of asking (Host territory) is exactly the IP address, each CDN node is after the request for the second time that receives the user, whether the host domain name (Host territory) that at first judgement is asked is the IP address, if be, do not carry out redirect, avoid thus the endless loop phenomenon of 302 redirects.
CDN node according to the embodiment of the present invention, can prevent endless loop by mark module.
The structured flowchart of the CDN node that Fig. 9 is one embodiment of the invention.
As shown in Figure 9, comprise receiver module 210, judge module 220, scheduler module 230, mark module 240 and service module 250 according to the CDN node of the embodiment of the present invention.
Particularly, receiver module 210 is for receiving the access request of client.Judge module 220 judges for the address according to client whether the access request of client belongs to the service of this node.Scheduler module 230 is for do not belong to this node when service in judgement, by interim redirect, according to the address of client and default scheduling strategy, client is dispatched to the CDN node under client.Mark module 240 shows that with the CDN node under client client accesses by interim redirect in interim redirect, adding mark.When serving for belong to this node in judgement, service module 250 provides service for client.
In one embodiment of the invention, the IP address that the affiliated CDN node of client is used in interim redirect.Wherein, historical redirect is HTTP 302 redirects.
According to the CDN node of the embodiment of the present invention, under the local dns of client configures accurately situation, the CDN node directly provides service by service module, without additionally having more once useless interim redirect, saves time.
Should be appreciated that each several part of the present invention can realize with hardware, software, firmware or their combination.In the above-described embodiment, a plurality of steps or method can realize with being stored in memory and by software or the firmware of suitable instruction execution system execution.For example, if realize with hardware, the same in another embodiment, can realize by any one in following technology well known in the art or their combination: have for data-signal being realized to the discrete logic of the logic gates of logic function, application-specific integrated circuit (ASIC) with suitable combinational logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) etc.
In the description of this specification, the description of reference term " embodiment ", " some embodiment ", " example ", " concrete example " or " some examples " etc. means to be contained at least one embodiment of the present invention or example in conjunction with specific features, structure, material or the characteristics of this embodiment or example description.In this manual, the schematic statement of above-mentioned term not necessarily referred to identical embodiment or example.And the specific features of description, structure, material or characteristics can be with suitable mode combinations in any one or more embodiment or example.
Although illustrated and described embodiments of the invention, those having ordinary skill in the art will appreciate that: in the situation that do not break away from principle of the present invention and aim can be carried out multiple variation, modification, replacement and modification to these embodiment, scope of the present invention is limited by claim and equivalent thereof.
Claims (15)
1. a large-scale cluster method for scheduling task, is characterized in that, comprises the following steps:
Client sends the DNS inquiry request to domain name system DNS;
Described DNS is dispatched according to the local dns configuration of described client, and returns to given content distributing network CDN nodal information to described client;
Described client is according to the described appointment CDN node of described appointment CDN nodal information access;
Described appointment CDN node judges whether to belong to the service of this node according to the address of described client; And
If judgement does not belong to the service of this node, described appointment CDN node is dispatched to CDN node described client under according to the address of described client and default scheduling strategy by described client by interim redirect.
2. large-scale cluster method for scheduling task as claimed in claim 1, is characterized in that, the CDN node under described appointment CDN node and described client has identical scheduling strategy.
3. large-scale cluster method for scheduling task as claimed in claim 1, is characterized in that, adds mark and show that with the CDN node under described client described client is by described interim redirect access in described interim redirect.
4. large-scale cluster method for scheduling task as described as claim 1-3 any one, is characterized in that, the IP address of the CDN node under described client is used in described interim redirect.
5. large-scale cluster method for scheduling task as described as claim 1-4 any one, is characterized in that, also comprises:
If judgement belongs to the service of this node, described appointment CDN node provides service for described client.
6. a large-scale cluster task scheduling system, is characterized in that, comprises client, domain name system DNS, a CDN node and the 2nd CDN node, wherein,
Described client, for to described DNS, sending the DNS inquiry request, and the described CDN nodal information returned according to described DNS accesses a described CDN node, and, when a described CDN node does not belong to the service of this node according to the judgement of the address of described client, access described the 2nd CDN node that a described CDN node is dispatched according to address and the default scheduling strategy of described client by interim redirect;
Described DNS, dispatched for the configuration of the local dns according to described client, and returned to a described CDN nodal information to described client;
A described CDN node, for according to the address of described client, judging whether to belong to the service of this node, and, when judgement does not belong to the service of this node, by interim redirect, according to the address of described client and default scheduling strategy, described client is dispatched to described the 2nd CDN node.
7. large-scale cluster task scheduling system as claimed in claim 1, is characterized in that, a described CDN node and described the 2nd CDN node have identical scheduling strategy.
8. large-scale cluster task scheduling system as claimed in claim 1, is characterized in that, adds mark to show that to described the 2nd CDN node described client is by described interim redirect access in described interim redirect.
9. large-scale cluster task scheduling system as described as claim 6-8 any one, is characterized in that, the IP address of described the 2nd CDN node is used in described interim redirect.
10. large-scale cluster task scheduling system as described as claim 6-9 any one, is characterized in that, a described CDN node also provides service for belong to this node when service in judgement for described client.
11. a CDN node, is characterized in that, comprising:
Receiver module, for receiving the access request of client;
Judge module, judge for the address according to described client whether the access request of described client belongs to the service of this node;
Scheduler module, for do not belong to this node when service in judgement, be dispatched to CDN node described client under according to the address of described client and default scheduling strategy by described client by interim redirect.
12. CDN node as claimed in claim 11, is characterized in that, the CDN node under described CDN node and described client has identical scheduling strategy.
13. CDN node as claimed in claim 11, is characterized in that, also comprises:
Mark module, show that with the CDN node under described client described client is by described interim redirect access for add mark in described interim redirect.
14. CDN node as described as claim 11-13, is characterized in that, the IP address of the CDN node under described client is used in described interim redirect.
15. CDN node as described as claim 11-14, is characterized in that, also comprises:
Service module, provide service for described client while serving for belong to this node in judgement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210189888.9A CN103491112A (en) | 2012-06-08 | 2012-06-08 | Large-scale cluster task scheduling method, large-scale cluster task scheduling system and CDN node |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210189888.9A CN103491112A (en) | 2012-06-08 | 2012-06-08 | Large-scale cluster task scheduling method, large-scale cluster task scheduling system and CDN node |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103491112A true CN103491112A (en) | 2014-01-01 |
Family
ID=49831070
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210189888.9A Pending CN103491112A (en) | 2012-06-08 | 2012-06-08 | Large-scale cluster task scheduling method, large-scale cluster task scheduling system and CDN node |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103491112A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105045866A (en) * | 2015-07-14 | 2015-11-11 | 深圳市云宙多媒体技术有限公司 | Short-URL query-based CDN node hopping method and system |
| CN105704171A (en) * | 2014-11-25 | 2016-06-22 | 北京奇虎科技有限公司 | System and method for realizing content delivery network (CDN) access |
| CN106453675A (en) * | 2016-09-19 | 2017-02-22 | 网宿科技股份有限公司 | 302 jump method, method and system for generating URL, and method and system for resolving domain name resolution request |
| CN106453671A (en) * | 2015-08-07 | 2017-02-22 | 中国电信股份有限公司 | Method, client, server and system for CDN scheduling |
| CN109639533A (en) * | 2019-01-08 | 2019-04-16 | 深圳市网心科技有限公司 | A kind of test method, system and the terminal device of subdispatch strategy |
| CN112671664A (en) * | 2020-12-04 | 2021-04-16 | 新浪网技术(中国)有限公司 | CDN scheduling system and method based on refined scheduling |
| CN113259413A (en) * | 2021-04-16 | 2021-08-13 | 卓望数码技术（深圳）有限公司 | CDN scheduling method, scheduling system and storage medium |
| US11184318B2 (en) | 2016-09-19 | 2021-11-23 | Wangsu Science & Technology Co., Ltd. | 302 redirecting method, URL generating method and system, and domain-name resolving method and system |
| CN114222168A (en) * | 2021-12-02 | 2022-03-22 | 上海哔哩哔哩科技有限公司 | Resource scheduling method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020116481A1 (en) * | 2001-02-20 | 2002-08-22 | Raymond Lee | Content delivery network system and method for building the same |
| CN101202684A (en) * | 2006-12-15 | 2008-06-18 | 中国电信股份有限公司 | Method for close judgment of user in content distributing network |
-
2012
- 2012-06-08 CN CN201210189888.9A patent/CN103491112A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020116481A1 (en) * | 2001-02-20 | 2002-08-22 | Raymond Lee | Content delivery network system and method for building the same |
| CN101202684A (en) * | 2006-12-15 | 2008-06-18 | 中国电信股份有限公司 | Method for close judgment of user in content distributing network |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105704171A (en) * | 2014-11-25 | 2016-06-22 | 北京奇虎科技有限公司 | System and method for realizing content delivery network (CDN) access |
| CN105704171B (en) * | 2014-11-25 | 2020-07-28 | 北京奇虎科技有限公司 | System and method for realizing CDN access |
| CN105045866A (en) * | 2015-07-14 | 2015-11-11 | 深圳市云宙多媒体技术有限公司 | Short-URL query-based CDN node hopping method and system |
| CN106453671A (en) * | 2015-08-07 | 2017-02-22 | 中国电信股份有限公司 | Method, client, server and system for CDN scheduling |
| US11184318B2 (en) | 2016-09-19 | 2021-11-23 | Wangsu Science & Technology Co., Ltd. | 302 redirecting method, URL generating method and system, and domain-name resolving method and system |
| CN106453675A (en) * | 2016-09-19 | 2017-02-22 | 网宿科技股份有限公司 | 302 jump method, method and system for generating URL, and method and system for resolving domain name resolution request |
| CN109639533A (en) * | 2019-01-08 | 2019-04-16 | 深圳市网心科技有限公司 | A kind of test method, system and the terminal device of subdispatch strategy |
| CN112671664A (en) * | 2020-12-04 | 2021-04-16 | 新浪网技术(中国)有限公司 | CDN scheduling system and method based on refined scheduling |
| CN112671664B (en) * | 2020-12-04 | 2022-08-19 | 新浪网技术(中国)有限公司 | CDN scheduling system and method based on refined scheduling |
| CN113259413A (en) * | 2021-04-16 | 2021-08-13 | 卓望数码技术（深圳）有限公司 | CDN scheduling method, scheduling system and storage medium |
| 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 |
| CN114222168A (en) * | 2021-12-02 | 2022-03-22 | 上海哔哩哔哩科技有限公司 | Resource scheduling method and system |
| CN114222168B (en) * | 2021-12-02 | 2024-03-12 | 上海哔哩哔哩科技有限公司 | Resource scheduling method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103491112A (en) | Large-scale cluster task scheduling method, large-scale cluster task scheduling system and CDN node | |
| CN102291447B (en) | Content distribution network load scheduling method and system | |
| US8037025B2 (en) | Method and approach to hosting versioned web services | |
| CN102968442A (en) | Network map tile service aggregation method and system | |
| CN104079630A (en) | Business server side load balancing method, client side, server side and system | |
| CN101989951A (en) | Method, system and equipment for selecting service node | |
| CN102497409B (en) | A kind of method of cloud computing system resource management | |
| CN102281190A (en) | Networking method for load balancing apparatus, server and client access method | |
| US20180176098A1 (en) | Messaging Queue Spinning Engine | |
| CN103544285A (en) | Data loading method and device | |
| US20130148596A1 (en) | Resource management system and method of centralized base station in mobile communication network | |
| CN100525378C (en) | Management method, system and device to update distributed set top box | |
| CN104750544B (en) | Applied to the process management system and process management method in distributed system | |
| CN105656964B (en) | The implementation method and device of data-pushing | |
| CN104869162A (en) | Method of resource downloading and apparatus thereof | |
| US20140164479A1 (en) | Smart redirection and loop detection mechanism for live upgrade large-scale web clusters | |
| CN104537045A (en) | Service distribution method and device based on distributive system | |
| CN103186536A (en) | Method and system for scheduling data shearing devices | |
| CN105530323B (en) | File upgrading method, related equipment and system | |
| CN102075584B (en) | Distributed file system and access method thereof | |
| CN103685576A (en) | Interaction method for CDN (Content Delivery Network) nodes and CDN management system | |
| EP2212786A1 (en) | Method and device for agile computing | |
| US20120226702A1 (en) | Data Query System and Constructing Method Thereof and Corresponding Data Query Method | |
| CN103259824A (en) | Method and device for synchronizing data route in distributed data storage system | |
| CN103685405B (en) | Network service system and the method updating firmware thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140101 |
|
| RJ01 | Rejection of invention patent application after publication |