CN107729514A - A kind of Replica placement node based on hadoop determines method and device - Google Patents
A kind of Replica placement node based on hadoop determines method and device Download PDFInfo
- Publication number
- CN107729514A CN107729514A CN201711007971.9A CN201711007971A CN107729514A CN 107729514 A CN107729514 A CN 107729514A CN 201711007971 A CN201711007971 A CN 201711007971A CN 107729514 A CN107729514 A CN 107729514A
- Authority
- CN
- China
- Prior art keywords
- node
- connection number
- load
- copy
- target
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/11—File system administration, e.g. details of archiving or snapshots
- G06F16/113—Details of archiving
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/13—File access structures, e.g. distributed indices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention discloses a kind of Replica placement node based on hadoop to determine method, apparatus, equipment and computer-readable recording medium, including:Target rack server is determined according to the copy type of target copy;Node to be placed is chosen from target rack server, forms node cluster to be placed;The node that work connection number is less than connection number threshold value is chosen from node cluster to be placed, and is less than in the node of connection number threshold value the really minimum node of real time load, the placement node as target copy from work connection number.It can be seen that, in this programme, when selection is prevented putting the node of copy, need to consider simultaneously the real time load and HDFS progress of work numbers of node, so as to effectively raise the reasonable layout of copy, compared to the Replica Placement Strategy of acquiescence, the Replica Placement Strategy of optimization has more specific object, minimum real time load node is have selected as far as possible, avoids the storage of high capacity node, improves the time of transfer copies.
Description
Technical field
The present invention relates to distributed file system copy technical field of memory, it is based on more specifically to one kind
Hadoop Replica placement node determines method, apparatus, equipment and computer-readable recording medium.
Background technology
At present, Hadoop is enterprise's big data analysis platform of current main-stream.Hadoop is using the distributed texts of HDFS
Part system carries out data storage.HDFS using master-slave architecture design pattern (master/slavearchitecture), one
Individual name node (NameNode) and some back end (DataNode) form HDFS clusters.Wherein HDFS is superfluous using three copies
Remaining mechanism ensures the security of data.HDFS acquiescence Replica Placement Strategy principle be:As much as possible by two of which data
Block copy is stored in a frame, another data block copy is stored in another frame, well in bandwidth resources
And balanced in terms of reliability.
But acquiescence Replica Placement Strategy has certain limitation, major embodiment is as follows:When choosing copy memory node
Random machine mode is employed, although HDFS have also contemplated that the load information that the work of back end counts in succession, but relatively easy, and
And just judged after memory node is randomly selected.Such Replica placement mode is random by the distribution for causing copy
Greatly, especially in isomerous environment it is very possible there is distributing more data trnascription node be poor-performing node, these feelings
Condition, which will further result in some nodes, has very high load, and some nodes but cause data transmission efficiency in idle condition
Decline.
Therefore, the placement node of copy how is determined, to improve the harmony of clustered node load, is finally reached lifting number
It is that those skilled in the art need to solve according to the purpose of efficiency of transmission.
The content of the invention
It is an object of the invention to provide a kind of Replica placement node based on hadoop determine method, apparatus, equipment and
Computer-readable recording medium, to determine the placement node of copy, the harmony of clustered node load is improved, is finally reached lifting
The purpose of data transmission efficiency.
To achieve the above object, the embodiments of the invention provide following technical scheme:
A kind of Replica placement node based on hadoop determines method, including:
Target rack server is determined according to the copy type of target copy;
Node to be placed is chosen from the target rack server, forms node cluster to be placed;
The node that work connection number is less than connection number threshold value is chosen from the node cluster to be placed, and from the work
Connect number and be less than in the node of connection number threshold value the really minimum node of real time load, the placement node as the target copy.
Wherein, the copy type according to target copy determines target rack server, including:
If the copy type of the target copy is first copy, rack server is randomly selected as the target
Rack server;
If the copy type of the target copy is second copy, from first pair corresponding with the target copy
In other rack servers outside the rack server of this placement, target rack server is chosen;
If the copy type of the target copy is the 3rd copy, first pair corresponding to the target copy is judged
Whether the rack server of this placement is identical in the rack server of second Replica placement;If identical, from it is described
In other rack servers outside the rack server of second Replica placement corresponding to target copy, target frame clothes are chosen
Business device;If differing, using the rack server of second Replica placement corresponding to the target copy as the target machine
Frame server.
Wherein, the node that work connection number is chosen from the node cluster to be placed and is less than connection number threshold value, bag
Include:
Determine the work connection number of each node in the node cluster to be placed;
The average operation that the node cluster to be placed is calculated according to the work connection number of each node connects number, by described in
Average operation connection number chooses work connection number from the node cluster to be placed and is less than company as the connection number threshold value
Connect the node of several threshold values.
It is wherein, described to be less than in the node of connection number threshold value the really minimum node of real time load from work connection number,
As the placement node of the target copy, including:
It is determined that each work connection number be less than the disk I/O load of the node of connection number threshold value, internal memory load, cpu load with
And network load;
According to each work connect number be less than connection number threshold value node disk I/O load, internal memory load, cpu load with
And network load and load factor, it is determined that each work connection number is less than the real time load of the node of connection number threshold value, and choose
Placement node of the minimum node of real time load as the target copy.
Wherein, the basis each work connection number be less than connection number threshold value node disk I/O load, internal memory load,
Cpu load and network load and load factor, it is determined that each work connection number is less than the negative in real time of the node of connection number threshold value
Carry, including:
Determine that rule determines that each work connection number is less than the real time load of the node of connection number threshold value using real time load;
The real time load determines that rule is:
W=λio×wio+λmem×wmem+λcpu×wcpu+λband×wband;
Wherein, W is real time load, wioFor disk I/O load, wmemFor internal memory load, wcpuFor cpu load, wbandFor network
Load, λioFor disk specific gravity factor, λmemFor internal memory specific gravity factor, λcpuFor CPU specific gravity factors, λbandFor network bandwidth proportion system
Number, λio+λmem+λcpu+λband=1, and λio、λmem、λcpu、λband∈[0,1]。
A kind of Replica placement node determining device based on hadoop, including:
Target rack server determining module, target rack server is determined for the copy type according to target copy;
Cluster determining module, for choosing node to be placed from the target rack server, form node to be placed
Cluster;
Node selection module, it is less than connection number threshold value for choosing work connection number from the node cluster to be placed
Node, and it is less than in the node of connection number threshold value the really minimum node of real time load from work connection number, as the mesh
Mark the placement node of copy.
Wherein, the node selection module includes:
Work connection number determining unit, for determining that the work of each node in the node cluster to be placed connects number;
Average operation connects number computing unit, and the node to be placed is calculated for connecting number according to the work of each node
The average operation connection number of cluster;
Node selection unit, for the average operation to be connected into number as the connection number threshold value, and wait to put from described
Put and the node that work connection number is less than connection number threshold value is chosen in node cluster.
Wherein, the node selection module includes:
Load determining unit, for determine each work connection number be less than the node of connection number threshold value disk I/O load,
Internal memory load, cpu load and network load;
Real time load determining unit, the disk I/O for connecting the node that number is less than connection number threshold value according to each work are born
Load, internal memory load, cpu load and network load and load factor, it is determined that each work connection number is less than connection number threshold value
The real time load of node, and choose placement node of the minimum node of real time load as the target copy.
A kind of Replica placement node based on hadoop determines equipment, including:
Memory, for storing computer program;Processor, above-mentioned copy is realized during for performing the computer program
Place the step of node determines method.
A kind of computer-readable recording medium, computer program is stored with the computer-readable recording medium, it is described
The step of above-mentioned Replica placement node determines method is realized when computer program is executed by processor.
By above scheme, a kind of Replica placement node determination side based on hadoop provided in an embodiment of the present invention
Method, including:Target rack server is determined according to the copy type of target copy;Choose and treat from the target rack server
Node is placed, forms node cluster to be placed;Work connection number is chosen from the node cluster to be placed and is less than connection number threshold
The node of value, and it is less than in the node of connection number threshold value the really minimum node of real time load from work connection number, as institute
State the placement node of target copy.
It can be seen that in this programme, choose it is anti-put the node of copy when, it is necessary to consider simultaneously node real time load and
HDFS progress of work numbers, so as to effectively raise the reasonable layout of copy, compared to the Replica Placement Strategy of acquiescence, optimization
Replica Placement Strategy has more specific object, have selected minimum real time load node as far as possible, avoids high capacity node
Storage, improve the time of transfer copies;The invention also discloses a kind of Replica placement node determining device based on hadoop,
Equipment and computer-readable recording medium, it can equally realize above-mentioned technique effect.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is that a kind of Replica placement node based on hadoop disclosed in the embodiment of the present invention determines that method flow is illustrated
Figure;
Fig. 2 is a kind of Replica placement node determining device structural representation based on hadoop disclosed in the embodiment of the present invention
Figure.
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 carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
The embodiment of the invention discloses a kind of Replica placement node based on hadoop to determine method, apparatus, equipment and meter
Calculation machine readable storage medium storing program for executing, to determine the placement node of copy, the harmony of clustered node load is improved, is finally reached lifting number
According to the purpose of efficiency of transmission.
Referring to Fig. 1, a kind of Replica placement node based on hadoop provided in an embodiment of the present invention determines method, including:
S101, target rack server determined according to the copy type of target copy;
Wherein, the copy type according to target copy determines target rack server, including:
If the copy type of the target copy is first copy, rack server is randomly selected as the target
Rack server;
If the copy type of the target copy is second copy, from first pair corresponding with the target copy
In other rack servers outside the rack server of this placement, target rack server is chosen;
If the copy type of the target copy is the 3rd copy, first pair corresponding to the target copy is judged
Whether the rack server of this placement is identical in the rack server of second Replica placement;If identical, from it is described
In other rack servers outside the rack server of second Replica placement corresponding to target copy, target frame clothes are chosen
Business device;If differing, using the rack server of second Replica placement corresponding to the target copy as the target machine
Frame server.
Specifically, before the node for determining copy to be placed, it is necessary first to determine rack server, in this programme, really
Determining rack server is determined according to the type information of copy, and the type information refers to which copy copy is, typically
For, the selection of the placement location of copy follows is placed on copy to ensure the principle of reliability in different frames as far as possible, at this
In embodiment, by taking most common 3 copy scheme as an example, that is to say, that the type of target copy can be:First copy, second
Individual copy or the 3rd copy;Specifically, the Replica placement selection strategy of this programme is as follows:
1) number of copies chosen is needed>0 and need select first copy:
Judge whether client node is back end;The node is selected to deposit if client node is back end
Copy, otherwise, rack server is determined at random, and the node of drop target copy is chosen by the method described in this programme;
2) if selection triplicate:
Then specify in institute's organic frame where removing the first authentic copy outside frame and choose target rack server, and pass through we
The method choice node of case;
3) if selection triplicate:
If node where the first and second copies is in same frame, the institute where specifying removing triplicate outside frame
Organic frame goes to choose node by the method described in this programme, otherwise specifies frame where triplicate to pass through described in this programme
Method goes to choose node.
It should be noted that when target rack server is chosen from multiple rack servers, selection that can be random,
It can be chosen according to predetermined selection rule, it is not specific herein to limit.
S102, node to be placed is chosen from the target rack server, form node cluster to be placed;
Specifically, this programme chooses a number of back end from specified shelf position, node to be placed is generated
Cluster, the cluster is used for therefrom determining the node of drop target copy, it is necessary to explanation, and the number of nodes in the cluster can be with
It is set in advance, and when choosing node, can also be chosen according to node selection rule set in advance, equally can also
Randomly select, it is not specific herein to limit.
S103, the node that work connection number is less than connection number threshold value is chosen from the node cluster to be placed, and from institute
State work connection number and be less than in the node of connection number threshold value the really minimum node of real time load, the placement as the target copy
Node.
Wherein, the node that work connection number is chosen from the node cluster to be placed and is less than connection number threshold value, bag
Include:
Determine the work connection number of each node in the node cluster to be placed;
The average operation that the node cluster to be placed is calculated according to the work connection number of each node connects number, by described in
Average operation connection number chooses work connection number from the node cluster to be placed and is less than company as the connection number threshold value
Connect the node of several threshold values.
In this programme, determine after cluster, it is necessary to further choose work connection number from from the cluster less than connection number
The node of threshold value, and choose the minimum node of real time load as locations of copies from the node less than connection number threshold value and place section
Point;It should be noted that the connection number threshold value in this programme can be that user is set in advance, equally can also be according to actual feelings
Condition dynamically changes, such as cluster average operation is connected into number as connection number threshold value, not specific herein to limit.
It is understood that the HDFS back end progresses of work are the connection of the work such as back end HDFS write-ins, reading
Number.Because these loads are the relations of ratio, some nodes may be due to better performances under isomerous environment, and its is some negative in real time
Load is in reduced levels, and a large amount of copies of cluster will be caused to be stored in indivedual high-performance nodes when joint behavior is seriously unbalanced
On.The load information can control the HDFS progresses of work carried out on a back end, suppress some back end and carry out excessively
HDFS service.
It can be seen that in this programme, choose it is anti-put the node of copy when, it is necessary to consider simultaneously node real time load and
HDFS progress of work numbers, so as to effectively raise the reasonable layout of copy, compared to the Replica Placement Strategy of acquiescence, optimization
Replica Placement Strategy has more specific object, have selected minimum real time load node as far as possible, avoids high capacity node
Storage, improve the time of transfer copies.
It is described to connect number less than true real time load in the node for connecting number threshold value most from the work based on above-described embodiment
Small node, as the placement node of the target copy, including:
It is determined that each work connection number be less than the disk I/O load of the node of connection number threshold value, internal memory load, cpu load with
And network load;
According to each work connect number be less than connection number threshold value node disk I/O load, internal memory load, cpu load with
And network load and load factor, it is determined that each work connection number is less than the real time load of the node of connection number threshold value, and choose
Placement node of the minimum node of real time load as the target copy.
Specifically, this programme determines that rule determines that each work connection number is less than the section of connection number threshold value using real time load
The real time load of point;The real time load determines that rule is:
W=λio×wio+λmem×wmem+λcpu×wcpu+λband×wband;
Wherein, W is real time load, wioFor disk I/O load, wmemFor internal memory load, wcpuFor cpu load, wbandFor network
Load, λioFor disk specific gravity factor, λmemFor internal memory specific gravity factor, λcpuFor CPU specific gravity factors, λbandFor network bandwidth proportion system
Number, λio+λmem+λcpu+λband=1, and λio、λmem、λcpu、λband∈[0,1]。
Specifically, the real time load of back end can be weighed by multiple indexs, in this programme, born with disk I/O
Carry, internal memory load, cpu load, this programme is illustrated exemplified by network load.Assuming that back end real time load is W, then:
W=λio×wio+λmem×wmem+λcpu×wcpu+λband×wband;
Wherein, wioFor disk I/O load, wmemFor internal memory load, wcpuFor cpu load, wbandFor network load, λio, λmem,
λcpu, λbandThen represent node disk, internal memory, CPU, the proportion shared by network bandwidth when weighing node workload, λio+
λmem+λcpu+λband=1, λio、λmem、λcpu、λband∈ [0,1], further, the selection of the weights in this programme, which uses, plans strategies for
Analytic hierarchy process (AHP) in determines that the weights of wherein real time load are defined as:λcpu=0.153, λmem=0.072, λio=
0.531、λband=0.245.
Replica placement node determining device provided in an embodiment of the present invention is introduced below, copy described below is put
Put node determining device and determine that method can be with cross-referenced with above-described Replica placement node.
Referring to Fig. 2, a kind of Replica placement node determining device based on hadoop provided in an embodiment of the present invention, including:
Target rack server determining module 100, for determining target frame service according to the copy type of target copy
Device;
Cluster determining module 200, for choosing node to be placed from the target rack server, form section to be placed
Point cluster;
Node selection module 300, it is less than connection number threshold for choosing work connection number from the node cluster to be placed
The node of value, and it is less than in the node of connection number threshold value the really minimum node of real time load from work connection number, as institute
State the placement node of target copy.
Wherein, the node selection module includes:
Work connection number determining unit, for determining that the work of each node in the node cluster to be placed connects number;
Average operation connects number computing unit, and the node to be placed is calculated for connecting number according to the work of each node
The average operation connection number of cluster;
Node selection unit, for the average operation to be connected into number as the connection number threshold value, and wait to put from described
Put and the node that work connection number is less than connection number threshold value is chosen in node cluster.
Wherein, the node selection module includes:
Load determining unit, for determine each work connection number be less than the node of connection number threshold value disk I/O load,
Internal memory load, cpu load and network load;
Real time load determining unit, the disk I/O for connecting the node that number is less than connection number threshold value according to each work are born
Load, internal memory load, cpu load and network load and load factor, it is determined that each work connection number is less than connection number threshold value
The real time load of node, and choose placement node of the minimum node of real time load as the target copy.
Based on above-mentioned any embodiment, this programme discloses a kind of Replica placement node based on hadoop and determines equipment,
Including:
Memory, for storing computer program;Processor, realize during for performing the computer program and above-mentioned state pair
The step of this placement node determines method.
This programme also discloses a kind of computer-readable recording medium, and computer is stored with computer-readable recording medium
Program, the computer program realize the step of above-mentioned Replica placement node determines method when being executed by processor.
It should be noted that the storage medium can include:USB flash disk, mobile hard disk, read-only storage (Read-Only
Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disc or CD etc. are various to deposit
Store up the medium of program code.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be and other
The difference of embodiment, between each embodiment identical similar portion mutually referring to.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (10)
1. a kind of Replica placement node based on hadoop determines method, it is characterised in that including:
Target rack server is determined according to the copy type of target copy;
Node to be placed is chosen from the target rack server, forms node cluster to be placed;
Work connection number is chosen from the node cluster to be placed and is less than the node of connection number threshold value, and is connected from the work
Number is less than in the node of connection number threshold value the really minimum node of real time load, the placement node as the target copy.
2. Replica placement node according to claim 1 determines method, it is characterised in that the pair according to target copy
This type determines target rack server, including:
If the copy type of the target copy is first copy, rack server is randomly selected as the target frame
Server;
If the copy type of the target copy is second copy, put from first copy corresponding with the target copy
In other rack servers outside the rack server put, target rack server is chosen;
If the copy type of the target copy is the 3rd copy, judge that first copy is put corresponding to the target copy
Whether the rack server put is identical in the rack server of second Replica placement;If identical, from the target
In other rack servers outside the rack server of second Replica placement corresponding to copy, target frame service is chosen
Device;If differing, using the rack server of second Replica placement corresponding to the target copy as the target frame
Server.
3. Replica placement node according to claim 1 determines method, it is characterised in that described from the node to be placed
The node that work connection number is less than connection number threshold value is chosen in cluster, including:
Determine the work connection number of each node in the node cluster to be placed;
The average operation that the node cluster to be placed is calculated according to the work connection number of each node connects number, will be described average
Work connection number chooses work connection number from the node cluster to be placed and is less than connection number as the connection number threshold value
The node of threshold value.
4. Replica placement node as claimed in any of claims 1 to 3 determines method, it is characterised in that it is described from
The work connection number is less than in the node of connection number threshold value the really minimum node of real time load, as putting for the target copy
Node is put, including:
It is determined that each work connection number is less than disk I/O load, internal memory load, cpu load and the net of the node of connection number threshold value
Network loads;
Number is connected according to each work and is less than the disk I/O load for the node for connecting number threshold value, internal memory load, cpu load and net
Network loads and load factor, it is determined that each work connection number is less than the real time load of the node of connection number threshold value, and chooses real-time
Load placement node of the minimum node as the target copy.
5. Replica placement node according to claim 4 determines method, it is characterised in that the basis each works connection
Number is less than the disk I/O load of the node of connection number threshold value, internal memory load, cpu load and network load and load factor, really
Fixed each work connection number is less than the real time load of the node of connection number threshold value, including:
Determine that rule determines that each work connection number is less than the real time load of the node of connection number threshold value using real time load;It is described
Real time load determines that rule is:
W=λio×wio+λmem×wmem+λcpu×wcpu+λband×wband;
Wherein, W is real time load, wioFor disk I/O load, wmemFor internal memory load, wcpuFor cpu load, wbandBorn for network
Carry, λioFor disk specific gravity factor, λmemFor internal memory specific gravity factor, λcpuFor CPU specific gravity factors, λbandFor network bandwidth proportion system
Number, λio+λmem+λcpu+λband=1, and λio、λmem、λcpu、λband∈[0,1]。
A kind of 6. Replica placement node determining device based on hadoop, it is characterised in that including:
Target rack server determining module, target rack server is determined for the copy type according to target copy;
Cluster determining module, for choosing node to be placed from the target rack server, form node cluster to be placed;
Node selection module, the section of connection number threshold value is less than for choosing work connection number from the node cluster to be placed
Point, and it is less than in the node of connection number threshold value the really minimum node of real time load from work connection number, as the target
The placement node of copy.
7. Replica placement node determining device according to claim 6, it is characterised in that the node selection module bag
Include:
Work connection number determining unit, for determining that the work of each node in the node cluster to be placed connects number;
Average operation connects number computing unit, and the node cluster to be placed is calculated for connecting number according to the work of each node
Average operation connection number;
Node selection unit, for the average operation to be connected into number as the connection number threshold value, and from the section to be placed
The node that work connection number is less than connection number threshold value is chosen in point cluster.
8. the Replica placement node determining device according to claim 6 or 7, it is characterised in that the node selection module
Including:
Load determining unit, for determining that each work connection number is less than disk I/O load, the internal memory of the node of connection number threshold value
Load, cpu load and network load;
Real time load determining unit, the disk I/O load of the node for being less than connection number threshold value according to each work connection number,
Internal memory load, cpu load and network load and load factor, it is determined that each work connection number is less than the node of connection number threshold value
Real time load, and choose placement node of the minimum node of real time load as the target copy.
9. a kind of Replica placement node based on hadoop determines equipment, it is characterised in that including:
Memory, for storing computer program;
Processor, realize that the Replica placement node as described in any one of claim 1 to 5 is true during for performing the computer program
The step of determining method.
10. a kind of computer-readable recording medium, it is characterised in that be stored with computer on the computer-readable recording medium
Program, realize that Replica placement node determines as described in any one of claim 1 to 5 when the computer program is executed by processor
The step of method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711007971.9A CN107729514A (en) | 2017-10-25 | 2017-10-25 | A kind of Replica placement node based on hadoop determines method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711007971.9A CN107729514A (en) | 2017-10-25 | 2017-10-25 | A kind of Replica placement node based on hadoop determines method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107729514A true CN107729514A (en) | 2018-02-23 |
Family
ID=61213909
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711007971.9A Pending CN107729514A (en) | 2017-10-25 | 2017-10-25 | A kind of Replica placement node based on hadoop determines method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107729514A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110032338A (en) * | 2019-03-20 | 2019-07-19 | 华中科技大学 | A kind of data copy laying method and system towards correcting and eleting codes |
CN110083672A (en) * | 2019-04-29 | 2019-08-02 | 广东广量测绘信息技术有限公司 | Geography information big data three-dimension visible sysem |
CN110278233A (en) * | 2018-03-16 | 2019-09-24 | 中移(苏州)软件技术有限公司 | A kind of load regulation method and device |
CN110535898A (en) * | 2018-05-25 | 2019-12-03 | 许继集团有限公司 | Copy storage, completion, node selecting method and management system in big data storage |
CN110674099A (en) * | 2019-09-25 | 2020-01-10 | 新华三大数据技术有限公司 | Copy balancing method and device, storage medium and electronic equipment |
CN110769272A (en) * | 2019-11-19 | 2020-02-07 | 深圳市网心科技有限公司 | Node adjusting method and related equipment thereof |
CN112632621A (en) * | 2020-12-30 | 2021-04-09 | 中国移动通信集团江苏有限公司 | Data access method, device, equipment and computer storage medium |
Citations (3)
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 |
CN103533058A (en) * | 2013-10-17 | 2014-01-22 | 南京大学镇江高新技术研究院 | HDFS (Hadoop distributed file system)/Hadoop storage cluster-oriented resource monitoring system and HDFS/Hadoop storage cluster-oriented resource monitoring method |
CN105025086A (en) * | 2015-06-11 | 2015-11-04 | 江南大学 | A cloud safe storing/retrieving method targeted at mobile intelligent terminals |
-
2017
- 2017-10-25 CN CN201711007971.9A patent/CN107729514A/en active Pending
Patent Citations (3)
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 |
CN103533058A (en) * | 2013-10-17 | 2014-01-22 | 南京大学镇江高新技术研究院 | HDFS (Hadoop distributed file system)/Hadoop storage cluster-oriented resource monitoring system and HDFS/Hadoop storage cluster-oriented resource monitoring method |
CN105025086A (en) * | 2015-06-11 | 2015-11-04 | 江南大学 | A cloud safe storing/retrieving method targeted at mobile intelligent terminals |
Non-Patent Citations (1)
Title |
---|
蔡燕冬 等: ""一种优化的Hadoop副本放置策略"", 《软件与算法》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110278233A (en) * | 2018-03-16 | 2019-09-24 | 中移(苏州)软件技术有限公司 | A kind of load regulation method and device |
CN110278233B (en) * | 2018-03-16 | 2022-06-03 | 中移(苏州)软件技术有限公司 | Load adjusting method and device |
CN110535898A (en) * | 2018-05-25 | 2019-12-03 | 许继集团有限公司 | Copy storage, completion, node selecting method and management system in big data storage |
CN110535898B (en) * | 2018-05-25 | 2022-10-04 | 许继集团有限公司 | Method for storing and complementing copies and selecting nodes in big data storage and management system |
CN110032338A (en) * | 2019-03-20 | 2019-07-19 | 华中科技大学 | A kind of data copy laying method and system towards correcting and eleting codes |
CN110083672A (en) * | 2019-04-29 | 2019-08-02 | 广东广量测绘信息技术有限公司 | Geography information big data three-dimension visible sysem |
CN110674099A (en) * | 2019-09-25 | 2020-01-10 | 新华三大数据技术有限公司 | Copy balancing method and device, storage medium and electronic equipment |
CN110769272A (en) * | 2019-11-19 | 2020-02-07 | 深圳市网心科技有限公司 | Node adjusting method and related equipment thereof |
CN112632621A (en) * | 2020-12-30 | 2021-04-09 | 中国移动通信集团江苏有限公司 | Data access method, device, equipment and computer storage medium |
CN112632621B (en) * | 2020-12-30 | 2024-08-27 | 中国移动通信集团江苏有限公司 | Data access method, device, equipment and computer storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107729514A (en) | A kind of Replica placement node based on hadoop determines method and device | |
CN108009016B (en) | Resource load balancing control method and cluster scheduler | |
CN104468803B (en) | A kind of virtual data center method for mapping resource and equipment | |
US10623481B2 (en) | Balancing resources in distributed computing environments | |
JP5998206B2 (en) | Scalable centralized dynamic resource distribution in cluster data grids | |
CN104038540B (en) | Method and system for automatically selecting application proxy server | |
CN103338252B (en) | Realizing method of distributed database concurrence storage virtual request mechanism | |
US9888065B2 (en) | Defragmenting clusters with reserved resources | |
CN103607424B (en) | Server connection method and server system | |
CN103595805A (en) | Data placement method based on distributed cluster | |
CN110162388A (en) | A kind of method for scheduling task, system and terminal device | |
CN106843745A (en) | Capacity expansion method and device | |
CN107766159A (en) | A kind of metadata management method, device and computer-readable recording medium | |
CN107562531A (en) | A kind of data balancing method and device | |
CN102637138B (en) | A kind of method for calculating scheduling virtual machine | |
CN106406987A (en) | Task execution method and apparatus in cluster | |
CN104679594B (en) | A kind of middleware distributed computing method | |
CN107346267A (en) | A kind of cpu performance optimization method and device based on NUMA architecture | |
US10666719B2 (en) | Using read weight affinities to optimize the reading of block cache data | |
CN103997515B (en) | Center system of selection and its application are calculated in a kind of distributed cloud | |
JP2014186364A (en) | Dispersion system | |
Liao et al. | A QoS-aware dynamic data replica deletion strategy for distributed storage systems under cloud computing environments | |
CN109491788A (en) | A kind of virtual platform implementation of load balancing and device | |
CN104657195B (en) | A kind of resource binding method and device | |
CN105915626B (en) | A kind of data copy initial placement method towards cloud storage |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180223 |
|
RJ01 | Rejection of invention patent application after publication |