CN106575267A - Process distributing device, process distributing program, and data processing system - Google Patents
Process distributing device, process distributing program, and data processing system Download PDFInfo
- Publication number
- CN106575267A CN106575267A CN201480081125.XA CN201480081125A CN106575267A CN 106575267 A CN106575267 A CN 106575267A CN 201480081125 A CN201480081125 A CN 201480081125A CN 106575267 A CN106575267 A CN 106575267A
- Authority
- CN
- China
- Prior art keywords
- server
- data
- identifier
- order
- unit
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Transfer Between Computers (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Computer And Data Communications (AREA)
Abstract
This load balancing device (200) stores an association table file in which respective identifiers of server devices (300) each storing content data to be used in data processing are associated with processing commands each designating data processing. The load balancing device receives a processing request packet from a client device (110), and selects, from the association table file, the identifier of a server device associated with the same processing command as the processing command included in the processing request packet. Then, the load balancing device transmits a distribution packet including the processing command to the server device identified by the identifier selected from the association table file.
Description
Technical field
The present invention relates to multiple server units are distributed with the technology of data processing.
Background technology
Exist with client terminal device, network (web) server unit, 3 layer network services of database server devices
Device system.In 3 layer network server systems, network service device is also referred to as front-end server.Also, database service
Device device is also referred to as back-end server.
Patent documentation 1 and non-patent literature 1 disclose by load equalizer is arranged in network server system to divide
Dissipate the technology of the load of multiple front-end servers.
Multiple data processings are distributed to multiple front-end servers by load equalizer, so that the load of each front-end server
It is impartial.
But, although load equalizer can disperse the load of multiple front-end servers, multiple rear ends cannot be disperseed
The load of server.Therefore, load equalizer cannot fully dispersed load in system entirety.
Non-patent literature 2 discloses the scatter storage technique of the load for disperseing multiple storage devices.
In scatter storage technique, guarantee respectively in a plurality of storage devices to disperse the memory area of storage, concentrate each
Constructing imaginary storage device, the multiplexing in imaginary storage device stores many numbers to the memory area that individual dispersion storage is used
According to.
In the case where scatter storage technique is applied to network server system, data processing may be assigned to and not deposit
The back-end server of storage target data.In this case, the back-end server of data processing is allocated for access target data, need
To be communicated with the back-end server of the target data that is stored with.And, the communication process becomes expense and prolongs data processing
Late, the handling capacity of system is reduced.
Prior art literature
Patent documentation
Patent documentation 1：Japanese Unexamined Patent Publication 2003-196178 publications
Non-patent literature
Non-patent literature 1：Wensong Zhang、“Linux Virtual Server for Scalable Network
Services”,Ottawa Linux Symposium 2000.
Non-patent literature 2：Vijay Bellur、“GlusterFS-Architecture&Roadmap”、[online]、
[August is retrieved on the 4th within 2014], the Internet (URL：http：//www.gluster.org/community/documentation/
Images/8/80/GlusterFS_Architecture_%26_Roadmap-Vijay_Bel lur-LinuxCon_EU_
2013.pdf)
The content of the invention
Invention problem to be solved
It is an object of the present invention to can be to the server unit distribution number of the content-data used in data storage process
According to process.
Means for solving the problems
The process distributor of the present invention has：Correspondence list file storage part, used in data storage is processed by its storage
Content-data server unit identifier server and specified data processing the corresponding table that is mapped of process order
File；Order receiving unit is processed, which is accepted for specifying the first process of the first data processing for using first content data to order
Order；Correspondence table management department, its select from the correspondence list file and with by it is described process that order receiving unit accepts first
Reason order identical processes the corresponding identifier server of order；And dispenser is processed, its request is using by the correspondence table
The server unit of the identifier server mark that management department is selected performs first data processing.
Invention effect
In accordance with the invention it is possible to the server unit distribution data of the content-data used in data storage process
Reason.
Description of the drawings
Fig. 1 is the synoptic diagram of the load system 100 in embodiment 1.
Fig. 2 is the functional structure chart of the load spreading device 200 in embodiment 1.
Fig. 3 is the functional structure chart of the server unit 300 in embodiment 1.
Fig. 4 is the flow chart of the action for illustrating the load spreading device 200 in embodiment 1.
Fig. 5 is the figure of that illustrates corresponding list file 291 in embodiment 1.
Fig. 6 is the figure of that illustrates server guide look file 292 in embodiment 1.
Fig. 7 is the flow chart of the action for illustrating the server unit 300 in embodiment 1.
Fig. 8 is the flow chart of the action for illustrating the server unit 300 in embodiment 1.
Fig. 9 is the figure of that illustrates data configuration file 391 in embodiment 1.
Figure 10 is the hardware structure diagram of the load spreading device 200 in embodiment 1.
Figure 11 is the structure chart of the client server system 101 in embodiment 2.
Figure 12 is the functional structure chart of the client terminal device 110 in embodiment 2.
Specific embodiment
Embodiment 1
Illustrate to distribute the server unit of the content-data used in data storage process the mode of data processing.
Fig. 1 is the synoptic diagram of the load system 100 in embodiment 1.
The summary of the load system 100 in embodiment 1 is illustrated according to Fig. 1.
Load system 100 (one of data handling system) is will to produce from more than one client terminal device 110
The multiple process request system that is distributed to multiple server units (300A～300D).
Load system 100 have more than one client terminal device 110, multiple server units (300A～
300D), load spreading device 200 (processing of distributor).These devices are communicated via network 109.
In the case where server unit is not differentiated between, each server unit is denoted as into server unit 300.
Client terminal device 110 is the computer that request performs data processing.Hereinafter, the data used in data processing are claimed
Make content-data.
Multiple server units (300A～300D) are carried out the computer of data processing.
It is many that the storage device (server storage section 390 of aftermentioned Fig. 3) that each server unit has constitutes dispersion storage
The dispersion memorizer (the dispersion memory area 399 of aftermentioned Fig. 3) of individual content-data.
Multiple content-data dispersions are stored in multiple server units.Also, each content-data multiplexing is stored in 2
In server unit above.
In FIG, content-data A381 and content-data B382 multiplexings are stored in server unit 300A and server dress
Put in 300C.Also, content-data C383 and content-data D384 multiplexings are stored in server unit 300B and server unit
In 300D.
Load spreading device 200 is that multiple process requests that request performs data processing are distributed to multiple server units
Computer.Load spreading device 200 is also referred to as load equalizer.
Load spreading device 200 receives to process request from client terminal device 110, and arbitrary server unit 300 is distributed
Request is processed, receives result from the server unit 300, result is responded to client terminal device 110.
One of the action of load system 100 is illustrated according to (1)～(5) of Fig. 1.
(1) request of client terminal device 110 performs the data processing C using content-data C383.
(2) load spreading device 200 will ask the process for performing data processing C to request assignment to arbitrary server dress
Put.Now, the content-data C383 used in data processing C is stored in server unit 300B and server unit 300D,
It is server unit 300B or server unit accordingly, as the appropriate server unit of the distribution destination for processing request
300D。
In the case where appropriate server unit is learnt, the process request of data processing C is divided by load spreading device 200
Dispensing server unit 300B or server unit 300D.
In the case where appropriate server unit is not known, the server dress lighter to load of load spreading device 200
Put the process request of distribution data processing C.The lighter server dress of load can be determined by existing load technology
Put.
Here, suppose that load spreading device 200 is not known appropriate server unit and asks the process of data processing C
Distribute to server unit 300A.
(3) server unit 300A selects storage content data according to data configuration file described later 391 (with reference to Fig. 3)
The server unit 300B (or server unit 300D) of C383.Then, server unit 300A by with server unit
300B (or server unit 300D) carries out communication to access content-data C383, performs data processing C.
(4) server unit 300A processes the implementing result and access information of C to 200 response data of load spreading device.
It is server that the access information is the appropriate server unit of the distribution destination of the process request for being denoted as data processing C
The information of device 300B (or server unit 300D).
(5) 200 data storage of load spreading device processes the access information of C, to the process of 110 response data of client terminal device
The implementing result of C.
In above-mentioned (5), the access information of data processing C is stored in load spreading device 200.Therefore, load
The process of the data processing C for hereafter producing can be requested assignment to server unit 300B (or server units by device 200
300D)。
Data processing C process request be assigned to server unit 300B in the case of, it is not necessary to as described above (3) that
Sample makes server unit 300A be communicated with server unit 300B.Therefore, will not produce and communicate between server unit
Process postpone, the handling capacity of system is improved.
Fig. 2 is the functional structure chart of the load spreading device 200 in embodiment 1.
The functional structure of the load spreading device 200 in embodiment 1 is illustrated according to Fig. 2.But, load
The functional structure of device 200 can not also be identical with the functional structure shown in Fig. 2.
Load spreading device 200 has request receiving unit 210 (processing of order receiving unit), correspondence table management department
220th, dispenser 230 (processing of dispenser), 290 (the one of correspondence list file storage part of load storage part are asked
Example).
Load storage part 290 storage load spreading device 200 use, generate or input and output data.For example, bear
The lotus dispersion storage of storage part 290 correspondence list file 291 and server guide look file 292 etc..
Correspondence list file 291 data storage is processed used in content-data server unit 300 identifier with
The process order of specified data processing is mapped.One of correspondence list file 291 is described below (with reference to Fig. 5).
The information such as IP address are mapped by server guide look file 292 with the identifier of server unit 300.Server
One of guide look file 292 is described below (with reference to Fig. 6).
Request receiving unit 210 receives the process request comprising the process order for specifying data processing from client terminal device 110
Packet, sends the process response point of the implementing result data comprising the implementing result for representing data processing to client terminal device 110
Group.
The packet that request packet is that request performs data processing is processed, the execution that respond packet is in response to data processing is processed
As a result packet.
There is correspondence table reference portion 221 (of corresponding server selector) and correspondence table to compile for correspondence table management department 220
Collect portion 222 (of corresponding server register), management correspondence list file 291.
Correspondence table select from correspondence list file 291 with reference to portion 221 and with process the process order that includes during request is grouped
The identifier of corresponding server unit 300 is ordered in identical process.
Access information data described later are asked the process order included in packet corresponding with processing by correspondence table editorial office 222
Get up to be registered in correspondence list file 291.
Request dispenser 230 is to the server unit 300 for utilizing the identifier mark selected with reference to portion 221 by correspondence table
Send comprising the allocation packets for processing order.Then, dispenser 230 is asked to receive at comprising data from the server unit 300
The implementing result packet of the implementing result of reason.
Allocation packets are to ask the packet of execution data processing, and implementing result packet is in response to the implementing result of data processing
Packet.
Fig. 3 is the functional structure chart of the server unit 300 in embodiment 1.
The functional structure of the server unit 300 in embodiment 1 is illustrated according to Fig. 3.But, server unit
300 functional structure can not also be identical with the functional structure shown in Fig. 3.
There is server unit 300 WEB server portion 301, DB servers portion 320, dispersion storage management portion 330 (to access
One of request unit), access information generating unit 340, server storage section 390 (deposit by content-data storage portion, data configuration file
One of storage portion).Also, WEB server portion 301 has WEB application portion 310.
390 storage server device 300 of server storage section uses, generate or input and output data.
For example, 390 data storage configuration file 391, server of server storage section guide look file 392 and content groups of data
380 etc..
Data configuration file 391 is by the identifier of the server unit 300 of the content-data that is stored with and the content-data
Identifier is mapped.One of data configuration file 391 is described below (with reference to Fig. 9).
The information such as IP address are mapped by server guide look file 392 with the identifier of server unit 300.Server
Guide look file 392 is the server guide look 292 identical file of file with load spreading device 200 (with reference to Fig. 2).
Content groups of data 380 is more than one content-data.Content groups of data 380 is stored in be guaranteed to disperse storage
In memory area.Hereinafter, will ensure that the memory area of dispersion storage referred to as disperses memory area 399.
WEB server portion 301 is performed for making server unit 300 as the WEB server of WEB server function
Program.
WEB application portion 310 has to perform asks receiving unit 311, processing execution unit 312, implementing result response portion 313.
Perform request receiving unit 311 allocation packets are received from load spreading device 200.
Processing execution unit 312 performs the data processing specified by the process order included in allocation packets.
Implementing result response portion 313 sends holding comprising the implementing result for representing data processing to load spreading device 200
The implementing result packet of row result data and the access information data generated by access information generating unit 340.
DB servers portion 320 is performed for making server unit 300 play work(as database server (DB servers)
The DB server programs of energy.
The content-data that dispersion storage management portion 330 is stored in accessing dispersion memory area 399.
In the case of the content-data for desiring access to is not stored in dispersion memory area 399, dispersion storage management portion 330
The identifier of server unit corresponding with the identifier of the content-data 300 is selected from data configuration file 391.Then,
Ask to access content-data using the server unit 300 of the identifier mark selected in dispersion storage management portion 330.
In the case where the access of content-data is requested from other server units 300, dispersion storage management portion 330 is visited
The content-data stored in asking dispersion memory area 399.
Access information generating unit 340 generates access information data, and the access information packet is made in processing containing data storage
The identifier of the server unit 300 of content-data.
Fig. 4 is the flow chart of the action for illustrating the load spreading device 200 in embodiment 1.
Illustrated according to actions of the Fig. 4 to the load spreading device 200 in embodiment 1.But, load spreading device
200 action can not also be identical with the action illustrated according to Fig. 4.
In S110, client terminal device 110 sends the process request that request performs data processing to load spreading device 200
Packet.
Then, 210 reception processing of the request receiving unit request packet of load spreading device 200.Process request packet to include
The process order of specified data processing.
For example, the GET methods of HTTP request are process order one, and the packets of the IP comprising the HTTP request are to process to ask
Ask packet one.HTTP is the abbreviation of HyperText Transfer Protocol (HTML (Hypertext Markup Language)).Also, IP
It is the abbreviation of Internet Protocol (Internet protocol).
After S110, process enters S120.
In S120, whether correspondence table is judged to be registered with correspondence list file 291 and is divided with request is processed with reference to portion 221
The process order identical included in group processes the corresponding identifier server of order.
Corresponding identifier server is referred to as into corresponding server identifier, by what is identified using corresponding server identifier
Server unit 300 is referred to as corresponding server.
For example, it is assumed that the process order included in processing request packet is HTTP as " GET/aaa HTTP/1.1 " asking
Ask.In this case, 2 corresponding server identifiers (server A, services are registered with correspondence list file 291 (with reference to Fig. 5)
Device C).
And, it is assumed that the process order included in processing request packet is that HTTP as " GET/ccc HTTP/1.1 " please
Ask.In this case, the unregistered corresponding server identifier in correspondence list file 291 (with reference to Fig. 5).
(YES) in the case of corresponding server identifier is registered with correspondence list file 291, process enters S121.
In the case of unregistered corresponding server identifier in correspondence list file 291 (no), process enters S122.
Fig. 5 is the figure of that illustrates corresponding list file 291 in embodiment 1.
According to Fig. 5 to embodiment 1 in of corresponding list file 291 illustrate.
Correspondence list file 291 will access field, access server hurdle, corresponding server hurdle and be mapped with request hurdle.
Request hurdle represents process order.
Access field to represent by the identifier of the content for processing the Data processing access that order is specified.
Access server hurdle represents that in more than one corresponding server, last time Data processing have accessed access number
According to corresponding server identifier.
Corresponding server hurdle represents the respective identifier of more than one corresponding server.
In S121 (with reference to Fig. 4), request dispenser 230 is selected from more than one corresponding server identifier arbitrarily
Corresponding server identifier.Now, dispenser 230 is asked to select from more than one corresponding server identifier negative
The identifier of the lighter server unit 300 of lotus.Can be by the service lighter to determine load of existing load technology
Device device 300.
The corresponding server identifier selected referred to as is distributed destination mark symbol, will be using distribution destination mark symbol
The server unit 300 of mark referred to as distributes destination's server.
After S121, process enters S130.
In S122, ask dispenser 230 that multiple servers of registration in file 292 (with reference to Fig. 6) are had a guide look of from server
Arbitrary identifier server is selected in identifier.Now, dispenser 230 is asked to select from multiple identifier servers
The identifier of the lighter server unit 300 of load.Can be by the existing load technology clothes lighter to determine load
Business device device 300.
The identifier server selected referred to as is distributed destination mark symbol, will be using distribution destination mark symbol mark
Server unit 300 referred to as distribute destination's server.
After S122, process enters S130.
Fig. 6 is the figure of that illustrates server guide look file 292 in embodiment 1.
One of the server guide look file 292 in embodiment 1 is illustrated according to Fig. 6.
The Informations such as IP address are mapped by server guide look file 292 with server hurdle.
Server hurdle represents the identifier of the server unit 300 that load system 100 has.
IP address hurdle represents the IP address of server unit 300.
In S130 (with reference to Fig. 4), request dispenser 230 from server guide look file 292 (with reference to Fig. 6), select and
IP address corresponding with the distribution destination mark symbol identical identifier server selected in S121 or S122.
Request dispenser 230 uses the IP address selected as sending destination address, and generation is sent to distribution destination
The packet of server.The packet is referred to as into allocation packets.
Allocation packets are the packets that request distribution destination server performs data processing.Allocation packets include process request
The process order included in packet.
Request dispenser 230 sends allocation packets to distribution destination server.
After S130, process enters S131.
In S131, request dispenser 230 is received comprising implementing result data from distribution destination server and accesses letter
The implementing result packet of breath data.
Implementing result tables of data shows the implementing result of data processing.
Access information packet data identifier containing access, access server identifier, corresponding server group identifier.
Access data identifier and represent that identification is accessed for the identifier of content-data.
Access server identifier represents the identifier of the server unit 300 that have accessed content-data.
Corresponding server group identifier is more than one corresponding server identifier.Corresponding server identifier is storage
With the 300 respective identifier of more than one server unit for being accessed for content-data identical content-data.
After S131, process enters S140.
In S140, correspondence table editorial office 222 is using the process order and implementing result point included in process request packet
The access information data (accessing data identifier, access server identifier, corresponding server group identifier) included in group, such as
Under it is described editor correspondence list file 291.
It is registered with correspondence list file 291 (with reference to Fig. 5) and asks what is included in packet comprising with processing in request hurdle
In the case of processing the respective record that order identical processes order, correspondence table editorial office 222 is according to access information data to phase
Should record and be updated.That is, correspondence table editorial office 222 is carried out to the access field of respective record using data identifier is accessed
Update, the access server hurdle of respective record is updated using access server identifier, identified using corresponding server
Symbol group is updated to the corresponding server hurdle of respective record.
Ask what is included in packet comprising with processing during request hurdle is not registered in correspondence list file 291 (with reference to Fig. 5)
In the case of processing the respective record that order identical processes order, correspondence table editorial office 222 uses and processes order and access letter
Breath data genaration new record.That is, correspondence table editorial office 222 is generated in request hurdle comprising processing order, in field is accessed
Comprising accessing data identifier, in access server hurdle comprising access server identifier, include in corresponding server hurdle
The new record of corresponding server group identifier.Then, correspondence table editorial office 222 is registered in new record in correspondence list file 291.
After S140, process enters S150.
In S150, receiving unit 210 is asked to generate the process respond packet of the implementing result that response data is processed, to conduct
Process the 110 transmission processe respond packet of client terminal device of request packet transmission side.Respond packet is processed comprising implementing result point
The implementing result data included in group.
After S150, load spreading device 200 is for a release for processing request packet.
Fig. 7, Fig. 8 are the flow charts of the action for illustrating the server unit 300 in embodiment 1.
Illustrated according to actions of the Fig. 7 and Fig. 8 to the server unit 300 in embodiment 1.But, server dress
The action for putting 300 can not also be identical with the action illustrated according to Fig. 7 and Fig. 8.
In S210, perform request receiving unit 311 and receive comprising the allocation packets for processing order.
After S210, process enters S220.
In S220, processing execution unit 312 starts to perform process logic according to the process order included in allocation packets.
It is the program for being also called module, process or function etc. to process logic, represents the order of data processing.For example, locate
Reason logic includes the order for generating the inquiry for being used to accessing content-data.
For example, in the case where process order is " GET/aaa HTTP/1.1 ", processing execution unit 312 selects multiple process
The process logic of the GET methods in logic, uses " aaa " as parameter, starts to perform process logic.
After S220, process enters S230.
In S230, processing execution unit 312 is generated for accessing the inquiry of content-data according to logic is processed, and inquiry is turned
Give DB servers portion 320.
After S230, process enters S231.
In S231, DB servers portion 320 receives inquiry, generates the access for the content-data specified in access challenge
Visit order is handed to dispersion storage management portion 330 by order.
After S231, process enters S232.
In S232, dispersion storage management portion 330 accepts the interview order, judges whether deposit in dispersion memory area 399
The content-data specified in containing visit order.The content-data specified in visit order is referred to as into corresponding contents.
Be stored with dispersion memory area 399 (YES) in the case of corresponding contents, and process enters S233.
In the case of corresponding contents are not stored in dispersion memory area 399 (no), process enters S235 (with reference to Fig. 8).
In S233, dispersion storage management portion 330 is according to visit order, the phase stored in accessing dispersion memory area 399
Answer content.
For example, storage management portion 330 is disperseed to read corresponding contents from dispersion memory area 399.
Dispersion storage management portion 330 implementing result of visit order is handed to into DB servers portion 320.For example, access life
The implementing result of order includes the corresponding contents read from dispersion memory area 399.
DB servers portion 320 accepts the interview the implementing result of order, generates inquiry using the implementing result of visit order
The implementing result of inquiry is handed to processing execution unit 312 by implementing result.For example, the implementing result of inquiry is comprising depositing from dispersion
The corresponding contents read in storage area domain 399.
After S233, process enters S234.
In S234, storage management portion 330 is disperseed to select from data configuration file 391 (with reference to Fig. 9) and interior with corresponding
Hold the corresponding identifier server group of identifier of identical content-data.For example, in the case where corresponding contents are content A,
Dispersion storage management portion 330 selects 2 identifier servers (server A, services from data configuration file 391 (with reference to Fig. 9)
Device C).
Then, disperse storage management portion 330 that the identifier of identifier server group and corresponding contents is handed to access letter
Breath generating unit 340.
Access information generating unit 340 receives the identifier of identifier server group and corresponding contents, generates access information number
According to access information data are handed to implementing result response portion 313.
Used as data identifier is accessed, comprising distribution, destination services identifier of the access information packet containing corresponding contents
, used as access server identifier, the identifier server group comprising acceptance is used as corresponding server identifier for the identifier of device
Group.
After S234, process enters S240.
Fig. 9 is the figure of that illustrates data configuration file 391 in embodiment 1.
One of the data configuration file 391 in embodiment 1 is illustrated according to Fig. 9.
The Informations such as content are mapped by data configuration file 391 with server hurdle.
Server represents 300 respective identifier of more than one server unit.
The respective identifier of more than one content-data stored in content representation server unit 300.
In S240 (with reference to Fig. 7), processing execution unit 312 receives the implementing result of inquiry from DB servers portion 320.
Processing execution unit 312 performs data processing according to the implementing result inquired.For example, processing execution unit 312 is used
The corresponding contents included in the implementing result of inquiry generate WEB page.
Then, processing execution unit 312 terminates to perform data processing, generates and represents that the implementing result of data processing performs knot
Fruit data.For example, processing execution unit 312 generates the implementing result data comprising WEB page generated by data processing.
After S240, process enters S250.
In S250, implementing result response portion 313 is accepted the interview information data from access information generating unit 340.
Then, implementing result response portion 313 generates the implementing result comprising implementing result data and access information data point
Group, sends implementing result packet to load spreading device 200.
After S250, release of the server unit 300 for an allocation packets.
In S235 (with reference to Fig. 8), dispersion storage management portion 330 is selected from data configuration file 391 (with reference to Fig. 9)
And identifier server group corresponding with the identifier of the corresponding contents identical content-data specified in visit order.
Then, disperse storage management portion 330 that arbitrary identifier server is selected from identifier server group.Now,
The identifier of the server unit 300 that dispersion storage management portion 330 can select load lighter from identifier server group.Energy
Enough server units 300 lighter to determine load by existing load technology.
The identifier server selected is referred to as into corresponding server identifier, will be identified using corresponding server identifier
Server unit 300 be referred to as corresponding server.
After S235, process enters S236.
In S236, dispersion storage management portion 330 is selected from server guide look file 392 and is identified with corresponding server
The corresponding IP address of symbol identical identifier server.Server guide look file 392 is and server illustrated in fig. 6 guide look is literary
292 identical file of part.
Dispersion storage management portion 330 uses the IP address selected as sending destination address, and generation is sent to correspondence clothes
The packet of business device.The packet is referred to as into access request packet.
Access request packet is the packet for asking corresponding server to perform visit order.Access request packet includes access life
Order.
Dispersion storage management portion 330 sends access request packet to corresponding server.
After S236, process enters S237.
In S237, the dispersion storage management portion 330 of corresponding server receives access request packet.
According to the visit order included in access request packet, it is right to access in the dispersion storage management portion 330 of corresponding server
The corresponding contents stored in the dispersion memory area 399 for answering server.
Then, the dispersion storage management portion 330 of corresponding server generates the access sound of the implementing result comprising visit order
Should be grouped, access response packet is sent to the server unit 300 as access request packet transmission side.
After S237, process enters S238.
In S238, the dispersion storage management portion 330 for distributing destination's server receives access response packet.
In distribution destination server, the visit order for disperseing storage management portion 330 to include during access response is grouped
Implementing result hand to DB servers portion 320.
DB servers portion 320 accepts the interview the implementing result of order, generates inquiry using the implementing result of visit order
The implementing result of inquiry is handed to processing execution unit 312 (identical with S233) by implementing result.
After S238, process enters S239.
In S239, distribute the dispersion storage management portion 330 of destination's server by the identifier of corresponding contents, from data
The identifier server group selected in configuration file 391, the corresponding server mark selected from identifier server group
Symbol hands to access information generating unit 340.
Access information generating unit 340 receives the identifier of corresponding contents, identifier server group, corresponding server mark
Symbol, generates access information data, access information data is handed to implementing result response portion 313 (identical with S234).
Access information packet contains the identifier of corresponding contents as data identifier is accessed, comprising identifier server group
As corresponding server group identifier, comprising corresponding server identifier as access server identifier.
After S239, process enters S240 (with reference to Fig. 7).S240 later process is as according to as Fig. 7 explanations.
Figure 10 is the hardware structure diagram of the load spreading device 200 in embodiment 1.
The hardware configuration of the load spreading device 200 in embodiment 1 is illustrated according to Figure 10.But, load point
200 hardware configuration put in bulk can not also be identical with the structure shown in Figure 10.
Load spreading device 200 is with arithmetic unit 901, auxilary unit 902, main storage means 903, communication dress
Put the computer of 904 and input/output unit 905.
Arithmetic unit 901, auxilary unit 902, main storage means 903, communicator 904 and input/output unit
905 are connected with bus 909.
Arithmetic unit 901 is carried out the CPU (Central Processing Unit) of program.
Auxilary unit 902 is, for example, ROM (Read Only Memory), flash memory or hard disk unit.
Main storage means 903 are, for example, RAM (Random Access Memory).
Communicator 904 is in a wired or wireless fashion via the Internet, LAN (LAN), telephone wire road network or other nets
Network is communicated.
Input/output unit 905 is, for example, mouse, keyboard, display equipment.
Program storage is in auxilary unit 902.
For example, operating system (OS) is stored in auxilary unit 902.Also, the work(that realization is illustrated as "～portion "
The program storage of energy is in auxilary unit 902.
Program storage is performed by arithmetic unit 901 in downloading to main storage means 903 in auxilary unit 902.
"～judgement ", "～judgement ", "～extraction ", "～detection ", "～setting ", "～registration ", "～
Selection ", "～generation ", "～input ", "～output " etc. represent the information of result, data, file, signal value
Or storage of variable values is in main storage means 903 or auxilary unit 902.
Client terminal device 110 and server unit 300 with 200 identical hardware configuration of load spreading device.
According to embodiment 1, for example, play following effect.
Load spreading device 200 has correspondence list file 291, thereby, it is possible to the process of data processing is requested assignment to
Appropriate server unit 300.That is, load spreading device 200 can be (corresponding to the content-data used in data storage process
Content) server unit 300 distribution data processing process request.
The corresponding contents if server unit 300 of the distribution destination asked as process is stored with, as process
The server unit 300 of the distribution destination of request need not carry out communication to other server units 300 to access in corresponding
Hold.Therefore, will not produce and postpone with the process communicated between server unit, the handling capacity of system is improved.
Embodiment 1 is of the mode of load system 100.
That is, a part for the structural element that load system 100 can not also be illustrated in having embodiment 1.Also,
Load system 100 can also have unaccounted structural element in embodiment 1.
Embodiment 2
The mode of the not client server system with load spreading device 200 is illustrated.
Below, mainly the item different from embodiment 1 is illustrated.The item for omitting the description and 1 phase of embodiment
Together.
Figure 11 is the structure chart of the client server system 101 in embodiment 2.
In fig. 11, client server system 101 (of data handling system) is with more than one client
Device 110 and multiple server units 300.
Figure 12 is the functional structure chart of the client terminal device 110 in embodiment 2.
In fig. 12, same with load spreading device 200 (with reference to Fig. 2), client terminal device 110 has request receiving unit
210th, correspondence table management department 220, request dispenser 230.
Also, client terminal device 110 has the client of storage correspondence list file 291 and server guide look file 292 etc.
Storage part 119 (one of correspondence list file storage part).
The action of the load spreading device 200 in the action of the client terminal device 110 in embodiment 2 and embodiment 1
(with reference to Fig. 4) is identical.
But, in S110, request receiving unit 210 replaces reception processing request packet and accepts user and be input to client
The process order of end device 110.
Also, in S150, request receiving unit 210 replaces transmission processe respond packet and implementing result is shown in picture
Data.
In embodiment 2, client terminal device 110 can play identical with the load spreading device 200 in embodiment 1
Effect.That is, client terminal device 110 can distribute to the server unit 300 of the content-data used in data storage process
The process request of data processing.
Embodiment 2 is of the mode of client server system 101.
That is, a part for the structural element that client server system 101 can not also be illustrated in having embodiment 2.
Also, client server system 101 can also have unaccounted structural element in embodiment 2.
Used in each embodiment, the processing sequence of the explanation such as flow chart is the process of the methods and procedures of each embodiment
One of order.The methods and procedures of each embodiment can also by it is a part of with each embodiment in the processing sequence that illustrates
Different processing sequences is realizing.
In each embodiment, "～portion " can be rewritten into "～process ", "～step ", "～program ", "～device " etc..
Label declaration
100：Load system；101：Client server system；109：Network；110：Client terminal device；119：Visitor
Family end storage part；200：Load spreading device；210：Request receiving unit；220：Correspondence table management department；221：Correspondence table reference portion；
222：Correspondence table editorial office；230：Request dispenser；290：Load storage part；291：Correspondence list file；292：Server
Guide look file；300：Server unit；301：WEB server portion；310：WEB application portion；311：Perform request receiving unit；312：
Processing execution unit；313：Implementing result response portion；320：DB servers portion；330：Dispersion storage management portion；340：Access information
Generating unit；380：Content groups of data；381：Content-data A；382：Content-data B；383：Content-data C；384：Content-data
D；390：Server storage section；391：Data configuration file；392：Server has a guide look of file；399：Dispersion memory area；901：
Arithmetic unit；902：Auxilary unit；903：Main storage means；904：Communicator；905：Input/output unit；909：Always
Line.
Claims (7)
1. it is a kind of to process distributor, it is characterised in that the process distributor has：
Correspondence list file storage part, which stores the server of the server unit of the content-data used in data storage is processed
The corresponding list file that identifier is mapped with the process order of specified data processing；
Order receiving unit is processed, which is accepted for specifying the first process of the first data processing for using first content data to order
Order；
Correspondence table management department, its select from the correspondence list file and with by it is described process that order receiving unit accepts first
Reason order identical processes the corresponding identifier server of order；And
Dispenser is processed, server dress of its request using the identifier server mark selected by the correspondence table management department
Put execution first data processing.
2. process distributor according to claim 1, it is characterised in that
Do not include and order identical is processed with described first in the correspondence list file process the corresponding server identification of order
Symbol, thus in the case of the non-selected identifier server of the correspondence table management department, the process dispenser asks multiple services
Any server unit in device device performs first data processing.
3. process distributor according to claim 1 and 2, it is characterised in that
The dispenser that processes accepts described first from the server unit of the request destination as first data processing
The identifier server of the server unit of the implementing result and the storage first content data of data processing,
It is right that the identifier server accepted by the process dispenser is processed order with described first by the correspondence table management department
Should get up to be registered in the correspondence list file.
4. process distributor according to claim 3, it is characterised in that
In the case where multiple identifier servers have been accepted by the process dispenser, the correspondence table management department will be described more
Individual identifier server is mapped with the described first process order and is registered in the corresponding list file.
5. the process distributor according to any one in Claims 1 to 4, it is characterised in that
Existing multiple and process order identical with described first and process the corresponding identifier server of order, thus it is described right
In the case of answering table management department to have selected multiple identifier servers, the process dispenser request is using the plurality of server
The server unit of any identifier server mark in identifier performs first data processing.
6. a kind of to process distribution program, the process distribution program uses correspondence list file, wherein,
The correspondence list file is the identifier server of the server unit of the content-data used in data storage is processed
The file being mapped with the process order of specified data processing,
The process distribution program is used to make computer perform following process：
Process order to receive treatment, accept for specifying the first process of the first data processing for using first content data to order
Order；
Corresponding server selection is processed, and selects and receive with being received treatment by the process order from the correspondence list file
First process order identical of reason processes the corresponding identifier server of order；And
Process distribution to process, request processes the identifier server selected mark using by the corresponding server selection
Server unit performs first data processing.
7. a kind of data handling system, it is characterised in that
The data handling system has process distributor and multiple server units,
The process distributor has：
Correspondence list file storage part, which stores the server of the server unit of the content-data used in data storage is processed
The corresponding list file that identifier is mapped with the process order of specified data processing；
Order receiving unit is processed, which is accepted for specifying the first process of the first data processing for using first content data to order
Order；
Dispenser is processed, and which does not include in the correspondence list file and the first process that order receiving unit accepts is processed with by described
In the case of the corresponding identifier server of order identical process order, any clothes in the plurality of server unit are asked
Business device device performs first data processing, accepts first data processing from the server unit as request destination
Implementing result and the storage first content data server unit identifier server；And
Corresponding server register, which will process identifier server and the described first process order that dispenser accepts by described
It is mapped and is registered in the correspondence list file,
The plurality of server unit has respectively：
Content-data storage portion, its storage content data；
Data configuration file storage part, its storage are interior with identification by the identifier server of the server unit of storage content data
Hold the data configuration file that the data identifier of data is mapped；
Processing execution unit, which is performed from first data processing for processing distributor request；
Access request portion, in the case which does not store the first content data in the content-data storage portion, from described
Identifier server corresponding with the data identifier of the first content data is selected in data configuration file, request is using choosing
The server unit of the identifier server mark selected out accesses the first content data；And
Implementing result response portion, which processes the first data processing that distribution means responsive is performed by the processing execution unit to described
Implementing result and the identifier server selected by the access request portion.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/JP2014/072716 WO2016031043A1 (en) | 2014-08-29 | 2014-08-29 | Process distributing device, process distributing program, and data processing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106575267A true CN106575267A (en) | 2017-04-19 |
| CN106575267B CN106575267B (en) | 2018-04-06 |
Family
ID=55398978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201480081125.XA Expired - Fee Related CN106575267B (en) | 2014-08-29 | 2014-08-29 | Handle distributor and data handling system |
Country Status (4)
| Country | Link |
|---|---|
| JP (1) | JP6203414B2 (en) |
| CN (1) | CN106575267B (en) |
| TW (1) | TWI531909B (en) |
| WO (1) | WO2016031043A1 (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1231855C (en) * | 2000-07-07 | 2005-12-14 | 株式会社日立制作所 | Device and method for dynamic distributing computer resource according to user's agreement |
| CN1855825A (en) * | 2005-04-27 | 2006-11-01 | 株式会社日立制作所 | Computer system |
| US7912954B1 (en) * | 2003-06-27 | 2011-03-22 | Oesterreicher Richard T | System and method for digital media server load balancing |
| JP2011186616A (en) * | 2010-03-05 | 2011-09-22 | Nec Corp | Distributed file system, and server, method, and program for remote file processing |
| CN102238143A (en) * | 2010-04-30 | 2011-11-09 | 鸿富锦精密工业（深圳）有限公司 | Master server and method for decoding multimedia data by using same |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SE0200417D0 (en) * | 2002-02-13 | 2002-02-13 | Ericsson Telefon Ab L M | A method and apparatus for reconfiguring a server system |
| JP2008071229A (en) * | 2006-09-15 | 2008-03-27 | Onkyo Corp | Network av system and controller |
| JP4677992B2 (en) * | 2007-01-19 | 2011-04-27 | 日本電気株式会社 | VIDEO DISTRIBUTION SYSTEM, VIDEO DISTRIBUTION METHOD, DISTRIBUTION MANAGEMENT SYSTEM, AND PROGRAM |
-
2014
- 2014-08-29 WO PCT/JP2014/072716 patent/WO2016031043A1/en active Application Filing
- 2014-08-29 CN CN201480081125.XA patent/CN106575267B/en not_active Expired - Fee Related
- 2014-08-29 JP JP2016545186A patent/JP6203414B2/en not_active Expired - Fee Related
- 2014-10-20 TW TW103136102A patent/TWI531909B/en not_active IP Right Cessation
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1231855C (en) * | 2000-07-07 | 2005-12-14 | 株式会社日立制作所 | Device and method for dynamic distributing computer resource according to user's agreement |
| US7912954B1 (en) * | 2003-06-27 | 2011-03-22 | Oesterreicher Richard T | System and method for digital media server load balancing |
| CN1855825A (en) * | 2005-04-27 | 2006-11-01 | 株式会社日立制作所 | Computer system |
| JP2011186616A (en) * | 2010-03-05 | 2011-09-22 | Nec Corp | Distributed file system, and server, method, and program for remote file processing |
| CN102238143A (en) * | 2010-04-30 | 2011-11-09 | 鸿富锦精密工业（深圳）有限公司 | Master server and method for decoding multimedia data by using same |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2016031043A1 (en) | 2016-03-03 |
| TWI531909B (en) | 2016-05-01 |
| JPWO2016031043A1 (en) | 2017-04-27 |
| TW201608389A (en) | 2016-03-01 |
| JP6203414B2 (en) | 2017-09-27 |
| CN106575267B (en) | 2018-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113906723B (en) | Multi-cluster portal | |
| US20200364608A1 (en) | Communicating in a federated learning environment | |
| US6078954A (en) | Server directed multicast communication method and system | |
| CN105122772B (en) | A kind of method and apparatus by head swap server state and client-side information | |
| US8180730B2 (en) | Arbitration token for managing data integrity and data accuracy of information services that utilize distributed data replicas | |
| CN109240830A (en) | Application intelligence request management based on server health and client-side information | |
| CN108182111A (en) | Task scheduling system, method and apparatus | |
| JP5724687B2 (en) | Information processing apparatus, server selection method, and program | |
| CN108933829A (en) | A kind of load-balancing method and device | |
| CN112579622B (en) | Method, device and equipment for processing service data | |
| CN108337275A (en) | Task distribution method, device and equipment for Distributor | |
| CN109726005A (en) | For managing method, server system and the computer program product of resource | |
| JP2016024612A (en) | Data processing control method, data processing control program, and data processing control apparatus | |
| CN114924751A (en) | Method and device for distributing service access request flow | |
| CN103248636B (en) | The system and method downloaded offline | |
| CN110336874A (en) | A kind of method, apparatus, server and the storage medium of access service | |
| CN106469091B (en) | For distributing the method and apparatus of available area | |
| CN111614792B (en) | Transparent transmission method, system, server, electronic device and storage medium | |
| CN110365783B (en) | File deployment method and device, network node and storage medium | |
| CN111126604A (en) | Model training method, device, server and storage medium | |
| WO2017167129A1 (en) | Method and device for distributing data center data to user | |
| JP6888478B2 (en) | Sorting system | |
| CN107969166A (en) | For asking the method and system of the limited Service Instance of access | |
| CN106575267B (en) | Handle distributor and data handling system | |
| CN109417566A (en) | Controlled seed dispersal based on constraint |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180406 Termination date: 20190829 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |