CN105245565A - Cloud system, management node and data obtaining method - Google Patents
Cloud system, management node and data obtaining method Download PDFInfo
- Publication number
- CN105245565A CN105245565A CN201510539917.3A CN201510539917A CN105245565A CN 105245565 A CN105245565 A CN 105245565A CN 201510539917 A CN201510539917 A CN 201510539917A CN 105245565 A CN105245565 A CN 105245565A
- Authority
- CN
- China
- Prior art keywords
- level storage
- computing node
- desired data
- data
- network delay
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a cloud system, a management node and a data obtaining method. The cloud system comprises the management node, at least one computing node and at least two secondary memories, wherein the secondary memories are placed at different geographical positions respectively; the management node receives a request, for needed data, of the computing node, obtains network delay between the computing node and each secondary memory, determines whether the secondary memories include the data needed by the computing node, and sends an identification of a target secondary memory to the computing node; and the computing node obtains needed data from the target secondary memory after receiving the identification. The target secondary memory comprises the data needed by the computing node, and network delay between the target secondary memory and the computing node is shortest. The data transmission speed in the cloud system can be improved.
Description
Technical field
The present invention relates to field of cloud computer technology, particularly a kind of method of cloud system, management node and data acquisition.
Background technology
Along with the fast development of computer technology, cloud service, cloud computing are widely used, and wherein cloud system is the basis of cloud service, cloud computing.Along with the continuous increase of cloud system user, the computing node of cloud system also constantly increases, and due to the randomness of cloud system user distribution, the distribution of cloud system computing node also becomes and becomes increasingly complex, and brings and huge challenge to the design of cloud system.
At present, generally arrange the memory device of cloud system in a fixing geographical position, when the user of various places needs to obtain the file in memory device, the mode that the file that obtain is transmitted by cross-region by cloud system, is transferred on the computing node of user.
When memory device and computing node distant time, network delay between memory device and computing node is larger, cause needing the longer time by the file transfer on memory device to computing node, such as when computing node creates virtual machine, the system image file creating virtual machine use is needed to be transferred to local computing node from the memory device in strange land, because system image file is general all larger, when computing node is far away apart from memory device, need the plenty of time could the transmission of completion system image file, therefore, the mode that this cloud system data store, when computing node obtains data from memory device, the transmission speed of cloud system data is slower.
Summary of the invention
The invention provides a kind of method of cloud system, management node and data acquisition, the speed of cloud system transfer of data can be improved.
Embodiments provide a kind of cloud system, comprising: management node, at least one computing node and at least two second-level storages;
Each second-level storage in described at least two second-level storages is deployed in different geographical position, for storing the Various types of data needed for described computing node;
Described management node, the request of desired data is obtained for receiving described computing node, obtain network delay between described computing node and each second-level storage described and whether each second-level storage comprises described computing node desired data described in judging, will described computing node desired data be comprised and and between described computing node the mark of the target second-level storage that network delay is the shortest send to described computing node;
Described computing node, for sending the request obtaining desired data to described management node, and after the mark receiving the described target second-level storage that described management node sends, obtains desired data from described target second-level storage.
Preferably, described management node, for according to the network delay between described computing node and each second-level storage described, according to network delay order from short to long, each second-level storage described is sorted, according to the ranking results of described second-level storage, judge whether comprise described computing node desired data in each second-level storage described one by one, and the mark first being judged the target second-level storage comprising described computing node desired data sends to described computing node.
Preferably, described management node, be further used for the ranking results according to described second-level storage, when described target second-level storage is not the second-level storage of sequence first, send long-range copy order, to be copied in the second-level storage of described sequence first by the described computing node desired data in described target second-level storage.
Preferably, described management node, be further used for every setting-up time, whether identically detect the data stored in each second-level storage described, if not, send long-range copy order, with mutually copying by data between each second-level storage described, between realization each second-level storage described, data is synchronous.
Preferably, described management node, for after the request receiving described computing node acquisition desired data, calculation command { " vcIP " is sent to each second-level storage described, " ID " }, wherein said vcIP is the mark of described computing node, and described ID is the mark of described desired data;
Described second-level storage, for receive described management node send calculation command { " vcIP ", " ID " } after, measure the network delay between self and described computing node vcIP, detect self according to described ID and whether comprise described desired data, and return result of calculation { " vcIP " to described management node, " secIP ", " ID ", " flag ", " time " }, wherein, described secIP is the mark of described second-level storage, described flag is the mark whether comprising desired data in second-level storage secIP, described time is the network delay between described second-level storage secIP and described computing node vcIP,
Described management node, for after receiving the result of calculation that second-level storage described in each returns, according to described network delay time, according to network delay order from short to long, each second-level storage described is sorted, according to the result that described second-level storage sorts, judge whether the flag that each second-level storage described is corresponding is YES successively, and the resource identifier URI first being judged the mark secIP that flag is the target second-level storage of YES and the described desired data comprised thereof sends to described computing node;
Described computing node, for according to the mark secIP of described target second-level storage and the resource identifiers, URIs of described desired data that comprises thereof, obtains described desired data from described target second-level storage.
Embodiments provide a kind of management node, comprising: receiving element, acquiring unit, judging unit and transmitting element;
Described receiving element, obtains desired data request for receiving external computing nodes;
Described acquiring unit, for obtaining the network delay between described computing node and each second-level storage outside;
Described judging unit, for judging whether comprise described computing node desired data in each second-level storage described;
Described transmitting element, for will comprise described computing node desired data and and between described computing node the mark of the target second-level storage that network delay is the shortest send to described computing node.
Preferably, this management node comprises further: sequencing unit;
Described sequencing unit, for according to the network delay between described computing node and each second-level storage described, according to network delay order from short to long, sorts to each second-level storage described;
Described judging unit, for the ranking results according to described sequencing unit, judges whether comprise described computing node desired data, until first is judged the target second-level storage comprising described computing node desired data in each second-level storage described one by one;
Described transmitting element, for the judged result according to described judging unit, sends to described computing node by the mark that described first is judged the target second-level storage comprising described computing node desired data.
Preferably, described transmitting element, be further used for the ranking results according to described sequencing unit, when described target second-level storage is not the second-level storage of sequence first, send long-range copy order, the described computing node desired data in described target second-level storage is copied in the second-level storage of described sequence first.
Preferably, described transmitting element, be further used for every setting-up time, whether identically detect the data stored in each second-level storage described, if not, send long-range copy order, with mutually copying by data between each second-level storage described, between realization each second-level storage described, data is synchronous.
Preferably, described transmitting element, for receive the request of described computing node acquisition desired data at described receiving element after, calculation command { " vcIP " is sent to each second-level storage described, " ID " }, wherein said vcIP is the mark of described computing node, and described ID is the mark of described desired data;
Described receiving element, for receiving the result of calculation { " vcIP " that second-level storage described in each returns, " secIP ", " ID ", " flag ", " time " }, wherein, described secIP is the mark of described second-level storage, and described flag is the mark whether comprising desired data in second-level storage secIP, and described time is the network delay between described second-level storage secIP and described computing node vcIP;
Described sequencing unit, for according to described network delay time, according to network delay order from short to long, sorts to each second-level storage described;
Described judging unit, for the ranking results according to described sequencing unit, judges whether the flag that each second-level storage described is corresponding is YES successively;
Described transmitting element, sends to described computing node for the resource identifier URI first being judged the mark secIP that flag is the target second-level storage of YES and the described desired data comprised thereof.
The embodiment of the present invention additionally provides a kind of data capture method, is applied in management node, comprises:
Receive the request that computing node obtains desired data;
Obtain second-level storage described in each and the network delay between described computing node;
Judge whether comprise described desired data in second-level storage described in each;
To described desired data be comprised and and between described computing node the mark of the target second-level storage that network delay is the shortest send to described computing node.
Preferably, after the network delay of described acquisition described in each between second-level storage and described computing node, comprise further: according to the network delay between described computing node and each second-level storage described, according to network delay order from short to long, each second-level storage described is sorted;
Describedly judge that whether comprising described desired data in memory at different levels described in each comprises: according to the ranking results of described second-level storage, judge whether comprise described desired data, until first is judged the current second-level storage comprising described desired data in each second-level storage described one by one.
Preferably, described will comprise described desired data and and after between described computing node, the mark of the target second-level storage that network delay is the shortest sends to described computing node, comprise further: according to the ranking results of described second-level storage, send copy order, the described desired data in described target second-level storage is copied in the second-level storage of described sequence first.
Preferably, further, every setting-up time, whether identically detect the data stored in each second-level storage described, if not, send copy order, by mutually copying of data between each second-level storage described, between realization each second-level storage described, data is synchronous.
Embodiments provide a kind of cloud system, the method of management node and data acquisition, at least two second-level storages are disposed in different address locations, when computing node will obtain desired data from second-level storage, computing node sends the request obtaining desired data to management node, management node obtains the network delay between computing node and each second-level storage, and judge the data that whether comprise in each second-level storage required for computing node, management node by the data not only comprised required for computing node but also and computing node between the mark of the shortest second-level storage of network delay send to this computing node, after this computing node receives the second-level storage mark of management node transmission, from the second-level storage of correspondence, required data are obtained according to this mark, like this, in different geographical position, multiple second-level storage is set, the spacing of computing node and second-level storage is nearer, network delay is corresponding shorter, computing node can from the second-level storage of close together obtain desired data, avoid the situation that in cloud system, the time delay of cross-region transmission data network is larger to occur, improve the speed of cloud system transfer of data.
Accompanying drawing explanation
Fig. 1 is a kind of cloud system schematic diagram that one embodiment of the invention provides;
Fig. 2 is a kind of management node schematic diagram that one embodiment of the invention provides;
Fig. 3 is the method flow diagram of a kind of data acquisition that one embodiment of the invention provides;
Fig. 4 is the method flow diagram of a kind of data acquisition that another embodiment of the present invention provides.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described.Obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
As shown in Figure 1, one embodiment of the invention provides a kind of cloud system, comprising: management node 101, at least one computing node 102 and at least two second-level storages 103;
Each second-level storage 103 in described at least two second-level storages 103 is deployed in different geographical position, for storing the Various types of data needed for described computing node 102;
Described management node 101, the request of desired data is obtained for receiving described computing node 102, obtain network delay between described computing node 102 and each second-level storage 103 described and whether each second-level storage 103 comprises described computing node 102 desired data described in judging, will described computing node 102 desired data be comprised and and between described computing node 102 mark of the target second-level storage 103 that network delay is the shortest send to described computing node 102;
Described computing node 102, for sending the request obtaining data to described management node 101, and after the mark receiving the target second-level storage 103 that described management node 101 sends, obtains desired data from described target second-level storage 103.
Embodiments provide a kind of cloud system, at least two second-level storages are disposed in different address locations, when computing node will obtain desired data from second-level storage, computing node sends the request obtaining desired data to management node, management node obtains the network delay between computing node and each second-level storage, and judge the data that whether comprise in each second-level storage required for computing node, management node by the data not only comprised required for computing node but also and computing node between the mark of the shortest second-level storage of network delay send to this computing node, after this computing node receives the second-level storage mark of management node transmission, from the second-level storage of correspondence, required data are obtained according to this mark, like this, in different geographical position, multiple second-level storage is set, the spacing of computing node and second-level storage is nearer, network delay is corresponding shorter, computing node can from the second-level storage of close together obtain desired data, avoid the situation that in cloud system, the time delay of cross-region transmission data network is larger to occur, improve the speed of cloud system transfer of data.
In an embodiment of the invention, management node is after getting the network delay between computing node and each second-level storage, according to network delay order from short to long, each second-level storage is sorted, according to ranking results, judge the data whether comprised in each second-level storage needed for computing node successively, if, do not judged next second-level storage, until find the target second-level storage comprising computing node desired data, stop judging, the mark of this target second-level storage is sent to computing node, computing node obtains desired data from this target second-level storage, like this, the path that guarantee computing node obtains desired data is that network delay is the shortest, thus improve the speed obtaining data.
In an embodiment of the invention, according to the ranking results of each second-level storage, when not comprising the data needed for computing node in the second-level storage of sequence first, management node sends copy command, computing node desired data will be comprised and computing node desired data in the shortest second-level storage of network delay copies in the second-level storage of sequence first, like this, when next time, this computing node needed to obtain the data identical with this again, this data are obtained by the second-level storage from sequence first, such computing node later stage is when again obtaining identical data, to ensure that it obtains this data from the shortest second-level storage of network delay, shorten the time needed for transfer of data.
In an embodiment of the invention, management node is every a setting-up time, will detect whether the data stored in each second-level storage are identical, if not identical, for each second-level storage, self not to be comprised but in the data Replica that comprises of other second-level storages to self storage, realize the synchronous of data between each second-level storage, thus after the data in a second-level storage are upgraded, ensure that the data in other each second-level storages are also up-to-date.
In an embodiment of the invention, management node is after the request receiving the acquisition desired data that computing node sends, calculation command { " vcIP " is sent to each second-level storage, " ID " }, wherein vcIP is the mark of computing node, ID is the mark of desired data, each second-level storage is after receiving calculation command, measure the network delay between self and computing node vcIP, and judge whether self comprises required data according to ID, result of calculation { " vcIP " is returned to management node, " secIP ", " ID ", " flag ", " time " }, in, described secIP is the mark of described second-level storage, described flag is the mark whether comprising desired data in second-level storage secIP, described time is the network delay between described second-level storage secIP and described computing node vcIP, management node is after receiving the result of calculation that each second-level storage returns, according to time from short to long, each second-level storage is sorted, and judge whether the flag that each second-level storage is corresponding is YSE successively according to ranking results, the resource identifier URI first being judged the mark secIP that flag is the second-level storage of YES and desired data sends to computing node, realize selection to second-level storage when computing node obtains desired data like this, ensure that the path obtaining desired data is that network delay is the shortest, improve the speed of transfer of data, reduce the time needed for transfer of data.
In embodiment the present invention one, the file stored in second-level storage has the advantages that single request data volume is large, read to write less more and do not modify after write, be generally ios file or system backup file that computing node creates virtual machine use, continue to avoid second-level storage to carry out the operation of cross-region data syn-chronization, affect the performance of cloud system.
As shown in Figure 2, one embodiment of the invention provides a kind of management node, comprising: receiving element 201, acquiring unit 202, judging unit 203 and transmitting element 204;
Receiving element 201, is connected with the computing node of outside, for receiving the request of the acquisition desired data that external computing nodes sends;
Acquiring unit 202, is connected with receiving element 201, for receive the acquisition desired data that computing node sends at receiving element 201 request after, obtain the network delay between computing node and each second-level storage outside;
Judging unit 203, is connected with receiving element 201, for judging the data whether comprised in each second-level storage needed for computing node;
Transmitting element 204, be connected with acquiring unit 202, judging unit 203 and external computing nodes, for the judged result of the network delay that obtains according to acquiring unit 202 and judging unit 203, will computing node desired data be comprised and and between computing node the mark of the target second-level storage that network delay is the shortest send to computing node.
In an embodiment of the invention, management node comprises further: sequencing unit;
Sequencing unit, is connected with acquiring unit 202 and judging unit 203, for the network delay between the computing node that obtains according to acquiring unit 203 and each second-level storage, according to network delay order from short to long, sorts to each second-level storage;
Correspondingly, judging unit 203 is according to the ranking results of sequencing unit, from first second-level storage, judge the data whether comprised in each second-level storage needed for computing node one by one, until find a target second-level storage comprising computing node desired data, then stop judging operation;
Correspondingly, the mark comprising the target second-level storage of computing node desired data found, according to the judged result of judging unit 203, is sent to computing node by transmitting element 204.
In an embodiment of the invention, according to the judged result of judging unit 203, if comprise computing node desired data and and the target second-level storage that between computing node, network delay is the shortest be not be positioned at during second-level storage sorts first second-level storage, the second-level storage that transmitting element 204 is positioned at first in second-level storage sequence sends long-range copy order, after this second-level storage receives copy command, in the data Replica needed for the computing node that described target second-level storage is comprised to self storage.
In an embodiment of the invention, pre-determine a setting-up time, every a setting-up time, whether transmitting element 204 detects the data stored in each second-level storage identical, if so, does not then deal with, if not, then transmitting element 204 sends copy order to each second-level storage, after each second-level storage receives copy command, self not to be comprised but in the data Replica that comprises of other one or more second-level storages to self storage.
In an embodiment of the invention, receive the request of the acquisition desired data that computing node sends at receiving element 201 after, transmitting element 204 sends calculation command { " vcIP " to each second-level storage, " ID " }, wherein vcIP is the mark of current computing node, and ID is the mark of current computing node desired data;
Receiving element 201, for receiving the result of calculation { " vcIP " that each second-level storage returns for calculation command, " secIP ", " ID ", " flag ", " time " }, wherein secIP is the mark of the second-level storage returning this result, flag is the mark whether comprising computing node desired data in the second-level storage returning this result, and time is the network delay returned between the second-level storage of this result and computing node;
Sequencing unit, for the time parameter in the result of calculation that returns according to each second-level storage, according to the order that time is ascending, each second-level storage is sorted, ranking results be the short second-level storage of network delay front, the long second-level storage of network delay is rear;
Judging unit 203, for according to the ranking results of sequencing unit to each second-level storage, from first second-level storage, judge whether the flag parameter in the result of calculation that each second-level storage returns is YES, if not, according to second-level storage clooating sequence successively, judge the flag parameter that next second-level storage returns, if so, stop judging, current second-level storage is target second-level storage;
Transmitting element 204, be further used for the judged result according to judging unit 203, the mark of target second-level storage is sent to computing node, meanwhile, the resource identifiers, URIs of the computing node desired data comprised by target second-level storage also sends to computing node, computing node, by the mark of target second-level storage and the resource identifiers, URIs of desired data, directly can obtain desired data in target second-level storage.
As shown in Figure 3, one embodiment of the invention provides a kind of data capture method, is applied in the management node that above-described embodiment provides, comprises:
Step 301: receive the request that computing node obtains desired data;
Step 302: obtain second-level storage described in each and the network delay between described computing node;
Step 303: judge whether comprise described desired data in second-level storage described in each;
Step 304: will described desired data be comprised and and between described computing node the mark of the target second-level storage that network delay is the shortest send to described computing node.
In embodiments of the present invention, after obtaining the request of the acquisition desired data that computing node sends, first the network delay between this computing node and each second-level storage is obtained, and judge in each second-level storage, whether to comprise this computing node desired data, the second-level storage that between selection and calculation node, network delay is the shortest in each second-level storage comprising computing node desired data is as target second-level storage, the mark of target second-level storage is sent to this computing unit, this computing unit obtains desired data from target second-level storage, like this, in each second-level storage, the second-level storage that between selection and calculation node, network delay is the shortest provides described data for computing node, farthest can reduce the network delay in data transmission procedure, improve the speed of cloud system transfer of data.
For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with drawings and the specific embodiments, the present invention is described in further detail.
As shown in Figure 4, one embodiment of the invention provides a kind of method of data acquisition, is applied in the management node that above-described embodiment provides, comprises:
Step 401: the request receiving the acquisition desired data that external computing nodes sends.
In an embodiment of the invention, each management node at least one computing node corresponding and at least two second-level storages, wherein each second-level storage is arranged in diverse geographic location, each second-level storage stores the data that computing node uses, these data mainly create ISO file or the system backup file of virtual machine use, when data needed for a computing node will obtain from second-level storage, first send to the receiving element in management node and obtain the request of desired data.Such as, corresponding three computing nodes of management node and three second-level storages, three computing nodes are respectively computing node 1, computing node 2 and computing node 3, three second-level storages are respectively second-level storage 1, second-level storage 2 and second-level storage 3, three second-level storages are arranged in place 1, place 2 and place 3 respectively, when computing node 1 will obtain the system backup file 1 creating virtual machine, send the request obtaining system backup file 1 to the receiving element in management node.
Step 402: obtain the situation whether comprising desired data in network delay between computing node and each second-level storage and each second-level storage.
In an embodiment of the invention, after receiving the request of the acquisition desired data that computing node sends, calculation command is sent to each second-level storage by transmitting element, calculation command is { " vcIP ", " ID " }, wherein vcIP is the mark of computing node, ID is the mark of desired data, each second-level storage is after receiving calculation command, the data calculating the network delay between self and computing node and self whether comprise needed for this computing node, result of calculation is returned and sends to receiving element, this result of calculation is { " vcIP ", " secIP ", " ID ", " flag ", " time " }, wherein secIP is the mark of second-level storage, flag is the mark whether comprising desired data, time is the network delay between computing node, management node obtains the network delay between each second-level storage and computing node and whether comprises the situation of desired data from the result of calculation received.Such as, receiving element is after receiving computing node 1 and obtaining the request of system backup file 1, calculation command { " vc1 " is sent to second-level storage 1 to 3 by transmitting element, " ghost1 " }, after second-level storage 1 to 3 receives calculation command, measure the network delay of self and computing node 1 respectively, and judge whether self comprises system backup file 1, by result of calculation occurrence value receiving element, the result of calculation that wherein second-level storage 1 sends is { " vc1 ", " sec1 ", " ghost1 ", " NO ", " 0.3 " }, the result of calculation that second-level storage 2 sends is { " vc1 ", " sec2 ", " ghost1 ", " YES ", " 0.4 " }, the result of calculation that second-level storage 3 sends is { " vc1 ", " sec3 ", " ghost1 ", " YES ", " 0.5 " }, the network delay obtaining second-level storage 1 from the result that three second-level storages return is 0.3s, do not comprise system backup file 1, the network delay of second-level storage 2 is 0.4s, comprise system backup file 1, the network delay of second-level storage 3 is 0.5s, comprise system backup file 1.
Step 403: the network delay corresponding according to each second-level storage, sorts to each second-level storage.
In an embodiment of the invention, when after the network delay that receiving element gets between each second-level storage and computing node, according to network delay order from short to long, individual second-level storage is sorted.Such as, according to the network delay between three second-level storages and computing node 1, according to network delay order from short to long, three second-level storages are sorted, second-level storage 1, second-level storage 2 and the network delay between second-level storage 3 and computing node 1 are respectively 0.3s, 0.4s and 0.5s, so ranking results is second-level storage 1-second-level storage 2-second-level storage 3.
Step 404: according to the ranking results of second-level storage, judges the desired data whether comprising computing node in the second-level storage of sequence first, if so, performs step 405, otherwise performs step 406.
In an embodiment of the invention, according to the sequence of each second-level storage, first judge in the result of calculation that the second-level storage of sequence first returns, whether flag is YES, if, illustrate that this second-level storage comprises the data needed for computing node, desired data is provided to computing node by this second-level storage, corresponding execution step 405, if not, the data do not comprised needed for computing node are then described in this second-level storage, there is provided desired data by other second-level storages below, perform step 406 accordingly.Such as, according to the ranking results of three second-level storages, first judge in the result of calculation that second-level storage 1 returns, whether flag is YES, because flag in the result of calculation that second-level storage 1 returns is NO, perform step 406 so corresponding.
Step 405: the resource identifiers, URIs of desired data described in the mark of the second-level storage of sequence first and this memory is sent to computing node, and receives current process.
In an embodiment of the invention, when the second-level storage of sequence first comprises the data needed for computing node, the resource identifiers, URIs of the computing node desired data having transmitting element the mark of the second-level storage of sequence first and this second-level storage to be comprised sends to computing node, and computing node obtains desired data according to the mark of second-level storage and the resource identifiers, URIs of desired data from the second-level storage of sequence first.Such as, if second-level storage 1 comprises the system backup file 1 needed for computing node 1, so the resource identifiers, URIs of the system backup file 1 in the mark sec1 of second-level storage 1 and second-level storage 1 is sent to computing node 1 by transmitting element, and corresponding computing node 1 directly obtains system backup file 1 from second-level storage 1.
Whether step 406: according to the ranking results of second-level storage, comprise described desired data in each second-level storage after judging the second-level storage of sequence first successively, until find the target second-level storage comprising described desired data.
In an embodiment of the invention, when not comprising described desired data in the second-level storage of sequence first, judging unit is according to second-level storage ranking results, from second second-level storage, judge whether the flag that each second-level storage is corresponding is YES successively, until find a corresponding flag to be the target second-level storage of YES, stop judging.Such as, judging unit is according to the ranking results of three second-level storages, after judging not comprise system backup file 1 in second-level storage 1, by detecting whether corresponding flag is YES, continue to judge whether comprise system backup file 1 in second-level storage 2 below and second-level storage 3 successively, when determining second-level storage 2 and comprising system backup file 1, stop judging.
Step 407: the resource identifiers, URIs of desired data described in target second-level storage mark and this second-level storage is sent to computing node.
In an embodiment of the invention, when judging the data that target second-level storage comprises needed for computing node, the resource identifiers, URIs of computing node desired data in the mark of target second-level storage and target memory is sent to computing node by transmitting element, and corresponding computing node directly obtains desired data from target second-level storage.Such as, when judging that second-level storage 2 comprises system backup file 1, the resource identifiers, URIs of system backup file 1 in the mark sec2 of second-level storage 2 and second-level storage 2 is sent to computing node 1 by transmitting element, and computing node 1 directly obtains system backup file 1 from second-level storage 2.
Step 408: the second-level storage to described sequence first sends order, is copied in the second-level storage of sequence first by desired data described in target second-level storage.
In an embodiment of the invention, by the mark of target second-level storage and comprising the resource identifiers, URIs of desired data send to computing node after, transmitting element sends copy order to the second-level storage of sequence first, the second-level storage of sequence first copies described desired data from target second-level storage, is stored in the storage of self.Such as, after transmitting element sends the resource representation symbol URI of system backup file 1 in the mark of second-level storage 2 and second-level storage 2 to computing node 1, copy order is sent to second-level storage 1, after second-level storage 1 receives this copy order, the system backup file 1 in second-level storage 2 is copied in the storage of self.
It should be noted that, in the embodiment of the present invention, whether identical transmitting element, according to predetermined setting-up time, every a setting-up time, detect the data stored in each second-level storage, if, do not deal with, if not, then send long-range copy order to each second-level storage, other second-level storages, after receiving copy order, to comprise but in self data Replica not to be covered to self storage by each second-level storage.Such as, setting-up time is 2 hours, not through 2 hours, whether the data that transmitting element detects in three second-level storages are identical, if identical, then do not deal with, if different, wherein second-level storage 1 comprises system backup file 2, but do not comprise system backup file 2 in second-level storage 2 and second-level storage 3, transmitting element then sends copy order to three second-level storages, and after receiving copy order, second-level storage 2 and second-level storage 3 are respectively by the storage of the system backup file 2 in second-level storage 1 self.
According to such scheme, a kind of method of cloud system, management node and data acquisition that embodiments of the invention provide, at least has following beneficial effect:
1, in the embodiment of the present invention, at least two second-level storages are disposed in different address locations, when computing node will obtain desired data from second-level storage, computing node sends the request obtaining desired data to management node, management node obtains the network delay between computing node and each second-level storage, and judge the data that whether comprise in each second-level storage required for computing node, management node by the data not only comprised required for computing node but also and computing node between the mark of the shortest second-level storage of network delay send to this computing node, after this computing node receives the second-level storage mark of management node transmission, from the second-level storage of correspondence, required data are obtained according to this mark, like this, in different geographical position, multiple second-level storage is set, the spacing of computing node and second-level storage is nearer, network delay is corresponding shorter, computing node can from the second-level storage of close together obtain desired data, avoid the situation that in cloud system, the time delay of cross-region transmission data network is larger to occur, improve the speed of cloud system transfer of data.
2, in the embodiment of the present invention, if during with the data do not comprised in sequence first second-level storage that computing node network delay is the shortest needed for computing node, after computing node obtains desired data from other target second-level storages, this desired data copies in the storage of self by first second-level storage that sorts from target second-level storage, like this, when identical computing node obtains identical desired data again, just can obtain desired data from the shortest sequence first second-level storage of network delay, the speed of further raising transfer of data, reduce the time obtaining data.
3, in the embodiment of the present invention, often through the time of a setting, the data stored in each second-level storage will be detected whether identical, if not identical, mutually copying then by data between each second-level storage, ensures that the data stored in each second-level storage are consistent, realizes the synchronous of data in each second-level storage, ensure that computing node is obtain from the shortest second-level storage of network delay with it when obtaining desired data, reduce the time of transfer of data.
4, in the embodiment of the present invention, the data stored in second-level storage have read to write less more, that single is not generally modified after reading the large and write of data volume is specific, the speed of such guarantee big data quantity transmission, and avoid after data write in second-level storage or amendment, when each second-level storage realizes data syn-chronization each other data continue copy, affect the performance of cloud system.
It should be noted that, in this article, the relational terms of such as first and second and so on is only used for an entity or operation to separate with another entity or operating space, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element " being comprised a 〃 〃 〃 〃 〃 〃 " limited by statement, and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical factor.
One of ordinary skill in the art will appreciate that: all or part of step realizing said method embodiment can have been come by the hardware that program command is relevant, aforesaid program can be stored in the storage medium of embodied on computer readable, this program, when performing, performs the step comprising said method embodiment; And aforesaid storage medium comprises: ROM, RAM, magnetic disc or CD etc. various can be program code stored medium in.
Finally it should be noted that: the foregoing is only preferred embodiment of the present invention, only for illustration of technical scheme of the present invention, be not intended to limit protection scope of the present invention.All any amendments done within the spirit and principles in the present invention, equivalent replacement, improvement etc., be all included in protection scope of the present invention.
Claims (10)
1. a cloud system, is characterized in that, comprising: management node, at least one computing node and at least two second-level storages;
Each second-level storage in described at least two second-level storages is deployed in different geographical position, for storing the Various types of data needed for described computing node;
Described management node, the request of desired data is obtained for receiving described computing node, obtain network delay between described computing node and each second-level storage described and whether each second-level storage comprises described computing node desired data described in judging, will described computing node desired data be comprised and and between described computing node the mark of the target second-level storage that network delay is the shortest send to described computing node;
Described computing node, for sending the request obtaining desired data to described management node, and after the mark receiving the described target second-level storage that described management node sends, obtains desired data from described target second-level storage.
2. cloud system according to claim 1, is characterized in that,
Described management node, for according to the network delay between described computing node and each second-level storage described, according to network delay order from short to long, each second-level storage described is sorted, according to the ranking results of described second-level storage, judge whether comprise described computing node desired data in each second-level storage described one by one, and the mark first being judged the target second-level storage comprising described computing node desired data sends to described computing node.
3. cloud system according to claim 2, is characterized in that,
Described management node, be further used for the ranking results according to described second-level storage, when described target second-level storage is not the second-level storage of sequence first, send long-range copy order, to copy in the second-level storage of described sequence first by the described computing node desired data in described target second-level storage;
And/or,
Described management node, be further used for every setting-up time, whether identically detect the data stored in each second-level storage described, if not, send long-range copy order, with mutually copying by data between each second-level storage described, between realization each second-level storage described, data is synchronous.
4., according to described cloud system arbitrary in claims 1 to 3, it is characterized in that,
Described management node, for after the request receiving described computing node acquisition desired data, send calculation command { " vcIP ", " ID " } to each second-level storage described, wherein said vcIP is the mark of described computing node, and described ID is the mark of described desired data;
Described second-level storage, for receive described management node send calculation command { " vcIP ", " ID " } after, measure the network delay between self and described computing node vcIP, detect self according to described ID and whether comprise described desired data, and return result of calculation { " vcIP " to described management node, " secIP ", " ID ", " flag ", " time " }, wherein, described secIP is the mark of described second-level storage, described flag is the mark whether comprising desired data in second-level storage secIP, described time is the network delay between described second-level storage secIP and described computing node vcIP,
Described management node, for after receiving the result of calculation that second-level storage described in each returns, according to described network delay time, according to network delay order from short to long, each second-level storage described is sorted, according to the result that described second-level storage sorts, judge whether the flag that each second-level storage described is corresponding is YES successively, and the resource identifier URI first being judged the mark secIP that flag is the target second-level storage of YES and the described desired data comprised thereof sends to described computing node;
Described computing node, for according to the mark secIP of described target second-level storage and the resource identifiers, URIs of described desired data that comprises thereof, obtains described desired data from described target second-level storage.
5. a management node, is characterized in that, comprising: receiving element, acquiring unit, judging unit and transmitting element;
Described receiving element, obtains desired data request for receiving external computing nodes;
Described acquiring unit, for obtaining the network delay between described computing node and each second-level storage outside;
Described judging unit, for judging whether comprise described computing node desired data in each second-level storage described;
Described transmitting element, for will comprise described computing node desired data and and between described computing node the mark of the target second-level storage that network delay is the shortest send to described computing node.
6. management node according to claim 5, is characterized in that, comprises further: sequencing unit;
Described sequencing unit, for according to the network delay between described computing node and each second-level storage described, according to network delay order from short to long, sorts to each second-level storage described;
Described judging unit, for the ranking results according to described sequencing unit, judges whether comprise described computing node desired data, until first is judged the target second-level storage comprising described computing node desired data in each second-level storage described one by one;
Described transmitting element, for the judged result according to described judging unit, sends to described computing node by the mark that described first is judged the target second-level storage comprising described computing node desired data.
7. management node according to claim 6, is characterized in that,
Described transmitting element, be further used for the ranking results according to described sequencing unit, when described target second-level storage is not the second-level storage of sequence first, send long-range copy order, the described computing node desired data in described target second-level storage is copied in the second-level storage of described sequence first;
And/or,
Described transmitting element, be further used for every setting-up time, whether identically detect the data stored in each second-level storage described, if not, send long-range copy order, with mutually copying by data between each second-level storage described, between realization each second-level storage described, data is synchronous.
8., according to the arbitrary described management node of claim 5 to 8, it is characterized in that,
Described transmitting element, for receive the request of described computing node acquisition desired data at described receiving element after, send calculation command { " vcIP ", " ID " } to each second-level storage described, wherein said vcIP is the mark of described computing node, and described ID is the mark of described desired data;
Described receiving element, for receiving the result of calculation { " vcIP " that second-level storage described in each returns, " secIP ", " ID ", " flag ", " time " }, wherein, described secIP is the mark of described second-level storage, and described flag is the mark whether comprising desired data in second-level storage secIP, and described time is the network delay between described second-level storage secIP and described computing node vcIP;
Described sequencing unit, for according to described network delay time, according to network delay order from short to long, sorts to each second-level storage described;
Described judging unit, for the ranking results according to described sequencing unit, judges whether the flag that each second-level storage described is corresponding is YES successively;
Described transmitting element, sends to described computing node for the resource identifier URI first being judged the mark secIP that flag is the target second-level storage of YES and the described desired data comprised thereof.
9. a method for data acquisition, is characterized in that, is applied in management node, comprises:
Receive the request that computing node obtains desired data;
Obtain second-level storage described in each and the network delay between described computing node;
Judge whether comprise described desired data in second-level storage described in each;
To described desired data be comprised and and between described computing node the mark of the target second-level storage that network delay is the shortest send to described computing node.
10. method according to claim 9, is characterized in that,
After the network delay of described acquisition described in each between second-level storage and described computing node, comprise further: according to the network delay between described computing node and each second-level storage described, according to network delay order from short to long, each second-level storage described is sorted;
Describedly judge that whether comprising described desired data in memory at different levels described in each comprises: according to the ranking results of described second-level storage, judge whether comprise described desired data, until first is judged the current second-level storage comprising described desired data in each second-level storage described one by one;
And/or,
Described will comprise described desired data and and after between described computing node, the mark of the target second-level storage that network delay is the shortest sends to described computing node, comprise further: according to the ranking results of described second-level storage, send copy order, the described desired data in described target second-level storage is copied in the second-level storage of described sequence first;
And/or,
Further, every setting-up time, whether identically detect the data stored in each second-level storage described, if not, send copy order, by mutually copying of data between each second-level storage described, between realization each second-level storage described, data is synchronous.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510539917.3A CN105245565A (en) | 2015-08-28 | 2015-08-28 | Cloud system, management node and data obtaining method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510539917.3A CN105245565A (en) | 2015-08-28 | 2015-08-28 | Cloud system, management node and data obtaining method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105245565A true CN105245565A (en) | 2016-01-13 |
Family
ID=55043063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510539917.3A Pending CN105245565A (en) | 2015-08-28 | 2015-08-28 | Cloud system, management node and data obtaining method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105245565A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110071949A (en) * | 2018-01-23 | 2019-07-30 | 阿里巴巴集团控股有限公司 | A kind of system, the method and apparatus of across geographic area management calculating application |
| CN115562911A (en) * | 2022-12-07 | 2023-01-03 | 中科方德软件有限公司 | Virtual machine data backup method, device, system, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932442A (en) * | 2012-10-26 | 2013-02-13 | 北京奇虎科技有限公司 | Equipment, system and method for distributed data storage system |
| CN103873546A (en) * | 2012-12-17 | 2014-06-18 | 国际商业机器公司 | Method and Apparatus for Controlling Data Storage Among Storage Centers |
-
2015
- 2015-08-28 CN CN201510539917.3A patent/CN105245565A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932442A (en) * | 2012-10-26 | 2013-02-13 | 北京奇虎科技有限公司 | Equipment, system and method for distributed data storage system |
| CN103873546A (en) * | 2012-12-17 | 2014-06-18 | 国际商业机器公司 | Method and Apparatus for Controlling Data Storage Among Storage Centers |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110071949A (en) * | 2018-01-23 | 2019-07-30 | 阿里巴巴集团控股有限公司 | A kind of system, the method and apparatus of across geographic area management calculating application |
| CN115562911A (en) * | 2022-12-07 | 2023-01-03 | 中科方德软件有限公司 | Virtual machine data backup method, device, system, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11275641B2 (en) | Automatic correlation of dynamic system events within computing devices | |
| CN102932415B (en) | Method and device for storing mirror image document | |
| CN103019960B (en) | Distributed caching method and system | |
| CN102857578B (en) | A kind of file uploading method of network hard disc, system and net dish client | |
| CN105701096A (en) | Index generation method, data inquiry method, index generation device, data inquiry device and system | |
| CN104298541A (en) | Data distribution algorithm and data distribution device for cloud storage system | |
| CN107085546B (en) | Data management method and device based on fault domain technology | |
| CN110928497B (en) | Metadata processing method, device and equipment and readable storage medium | |
| CN109714239B (en) | Management message issuing method, VNFM (virtual network management frequency) equipment and server | |
| CN104348679A (en) | Bucket testing method, device and system | |
| CN105159800A (en) | Data recovery method, data backup method, data recovery equipment, data backup equipment, data recovery system and data backup system | |
| CN105468660A (en) | Reading method for distributed file system, client device and distributed file system | |
| CN104216759A (en) | Method and device for generating logical disk of virtual machine | |
| CN105117345A (en) | Interface testing method and device for application program | |
| CN110008197A (en) | A kind of data processing method, system and electronic equipment and storage medium | |
| CN105245565A (en) | Cloud system, management node and data obtaining method | |
| CN104424316A (en) | Data storage method, data searching method, related device and system | |
| CN104283966A (en) | Data distribution algorithm and device of cloud storage system | |
| CN104580428A (en) | Data routing method, data management device and distributed storage system | |
| CN106354587A (en) | Mirror image server and method for exporting mirror image files of virtual machine | |
| CN111046004B (en) | Data file storage method, device, equipment and storage medium | |
| CN107547605B (en) | message reading and writing method based on node queue and node equipment | |
| CN105205062A (en) | Data storage method and data reading method and device | |
| CN103561128A (en) | FC ID recycling method and network equipment | |
| CN102447589B (en) | Method and device for aggregating records |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160113 |
|
| WD01 | Invention patent application deemed withdrawn after publication |