CN105516242A - Storage resource distribution method and storage resource distribution system - Google Patents

Storage resource distribution method and storage resource distribution system Download PDF

Info

Publication number
CN105516242A
CN105516242A CN201510821385.2A CN201510821385A CN105516242A CN 105516242 A CN105516242 A CN 105516242A CN 201510821385 A CN201510821385 A CN 201510821385A CN 105516242 A CN105516242 A CN 105516242A
Authority
CN
China
Prior art keywords
storage resources
target service
service
described target
storage
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
Application number
CN201510821385.2A
Other languages
Chinese (zh)
Inventor
陈安伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201510821385.2A priority Critical patent/CN105516242A/en
Publication of CN105516242A publication Critical patent/CN105516242A/en
Priority to PCT/CN2016/106651 priority patent/WO2017088717A1/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/40—Support for services or applications
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/104—Peer-to-peer [P2P] networks
    • H04L67/1074—Peer-to-peer [P2P] networks for supporting data block transmission mechanisms

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

One embodiment of the invention discloses a storage resource distribution method and a storage resource distribution system. The method comprises a storage resource distribution request is received (carrying service type and service characteristic data); a resource distribution algorithm in accordance with the target service is selected from preset resource distribution algorithms on the basis of the service type of the target service; the service characteristic data of the target service is input and calculated with the selected resource distribution algorithm so that storage resource configuration data is obtained; then the storage resource in accordance with the storage resource configuration data is distributed by an instruction storage system. By use of the method and system, storage resource is automatically distributed to each service in a service system so that resource distribution efficiency is increased.

Description

A kind of storage resources distribution method and storage resources distribution system
Technical field
The present invention relates to the technical field of memory of data center, particularly relate to a kind of storage resources distribution method and storage resources distribution system.
Background technology
Storage system is the core infrastructure of data center (DataCenterNetwork, DCN) environment, is the final supporting body of data center's data access.Along with the development of cloud computing, storage system can not simply again build according to general service operation demand, but will realize the elasticity of scheduling of resource, good retractility, operation expanding flexibly and quick deliverability for miscellaneous service system.
In prior art, in order to give the storage resources that different traffic assignments is different, ensure the normal operation of each operation system, keeper needs according to each business in the past to the service condition of storage resources, evaluate the storage resources of each service needed now, then manually configuration mode gives each traffic assignments storage resources.
But realized the distribution of storage resources by this manual configuration, keeper needs to take much time and analyzes each business storage resources service condition in the past, and layoutprocedure is complicated, inefficiency.
Summary of the invention
Embodiments provide a kind of storage resources distribution method and storage resources distribution system, can be each traffic assignments storage resources in operation system automatically according to service feature, improve the efficiency of Resourse Distribute.
First aspect, provides a kind of storage resources distribution method, comprising:
Storage resources distribution system receives the resource allocation request (comprising the type of service of described target service and the service feature data of described target service) for target service, and according to the type of service of described target service, from preset algorithm set, select the storage resources allocation algorithm for described target service; Afterwards, with the service feature data of described target service for input, the storage resources allocation algorithm that storage resources distribution system utilizes described target service corresponding, calculates the storage resources configuration data that described target service is corresponding; Finally, storage resources distribution system sends resource allocation request to storage system, described resource allocation request comprises described storage resources configuration data, and described resource allocation request is used to indicate described storage system and the storage resources meeting described storage resources configuration data is distributed to described target service.
Concrete, described preset algorithm set comprises multiple storage resources allocation algorithm, wherein, and the corresponding a kind of type of service of each storage resources allocation algorithm, the storage resources configuration data that the business for calculating corresponding service type is corresponding; Described storage resources configuration data can comprise: for the first numerical value of configuration store capacity and the second value for configuration store performance.
Understandable, the storage resources distribution method that first aspect describes can realize being described target service memory allocated resource automatically, improves the efficiency of Resourse Distribute.
Concrete, after send described storage resources configuration data to storage system, storage resources distribution system can also receive the storage resources configuration result that storage system returns, and send described storage resources configuration result to operation system, in order to will distribute to the storage resources informing business system of described target service, the storage resources making the described target service operated in operation system can use described storage resources configuration result to represent.
In conjunction with first aspect, in the implementation that the first is possible, after sending resource allocation request to storage system, described storage resources distribution system is all right: monitor the service condition of described target service to the described storage resources distributed; Storage resources configuration data corresponding to described target service is adjusted according to described service condition; Resource allocation request is resend to described storage system, the described resource allocation request resend comprises the described storage resources configuration data after adjustment, described in the resource allocation request that resends be used to indicate described storage system the storage resources meeting the described storage resources configuration data after adjustment distributed to described target service.
Understandable, by after giving described target service memory allocated resource, monitor the actual service condition of described target service to memory allocated resource, and adjust storage resources corresponding to described target service according to actual service condition, can ensure that the final storage resources distributed can meet the demand of described target service, can the wasting of resources be avoided again, realize making rational use of resources.
In conjunction with the first possible implementation of first aspect, in the implementation that the second is possible, describedly adjust storage resources configuration data corresponding to described target service according to described service condition, comprising: adjust storage resources allocation algorithm corresponding to described target service according to described service condition; With the service feature data of described target service for input, the storage resources allocation algorithm after utilization adjustment calculates storage resources configuration data corresponding to the described target service after adjustment.
Understandable, by going to adjust storage resources allocation algorithm corresponding to described target service to the actual service condition of memory allocated resource according to described target service, can realize training described storage resources allocation algorithm, make the result of calculation of described storage resources allocation algorithm closer to the storage resources of described target service actual demand, further ensure reasonable distribution and the utilization of storage resources.
In conjunction with the implementation that the second of first aspect is possible, describedly adjust storage resources allocation algorithm corresponding to described target service according to described service condition and can comprise following several possible implementation:
If the memory capacity of the described storage resources distributed described in described service condition shows does not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, reduce the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
If the memory property of the described storage resources distributed described in described service condition shows does not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory property and the positive correlation degree of memory property, or, reduce the negative correlation variable of memory property and the negative correlation degree of memory property.
If described service condition shows that the ratio of the memory capacity of the storage resources of the actual use of described target service and the memory capacity of the described described storage resources distributed is lower than first threshold, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, improve the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
If described service condition shows that the ratio of the memory property of the storage resources of the actual use of described target service and the memory property of the described described storage resources distributed is lower than Second Threshold, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory property and the positive correlation degree of memory property, or, improve the negative correlation variable of memory property and the negative correlation degree of memory property.
In conjunction with the third possible implementation of first aspect, improve the positive correlation variable of result of calculation (memory capacity or memory property) and the positive correlation degree of result of calculation and at least can comprise following two kinds of possible implementations: the positive correlation degree being improved described positive correlation variable and result of calculation by the coefficient increasing described positive correlation variable; The positive correlation degree of described positive correlation variable and result of calculation is improved by the index increasing described positive correlation variable.
Reduce the positive correlation variable of result of calculation (memory capacity or memory property) and the positive correlation degree of result of calculation and at least can comprise following two kinds of possible implementations: the positive correlation degree being reduced described positive correlation variable and result of calculation by the coefficient reducing described positive correlation variable; The positive correlation degree of described positive correlation variable and result of calculation is reduced by the index reducing described positive correlation variable.
Similar, the negative correlation degree of the negative correlation variable and result of calculation that improve result of calculation (memory capacity or memory property) at least can comprise following two kinds of possible implementations: the negative correlation degree being improved described negative correlation variable and result of calculation by the coefficient increasing described negative correlation variable; The negative correlation degree improving described negative correlation variable and result of calculation is changed by the index increasing described negative correlation variable.
Reduce the negative correlation variable of result of calculation (memory capacity or memory property) and the negative correlation degree of result of calculation and at least can comprise following two kinds of possible implementations: the negative correlation degree being reduced described negative correlation variable and result of calculation by the coefficient reducing described negative correlation variable; The negative correlation degree reducing described negative correlation variable and result of calculation is changed by the index reducing described negative correlation variable.
Second aspect, provides a kind of storage resources distribution system, and described storage resources distribution system comprises the module for performing method described in first aspect.
The third aspect, provides a kind of server, for performing the storage resources distribution method that first aspect describes.Described server comprises: input unit, output device, R-T unit, memory and the processor be coupled with described memory, wherein, described input unit is used for the input data of receiving management person, described output device is used for externally exporting data, described R-T unit is used for carrying out data communication with described operation system or described storage system, described memory is for storing the code of the storage resources distribution method of first aspect description, described processor is for performing the program code in described memory, namely the storage resources distribution method that first aspect describes is performed.
Fourth aspect, provide a kind of computer-readable recording medium, described readable storage medium storing program for executing stores the program code realizing the storage resources distribution method that first aspect describes, this program code comprises the execution instruction running the storage resources distribution method that first aspect describes.
5th aspect, provide a preset algorithm set, described preset algorithm set can comprise multiple storage resources allocation algorithm, and described multiple storage resources allocation algorithm is that empirical data according to each type of service (comprising service feature data, the memory capacity of storage resources be assigned with and memory property) analysis modeling draws.Wherein, the corresponding a kind of type of service of each storage resources allocation algorithm, the storage resources configuration data that the business for calculating corresponding service type is corresponding; Described storage resources allocation algorithm can comprise: the first algorithm and the second algorithm, wherein: the first algorithm is used for calculating memory capacity, namely the first numerical value in described storage resources configuration data, second algorithm is used for calculating memory property, the second value namely in described storage resources configuration data.
In possible execution modes more of the present invention, it can be the request that keeper is sent to storage resources distribution system by the working software that storage resources distribution system provides that described storage resources distributes request, also can be the request that operation system sends directly to storage resources distribution system.
In possible execution modes more of the present invention, described service feature data can comprise: business scale, and the data (peak value trading volume every day as ebanking services), business datum resting period etc. that characterize traffic fluctuations represent the data of service feature.
In possible execution modes more of the present invention, described memory capacity can be represented by the number of the capacity of single storage volume and storage volume; Described memory property can comprise: input-output operation number of times (IOPS) per second, bandwidth, I/O time delay etc.
Implement the embodiment of the present invention, described storage resources distribution system is after reception storage resources distributes request (carrying type of service and service feature data), from default resource allocation algorithm, the resource allocation algorithm for described target service is selected according to the type of service of described target service, and with the service feature data of described target service for input, the described resource allocation algorithm selected is utilized to calculate the storage resources configuration data of described target service, the storage resources meeting described storage resources configuration data is distributed to described target service by then instruction storage system.It is each traffic assignments storage resources in operation system automatically that the present invention program can realize according to service feature, improves the efficiency of Resourse Distribute.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described.
Fig. 1 is the network architecture schematic diagram of the data center that the embodiment of the present invention relates to;
Fig. 2 is the schematic flow sheet of a kind of storage resources distribution method that the embodiment of the present invention provides;
Fig. 3 is the schematic diagram of the service feature data of several types of service that the embodiment of the present invention relates to;
Fig. 4 is the schematic flow sheet of the another kind of storage resources distribution method that the embodiment of the present invention provides;
Fig. 5 is the module diagram of a kind of storage resources distribution system that the embodiment of the present invention provides;
Fig. 6 is the structural representation of a kind of server that the embodiment of the present invention provides.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly described.Obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.
For the ease of understanding the embodiment of the present invention, below first composition graphs 1 introduce application scenarios that the embodiment of the present invention relates to and the technical problem that described application scenarios exists in detail.
Fig. 1 shows the basic network framework of the data center that the embodiment of the present invention relates to, and wherein, storage system supports for operation system provides to store.
The storage architecture of storage system can be that cloud stores, in order to realize Distributed Storage and management.More excellent, the storage architecture of storage system can also be that software definition stores (SoftwareDefinedStorage, SDS), the i.e. separation of software and hardware in storage architecture, namely being separated of data Layer and key-course, management to storage resources and scheduling is realized, as rolled up migration etc. flexibly, and without the need to considering hardware device itself by software.
In the embodiment of the present invention, the network configuration of storage system can comprise: direct-connected storage (DirectAttachedStorage, DAS), network connects storage (NetworkAttachedStorage, and storage area network (StorageAreaNetwork, SAN) NAS).At present, for the operational environment of this large buffer memory of data center, the application of SAN is usually more extensive.As shown in Figure 1, the IPSAN of SAN can be transmission medium be IP network also can be transmission medium is FC (FiberChannel) SAN of optical fiber.
In the embodiment of the present invention, storage array in storage system can comprise the storage medium of multiple storage class, as solid state hard disc (SolidStateDrive, SSD), Serial Attached SCSI (SAS) (SerialAttachedSCSI, SAS) hard disk, mechanical hard disk etc.
Operation system generally includes multiple business platform, meets all kinds of business demand of user.Understandable, dissimilar business is normally different to the demand of storage resources, is mainly reflected in memory capacity and memory property.Such as, financial business requires higher to the response speed that data store, and namely requires high to the input-output operation number of times (Input/OutputOperationsPerSecond, IOPS) per second of storage resources.Again such as, because the data volume of business is large, the memory capacity of video monitoring service demand is larger.
In prior art, when memory allocated resource, keeper needs to take into full account the demand of dissimilar business to memory capacity and memory property.In the embodiment of the present invention, except aforesaid IOPS, the memory property of storage resources also comprises: readwrite bandwidth, fault freedom (i.e. failover capability) etc.
During concrete practice, keeper needs to collect each business in the past to the actual service condition of storage resources, and analyzes the demand of each business to storage resources according to this.Obviously, this working method needs the at substantial time, and analytic process is complicated.
For problems of the prior art, as shown in Figure 1, the embodiment of the present invention adds storage resources distribution system in existing data center network architecture, and provides a kind of storage resources distribution method.In the present invention program, described storage resources distribution system is after reception storage resources distributes request (carrying type of service and service feature data), from default resource allocation algorithm, the resource allocation algorithm for described target service is selected according to the type of service of target service, and with the service feature data of described target service for input, the described resource allocation algorithm selected is utilized to calculate the storage resources configuration data of described target service, the storage resources meeting described storage resources configuration data is distributed to described target service by then instruction storage system.It is each traffic assignments storage resources in operation system automatically that the present invention program can realize according to service feature, improves the efficiency of Resourse Distribute.Below be described in detail respectively.
The schematic flow sheet of the storage resources distribution method that the embodiment of the present invention provides is shown see Fig. 2, Fig. 2.Described method comprises:
S101, the storage resources that the working software that keeper is provided by storage resources distribution system sends for target service to storage resources distribution system distributes request.Accordingly, the storage resources that storage resources distribution system receives for described target service distributes request.Described request comprises: the type of service of described target service and the service feature data of described target service.
In a kind of possible implementation of the embodiment of the present invention, it also can be that operation system sends that described storage resources distributes request.
S103, after receiving described storage resources and distributing request, storage resources distribution system according to the type of service of described target service, can select the storage resources allocation algorithm for described target service from preset algorithm set.Concrete, described preset algorithm set comprises multiple storage resources allocation algorithm, wherein, and the corresponding a kind of type of service of each storage resources allocation algorithm, the storage resources configuration data that the business for calculating corresponding service type is corresponding.Here, described storage resources configuration data can comprise: for the first numerical value of configuration store capacity with for configuration store performance second value.
S105, storage resources distribution system with the service feature data of described target service for input, can utilize the storage resources allocation algorithm selected in S103, calculates the storage resources configuration data that described target service is corresponding.Here, described service feature data can comprise: business scale, and the data (peak value trading volume every day as ebanking services), business datum resting period etc. that characterize traffic fluctuations represent data of service feature etc.
S107, storage resources distribution system can send resource allocation request to described storage system, and described resource allocation request comprises described storage resources configuration data.
S109, accordingly, storage system receives the described resource allocation request that storage resources distribution system sends, and responds described resource allocation request, and the storage resources meeting described storage resources configuration data is distributed to described target service.
Concrete, before giving described target service memory allocated resource, storage system can obtain the state information of whole storage array in storage system, and chooses the storage resources meeting described storage resources configuration data according to described state information.Such as, the state information of the storage array in storage system is as shown in table 1:
Storage volume Bandwidth Media type RAID rank Capacity can be write Network type
Volume 1 0.1GB/S SSD RAID0 8.5GB IP SAN
Volume 2 0.5GB/S SAS RAID3 15GB FC SAN
… … … … … …
Table 1
Such as, the memory capacity that target service described in the first numeric representation in described storage resources configuration data needs is: 8GB, and the second value in described storage resources configuration data represents the memory bandwidth of described target service needs and is: 0.3GB/S, so, according to the storaging state information of the storage system shown in table 1, " volume 2 " can be distributed to described target service by storage system.Example is only a kind of implementation of the embodiment of the present invention, can be different in practical application, and such as, multiple storage volume can also be distributed to a business by storage system, should not form restriction here.
Needs illustrate, table 1 is only a kind of implementation of the embodiment of the present invention, can be different in practical application, should not form restriction.
S111, after giving described target service memory allocated resource, storage resources configuration result also can be returned to storage resources distribution system by storage system.Concrete, described storage resources configuration result mainly can comprise: the capacity information of the storage resources distributed and positional information.Further, described storage resources configuration result also can comprise: the storage class, readwrite performance (as IOPS, I/O delay, bandwidth), interface type etc. of the storage resources distributed.
S113, accordingly, storage resources distribution system receives the storage resources configuration result that storage system returns, and send described storage resources configuration result to operation system, in order to will distribute to the storage resources informing business system of described target service, the storage resources making the described target service operated in operation system can use described storage resources configuration result to represent.
Such as, operation system determines reading and writing data address according to the positional information (numbering as storage volume) of the storage resources distributed in described configuration result.Again such as, operation system determines suitable storage protocol according to the storage class, interface type etc. of the storage resources distributed.Example is only a kind of implementation of the embodiment of the present invention, can also be different in practical application, should not form restriction.
S115, preferably, after receiving described storage resources configuration result, operation system can also generate the mapping item of the storage resources distributed according to described storage resources configuration result, be convenient to described target service use intuitively described in the storage resources that distributed.Be to be understood that, described mapping item is quoting of the described storage resources distributed.In specific implementation, described mapping item can show as the virtual memory dish (as virtual drive) in business system server.
S117, optionally, after the mapping item of the storage resources distributed described in generating, execution result can be returned to storage resources distribution system by operation system.Optionally, if operation system does not perform S115, then the reception that operation system can only return for described storage resources configuration result is replied to storage resources distribution system.
S119, finally, optionally, described storage resources configuration result can be returned to keeper by storage resources distribution system.
In one implementation, the working software (for submitting the software of the resource allocation request of each business to) of keeper can run directly in storage resources distribution system.Described storage resources configuration result can output in described working software by storage resources distribution system.
In another kind of implementation, the working software of keeper may operate on a communication equipment interconnected with storage resources distribution system.Described storage resources configuration result can be sent to described communication equipment by storage resources distribution system.Here, interconnected refer to exist between described communication equipment and storage resources distribution system communicate to connect.
The following detailed description of the storage resources allocation algorithm in the embodiment of the present invention.
In the embodiment of the present invention, storage resources allocation algorithm is for calculating the storage resources configuration data of a service needed.Storage resources configuration data comprises: for the first numerical value of configuration store capacity and the second value for configuration store performance.
In the embodiment of the present invention, described memory capacity can be represented by the number of the capacity of single storage volume and storage volume; Described memory property can comprise: input-output operation number of times per second, bandwidth, I/O time delay etc.So, above-mentioned first numerical value can be expressed as: { LunSize, LunNum}, LunSize represent the capacity of single memory disc, and LunNum represents memory disc quantity; Above-mentioned second value can be expressed as: rIOPS, wIOPS, BandWidth, ioDelay ..., rIOPS represents read operation number of times per second, and wIOPS represents write operation number of times per second, and BandWidth represents bandwidth, and ioDelay represents that I/O postpones.
Understandable, the storage resources of a service needed is relevant to the service feature of this business.Therefore, the memory capacity of a service needed and memory property all can be represented by following Mathematical Modeling:
Memory capacity=f (service feature data)
Memory property=g (service feature data)
Wherein, f, g represent that variable is the function of described service feature data respectively.
Described service feature data represent by a set C, that is: C={c1, c2, c3 ....For the business of different service types, described service feature data may there are differences, and the element namely gathered in C is different.As shown in Figure 3, the service feature data of ebanking services not only comprise business scale, also comprise: the service features such as single business number, every user 3 annual data amount, peak value any active ues percentage; The service feature data of video traffic not only comprise business scale, also comprise: the service features such as holding time, code check coefficient of variation.Understandable, the memory capacity of the needs of ebanking services is all relevant to the service feature data of aforementioned ebanking services with memory property, and the memory capacity that the service feature data of video traffic need is all relevant to the service feature data of aforementioned video business with memory property.
During concrete enforcement, can according to the empirical data of the business of each type of service, that is: history service characteristic and memory capacity, the memory property of historical storage resource that have been assigned with, set up out the storage resources allocation algorithm that the business of each type of service is corresponding.
Such as, through the analysis of the empirical data of ebanking services, the storage resources allocation algorithm setting up out ebanking services is as follows:
LunSize＝businessScale*10000*numPerBusiness/10^6)*2*120％；
LunNum＝((businessScale*10000*numPerBusiness/10^6)*2*120％)/240；
rIOPS＝(businessScale*10000*numPerBusiness*pctPeak*elpsPerTran/peakHour/60)*0.5*0.1；
wIOPS＝(businessScale*10000*numPerBusiness*pctPeak*elpsPerTran/peakHour/60)*0.5*2；
Wherein, businessScale represents registered user's number, and numPerBusiness represents everyone trading volume every day, and pctPeak represents peak value any active ues percentage, and elpsPerTran represents online hours of often concluding the business, and peakHour represents peak value duration every day.
Needs illustrate, above-mentioned example is only a kind of implementation of the embodiment of the present invention, can also be different in practical application, should not form restriction.Except above-mentioned rIOPS and wIOPS, other memory properties (as BandWidth and ioDelay) that ebanking services needs also can by drawing the empirical data modeling of ebanking services.
From foregoing, storage resources allocation algorithm in the embodiment of the present invention can comprise: the first algorithm and the second algorithm, wherein: the first algorithm is used for calculating memory capacity, namely the first numerical value in described storage resources configuration data, second algorithm is used for calculating memory property, the second value namely in described storage resources configuration data.Concrete, for the ebanking services of aforementioned exemplary, described first algorithm can be expressed as algorithm set A1={a11, an a12}, and wherein, a11, a12 can represent the algorithm for calculating LunSize, LunNum respectively; Same, described second algorithm can be expressed as algorithm set A2={a21, a22, a23, an a24}, and wherein, a21, a22, a23 and a24 can represent the algorithm for calculating rIOPS, wIOPS, BandWidth and ioDelay respectively.
Needs illustrate, for the business of different service types, the algorithm set A1 characterizing described first algorithm may be different, and the algorithm set A2 characterizing described second algorithm also may difference.
Because the storage resources allocation algorithm in described preset algorithm set draws according to the empirical data analysis modeling of each type of service, therefore, discrepancy may be there is between the storage resources that the storage resources configuration data utilizing storage resources allocation algorithm corresponding to the described target service in described preset algorithm set to calculate characterizes and the storage resources of the current actual needs of described target service.
In order to ensure reasonable distribution and the Appropriate application of storage resources, in a further embodiment, the storage resources allocation algorithm for described target service is being selected from preset algorithm set, the storage resources configuration data of described target service is calculated according to the algorithm selected, and after the storage resources meeting described storage resources configuration data distributed to described target service by instruction storage system, storage resources distribution system is all right: monitor the actual service condition of described target service to the described storage resources distributed, and adjust storage resources configuration data corresponding to described target service according to described actual service condition, the described storage resources configuration data after adjustment is sent to described storage system, make described storage system that the storage resources meeting the described storage resources configuration data after adjustment is distributed to described target service.
Concrete, as shown in Figure 4, after the storage resources meeting described storage resources configuration data is distributed to described target service, the storage resources distribution method that the embodiment of the present invention provides also can further comprise:
S121, storage resources distribution system gathers the actual service condition of described target service to storage resources from storage system.
S123, accordingly, the actual service condition of described target service to storage resources is returned to storage resources distribution system by storage system.
In specific implementation, the actual service condition (taken capacity, reading and writing data log information etc.) of described target service to storage resources can be recorded in journal file by storage system.Further, in order to verify the service condition of the storage resources collected from storage system, storage resources distribution system also can gather the actual service condition of described target service to storage resources from operation system, and whether compare the actual service condition collected from both consistent, if consistent, then determine that the described actual service condition collected is correct, otherwise, S121 can be re-executed, until collection result is correct.
S125, receive described target service to the actual service condition of storage resources after, storage resources distribution system can adjust storage resources allocation algorithm corresponding to described target service according to described actual service condition, and utilize the storage resources allocation algorithm after adjustment to calculate storage resources configuration data corresponding to the described target service after adjustment, can realize adjusting storage resources configuration data corresponding to described target service according to described actual service condition, and can realize training described storage resources allocation algorithm, make the result of calculation of described storage resources allocation algorithm closer to the storage resources of described target service actual demand, the reasonable distribution of further raising storage resources and utilization.
S127, storage resources distribution system can send resource allocation request to storage system again, the described resource allocation request resend comprises storage resources configuration data corresponding to the described target service after adjustment, for reconfiguring storage resources corresponding to described target service.
S129, accordingly, the resource allocation request resend described in storage system receives, and the resource allocation request described in response, adjust storage resources corresponding to described target service according to the described storage resources configuration data after adjustment.
S131, after the storage resources that the described target service of adjustment is corresponding, adjustment result also can be returned to storage resources distribution system by storage system.Concrete, described adjustment result mainly can comprise: the capacity information of the storage resources after adjustment and positional information.Further, described adjustment result also can comprise: the storage class, readwrite performance (as IOPS, I/O delay, bandwidth), interface type etc. of the storage resources after adjustment.
S133, after receiving the described adjustment result that storage system returns, storage resources distribution system can send described adjustment result to operation system, in order to by the storage resources informing business system of described target service after adjustment, making to operate in described target service in operation system can storage resources after Use Adjustment.
S135, optionally, storage resources distribution system also can send described adjustment result to operation system.
Be to be understood that, if described adjustment result shows that storage system carries out storage resources recovery or dilatation to described target service, operation system only need change the mapping relations of the storage resources in described target service and storage system, without the need to change operation system generates described in the mapping item (as virtual drive) of storage resources that distributed.
In the embodiment of the present invention, storage resources distribution system roughly adjusts storage resources allocation algorithm corresponding to described target service by following several implementation:
If described actual service condition shows that the memory capacity of the storage resources distributing to described target service does not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, reduce the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
If the memory property that described actual service condition shows to distribute to described target service does not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory property and the positive correlation degree of memory property, or, reduce the negative correlation variable of memory property and the negative correlation degree of memory property.
If described actual service condition shows the memory capacity of the storage resources of the actual use of described target service and distributes to the ratio of memory capacity of storage resources of described target service lower than first threshold, that is: described target service does not make full use of the memory capacity of having distributed, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, improve the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
If described actual service condition shows that the ratio of the memory property of the storage resources of the actual use of described target service and the memory property of the described described storage resources distributed is lower than Second Threshold, that is: described target service does not make full use of the memory property of having distributed, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory property and the positive correlation degree of memory property, or, improve the negative correlation variable of memory property and the negative correlation degree of memory property.
Be to be understood that, positive correlation refers to that the change of equidirectional (Two Variables becomes large simultaneously or diminishes) occurs along with the change of another variable a variable.Such as, y=a*x (constant a is greater than 0), then variable y and variable x is positive correlation, and wherein, a is larger, and the positive correlation degree of y and x is higher.Again such as, g=x b(constant b is greater than 0), then variable g and variable x (x>0) positive correlation, wherein, b is larger, and the positive correlation degree of g and x is higher.
Be to be understood that, negative correlation refers to that reciprocal change occurs along with the change of another variable a variable.Such as, y=a*x (constant a is less than 0), then variable y and variable x is negative correlativing relation, and wherein, a is less, and the negative correlation degree of y and x is higher.Again such as, g=x b(constant b is less than 0), then variable g and variable x (x>0) negative correlation, wherein, b is less, and the negative correlation degree of g and x is higher.
Below for the storage resources allocation algorithm that aforementioned ebanking services is corresponding, be described:
From memory capacity computational algorithm: LunSize=businessScale*10000*numPerBusiness/10^6) * 2*120%, businessScale, numPerBusiness are the positive correlation variablees of LunSize.
In specific implementation, if the LunSize that memory capacity algorithm calculates can not meet the demand of described ebanking services to memory capacity, so, by increasing the constant factor such as " 10000 ", " 2 ", " 1/10^6 " and " 120% ", the positive correlation degree of LunSize and businessScale, numPerBusiness can be improved.
In specific implementation, if described ebanking services can not make full use of the memory capacity that LunSize characterizes, so, by reducing the constant factor such as " 10000 ", " 2 ", " 1/10^6 " and " 120% ", the positive correlation degree of LunSize and businessScale, numPerBusiness can be reduced.
In practical application, except the coefficient by changing positive correlation variable changes except the positive correlation degree of positive correlation variable and result of calculation, the positive correlation degree of positive correlation variable and result of calculation also can be changed by the index changing positive correlation variable.
Such as, by aforementioned memory capacity algorithm is changed into: LunSize=businessScale 2* 10000*numPerBusiness 2/ 10^6) * 2*120% improves the positive correlation degree of LunSize and businessScale, numPerBusiness.Again such as, by aforementioned memory capacity algorithm is changed into: LunSize=businessScale 0.5* 10000*numPerBusiness 0.5/ 10^6) * 2*120% reduces the positive correlation degree of LunSize and businessScale, numPerBusiness.Example is only a kind of implementation of the embodiment of the present invention, can also be different in practical application, should not form restriction.In practical application, other modes can also be taked to change the positive correlation degree of positive correlation variable and result of calculation, and the embodiment of the present invention is not restricted.
Similarly, computational algorithm from read operation number of times (one of memory property) per second: rIOPS=(businessScale*10000*numPerBusiness*pctPeak*elpsPerTran/p eakHour/60) * 0.5*0.1, businessScale, numPerBusiness, pctPeak and elpsPerTran are the positive correlation variable of rIOPS, and peakHour is the negative correlation variable of rIOPS.
In specific implementation, if the rIOPS that aforementioned algorism calculates can not meet described ebanking services to storing the demand reading performance, so, the positive correlation degree of positive correlation variable and rIOPS can be improved by the constant factor (or index) increasing positive correlation variable, also can be reduced the negative correlation degree of negative correlation variable and rIOPS by the constant factor (or index) reducing negative correlation variable.
In specific implementation, if the rIOPS that the read operation number of times per second of described ebanking services reality calculates much smaller than aforementioned algorism, so, the positive correlation degree of positive correlation variable and rIOPS can be reduced by the constant factor (or index) reducing positive correlation variable, also can be improved the negative correlation degree of negative correlation variable and rIOPS by the constant factor (or index) increasing negative correlation variable.
Similarly, other storage resources allocation algorithms (as computational algorithms such as LunNum, wIOPS, BandWidth, ioDelay) also can adjust with reference to above-mentioned implementation, repeat no more here.
In a kind of possible implementation of the present invention, collect described target service to the actual service condition of storage resources after, except passing through storage resources allocation algorithm corresponding to the described target service of adjustment, and recalculate described target service to outside storage resources configuration data (i.e. S125), storage resources distribution system directly can also adjust according to the actual service condition of described target service to storage resources (directly increase numerical value or reduce numerical value) storage resources configuration data that S105 calculates, and the storage resources configuration data after adjustment is sent to storage system, storage system is made to adjust storage resources corresponding to described target service.
Implement the embodiment of the present invention, described storage resources distribution system is after reception storage resources distributes request, from default resource allocation algorithm, the resource allocation algorithm for described target service is selected according to the type of service of described target service, and with the service feature data of described target service for input, the described resource allocation algorithm selected is utilized to calculate the storage resources configuration data of described target service, the storage resources meeting described storage resources configuration data is distributed to described target service by then instruction storage system, can realize being each traffic assignments storage resources in operation system automatically, improve the efficiency of Resourse Distribute.And, after the storage resources meeting described storage resources configuration data is distributed to described target service, can also go to adjust storage resources allocation algorithm corresponding to described target service to the actual service condition of memory allocated resource according to described target service, can realize training described storage resources allocation algorithm, make the result of calculation of described storage resources allocation algorithm closer to the storage resources of described target service actual demand, further ensure reasonable distribution and the utilization of storage resources.
In order to coordinate the storage resources allocation algorithm implemented the embodiment of the present invention and provide, embodiments provide a kind of storage resources distribution system, for realizing the concrete steps in Fig. 2 or Fig. 4 embodiment of the method corresponding respectively.As shown in Figure 5, described storage resources distribution system can comprise: receiver module 501, algorithms selection module 503, computing module 505 and resource distribution module 507.Wherein:
Receiver module 501, for receiving the resource allocation request for target service; Described request comprises the type of service of described target service and the service feature data of described target service;
Algorithms selection module 503, for the type of service according to described target service, selects the storage resources allocation algorithm for described target service from preset algorithm set 520; Preset algorithm set 520 comprises multiple storage resources allocation algorithm, wherein, and the corresponding a kind of type of service of each storage resources allocation algorithm, the storage resources configuration data that the business for calculating corresponding service type is corresponding; Described storage resources configuration data can comprise: for the first numerical value of configuration store capacity and the second value for configuration store performance;
Computing module 505, for the service feature data of described target service for input, utilize the storage resources allocation algorithm that described target service is corresponding, calculate the storage resources configuration data that described target service is corresponding;
Resource distribution module 507, for sending described storage resources configuration data to storage system, makes described storage system that the storage resources meeting described storage resources configuration data is distributed to described target service.
Because the storage resources allocation algorithm in preset algorithm set 520 draws according to the empirical data analysis modeling of each type of service, therefore, discrepancy may be there is between the storage resources that the storage resources configuration data that computing module 505 utilizes storage resources allocation algorithm corresponding to the described target service in preset algorithm set 520 to calculate characterizes and the storage resources of the current actual needs of described target service.
In order to ensure reasonable distribution and the Appropriate application of storage resources, further, described storage resources distribution system is comprising: outside receiver module 501, algorithms selection module 503, computing module 505 and resource distribution module 507, also can comprise: monitoring modular 509 and adjusting module 511, wherein:
Monitoring modular 509, be used in resource distribution module 507 and send described storage resources configuration data to storage system, make after the storage resources meeting described storage resources configuration data distributed to described target service data by described storage system, to monitor the service condition of described target service to the described storage resources distributed;
Adjusting module 511, the actual service condition that can be used for monitoring according to monitoring modular 509 adjusts storage resources configuration data corresponding to described target service, and triggers resource configuration module 507 sends the described storage resources configuration data after adjustment to described storage system, makes described storage system that the storage resources meeting the described storage resources configuration data after adjustment is distributed to described target service.
In specific implementation, the actual service condition that adjusting module 511 can monitor according to monitoring modular 509 adjusts storage resources allocation algorithm corresponding to described target service, and trigger computing module 505 with the service feature data of described target service for input, storage resources allocation algorithm after utilization adjustment calculates storage resources configuration data corresponding to the described target service after adjustment, can realize adjusting storage resources configuration data corresponding to described target service according to described actual service condition, and can realize training described storage resources allocation algorithm, make the result of calculation of described storage resources allocation algorithm closer to the storage resources of described target service actual demand, the reasonable distribution of further raising storage resources and utilization.
Concrete, several implementations that adjusting module 511 adjusts storage resources allocation algorithm corresponding to described target service with reference to the related content in figure 2 embodiment, can repeat no more here.
Needs illustrate, each software module that foregoing describes can be made up of the submodule of the sub-step of the refinement more performed in Fig. 2 or Fig. 4 embodiment of the method, is not restricted here.
For the ease of implementing the embodiment of the present invention, the invention provides a kind of server, for realizing the storage resources distribution method described in Fig. 2 or Fig. 4 embodiment.See Fig. 6, server 60 can comprise: input unit 603, output device 604, R-T unit 605, memory 602 and the processor 601 (quantity of the processor 601 in server 60 can be one or more, for a processor in Fig. 6) be coupled with memory 602.In some embodiments of the invention, input unit 603, output device 604, R-T unit 605, memory 602 are connected by bus or alternate manner with processor 601, wherein, to be connected by bus in Fig. 6.
Wherein, input unit 603, for receiving outside input data.In specific implementation, input unit 101 can comprise keyboard, mouse, photoelectricity input unit, acoustic input dephonoprojectoscope, touch-type input device, scanner etc.Output device 604, for externally exporting data.In specific implementation, output device 604 can comprise display, loud speaker, printer etc.R-T unit 605, for other equipment sending datas or from other equipment receiving datas.In specific implementation, R-T unit 605 can comprise the transceiving device such as radio receiving transmitting module, wired transceiver module.Memory 602 is for program code stored, and in specific implementation, memory 602 can adopt read-only memory (ReadOnlyMemory, ROM), can be used for storing the program code of the embodiment of the method realizing earlier figures 2 or Fig. 4 difference correspondence.Processor 601, such as CPU, be stored in memory 602 Program code execution following steps for calling:
By the resource allocation request for target service that input unit 603 receiving management person submits to; Described request comprises the type of service of described target service and the service feature data of described target service;
According to the type of service of described target service, from preset algorithm set, select the storage resources allocation algorithm for described target service; Described preset algorithm set comprises multiple storage resources allocation algorithm, wherein, and the corresponding a kind of type of service of each storage resources allocation algorithm, the storage resources configuration data that the business for calculating corresponding service type is corresponding; Described storage resources configuration data can comprise: for the first numerical value of configuration store capacity and the second value for configuration store performance;
With the service feature data of described target service for input, utilize the storage resources allocation algorithm that described target service is corresponding, calculate the storage resources configuration data that described target service is corresponding;
Send described storage resources configuration data by R-T unit 605 to storage system, make described storage system that the storage resources meeting described storage resources configuration data is distributed to described target service.
In the embodiment of the present invention, after sending described storage resources configuration data by R-T unit 605 to storage system, processor 601 can also receive the storage resources configuration result of storage system transmission by R-T unit 605, and send described storage resources configuration result by R-T unit 605 to operation system, in order to will distribute to the storage resources informing business system of described target service, the storage resources making the described target service operated in operation system can use described storage resources configuration result to represent.
Need to illustrate, the operation system in described operation system and Fig. 1, the storage system in described storage system and Fig. 1.
In the embodiment of the present invention, storage resources allocation algorithm is for calculating the storage resources configuration data of a service needed.Understandable, the storage resources of a service needed is relevant to the service feature of this business.During concrete enforcement, can according to the empirical data of the business of each type of service, that is: history service characteristic and memory capacity, the memory property of historical storage resource that have been assigned with, set up out the storage resources apportion model that the business of each type of service is corresponding.
Because the storage resources allocation algorithm in described preset algorithm set draws according to the empirical data analysis modeling of each type of service, therefore, discrepancy may be there is between the storage resources that the storage resources configuration data that processor 601 utilizes storage resources allocation algorithm corresponding to the described target service in described preset algorithm set to calculate characterizes and the storage resources of the current actual needs of described target service.
In order to ensure reasonable distribution and the Appropriate application of storage resources, after the storage resources meeting described storage resources configuration data is distributed to described target service, processor 601 can also monitor the actual service condition of described target service to the described storage resources distributed, and adjust storage resources configuration data corresponding to described target service according to described actual service condition, the described storage resources configuration data after adjustment is sent to described storage system finally by R-T unit 605, make described storage system that the storage resources meeting the described storage resources configuration data after adjustment is distributed to described target service.
Concrete, processor 601 can adjust storage resources allocation algorithm corresponding to described target service according to described actual service condition, and utilize the storage resources allocation algorithm after adjustment to calculate storage resources configuration data corresponding to the described target service after adjustment, can realize adjusting storage resources configuration data corresponding to described target service according to described actual service condition, and can realize training described storage resources allocation algorithm, make the result of calculation of described storage resources allocation algorithm closer to the storage resources of described target service actual demand, the reasonable distribution of further raising storage resources and utilization.
Concrete, the implementation that processor 601 adjusts storage resources allocation algorithm corresponding to described target service with reference to the related content in figure 2 embodiment, can repeat no more here.
Intelligible, the detailed step that processor 601 performs also can refer to the particular content in Fig. 2 or Fig. 4 embodiment of the method corresponding respectively, repeats no more here.
In sum, implement the embodiment of the present invention, described storage resources distribution system is after reception storage resources distributes request, from default resource allocation algorithm, the resource allocation algorithm for described target service is selected according to the type of service of described target service, and with the service feature data of described target service for input, the described resource allocation algorithm selected is utilized to calculate the storage resources configuration data of described target service, the storage resources meeting described storage resources configuration data is distributed to described target service by then instruction storage system, can realize according to service feature is each traffic assignments storage resources in operation system automatically, improve the efficiency of Resourse Distribute.
One of ordinary skill in the art will appreciate that all or part of flow process realized in above-described embodiment method, that the hardware that can carry out instruction relevant by computer program has come, described program can be stored in computer read/write memory medium, this program, when performing, can comprise the flow process of the embodiment as above-mentioned each side method.Wherein, described storage medium can be magnetic disc, CD, read-only store-memory body (Read-OnlyMemory, ROM) or random store-memory body (RandomAccessMemory, RAM) etc.

Claims (14)

1. a storage resources distribution method, is characterized in that, comprising:
Receive the resource allocation request for target service; Described request comprises the type of service of described target service and the service feature data of described target service;
According to the type of service of described target service, from preset algorithm set, select the storage resources allocation algorithm for described target service; Described preset algorithm set comprises multiple storage resources allocation algorithm, wherein, and the corresponding a kind of type of service of each storage resources allocation algorithm;
With the service feature data of described target service for input, utilize the storage resources allocation algorithm that described target service is corresponding, calculate the storage resources configuration data that described target service is corresponding, described storage resources configuration data can comprise: for the first numerical value of configuration store capacity and the second value for configuration store performance;
Resource allocation request is sent to storage system, described resource allocation request comprises described storage resources configuration data, and described resource allocation request is used to indicate described storage system and the storage resources meeting described storage resources configuration data is distributed to described target service.
2. the method for claim 1, is characterized in that, also comprises:
Monitor the service condition of described target service to the described storage resources distributed;
Storage resources configuration data corresponding to described target service is adjusted according to described service condition;
Resource allocation request is resend to described storage system, the described resource allocation request resend comprises the described storage resources configuration data after adjustment, described in the resource allocation request that resends be used to indicate described storage system the storage resources meeting the described storage resources configuration data after adjustment distributed to described target service.
3. method as claimed in claim 2, is characterized in that, describedly adjusts storage resources configuration data corresponding to described target service according to described service condition, comprising:
Storage resources allocation algorithm corresponding to described target service is adjusted according to described service condition;
With the service feature data of described target service for input, the storage resources allocation algorithm after utilization adjustment calculates storage resources configuration data corresponding to the described target service after adjustment.
4. method as claimed in claim 3, is characterized in that, describedly adjusts storage resources allocation algorithm corresponding to described target service according to described service condition, comprising:
If the memory capacity of the described storage resources distributed described in described service condition shows does not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, reduce the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
5. the method as described in claim 3 or 4, is characterized in that, describedly adjusts storage resources allocation algorithm corresponding to described target service according to described service condition, comprising:
If the memory property of the described storage resources distributed described in described service condition shows does not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory property and the positive correlation degree of memory property, or, reduce the negative correlation variable of memory property and the negative correlation degree of memory property.
6. the method according to any one of claim 3-5, is characterized in that, describedly adjusts storage resources allocation algorithm corresponding to described target service according to described service condition, comprising:
If described service condition shows that the ratio of the memory capacity of the storage resources of the actual use of described target service and the memory capacity of the described described storage resources distributed is lower than first threshold, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, improve the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
7. the method according to any one of claim 3-6, is characterized in that, the described coefficient adjusting the variable in storage resources allocation algorithm corresponding to described target service according to described service condition, comprising:
If described service condition shows that the ratio of the memory property of the storage resources of the actual use of described target service and the memory property of the described described storage resources distributed is lower than Second Threshold, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory property and the positive correlation degree of memory property, or, improve the negative correlation variable of memory property and the negative correlation degree of memory property.
8. a storage resources distribution system, is characterized in that, comprising:
Receiver module, for receiving the resource allocation request for target service; Described request comprises the type of service of described target service and the service feature data of described target service;
Algorithms selection module, for the type of service according to described target service, selects the storage resources allocation algorithm for described target service from preset algorithm set; Described preset algorithm set comprises multiple storage resources allocation algorithm, wherein, and the corresponding a kind of type of service of each storage resources allocation algorithm;
Computing module, for with the service feature data of described target service for input, utilize the storage resources allocation algorithm that described target service is corresponding, calculate the storage resources configuration data that described target service is corresponding, described storage resources configuration data can comprise: for the first numerical value of configuration store capacity and the second value for configuration store performance;
Resource distribution module, for sending resource allocation request to storage system, described resource allocation request comprises described storage resources configuration data, and described resource allocation request is used to indicate described storage system and the storage resources meeting described storage resources configuration data is distributed to described target service.
9. system as claimed in claim 8, also comprises: monitoring modular and adjusting module, wherein:
Described monitoring modular, for sending described storage resources configuration data to storage system in described resource distribution module, make after the storage resources meeting described storage resources configuration data distributed to described target service data by described storage system, to monitor the service condition of described target service to the described storage resources distributed;
Described adjusting module, for according to described monitoring module monitors to described service condition adjust storage resources configuration data corresponding to described target service, and trigger described resource distribution module and resend resource allocation request to described storage system, the described resource allocation request resend comprises the described storage resources configuration data after adjustment, described in the resource allocation request that resends be used to indicate described storage system the storage resources meeting the described storage resources configuration data after adjustment distributed to described target service.
10. system as claimed in claim 9, described adjusting module, specifically for: according to described monitoring module monitors to described service condition adjust storage resources allocation algorithm corresponding to described target service, and triggering described computing module with the service feature data of described target service for input, the storage resources allocation algorithm after utilization adjustment calculates storage resources configuration data corresponding to the described target service after adjustment.
11. systems as claimed in claim 10, described adjusting module, specifically for: if described service condition show described in the memory capacity of described storage resources of having distributed do not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, reduce the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
12. systems as described in claim 10 or 11, described algorithm adjusting module, specifically for: if described service condition show described in the memory property of described storage resources of having distributed do not meet the demand of described target service, then in the storage resources allocation algorithm that described target service is corresponding, improve the positive correlation variable of memory property and the positive correlation degree of memory property, or, reduce the negative correlation variable of memory property and the negative correlation degree of memory property.
13. systems according to any one of claim 10-12, described algorithm adjusting module, specifically for: if described service condition shows that the ratio of the memory capacity of the storage resources of the actual use of described target service and the memory capacity of the described described storage resources distributed is lower than first threshold, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory capacity and the positive correlation degree of memory capacity, or, improve the variable of negative correlation and the negative correlation degree of memory capacity of memory capacity.
14. systems according to any one of claim 10-13, described algorithm adjusting module, specifically for: if described service condition shows that the ratio of the memory property of the storage resources of the actual use of described target service and the memory property of the described described storage resources distributed is lower than Second Threshold, then in the storage resources allocation algorithm that described target service is corresponding, reduce the positive correlation variable of memory property and the positive correlation degree of memory property, or, improve the negative correlation variable of memory property and the negative correlation degree of memory property.
CN201510821385.2A 2015-11-23 2015-11-23 Storage resource distribution method and storage resource distribution system Pending CN105516242A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201510821385.2A CN105516242A (en) 2015-11-23 2015-11-23 Storage resource distribution method and storage resource distribution system
PCT/CN2016/106651 WO2017088717A1 (en) 2015-11-23 2016-11-21 Storage resource allocation method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510821385.2A CN105516242A (en) 2015-11-23 2015-11-23 Storage resource distribution method and storage resource distribution system

Publications (1)

Publication Number Publication Date
CN105516242A true CN105516242A (en) 2016-04-20

Family

ID=55723873

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510821385.2A Pending CN105516242A (en) 2015-11-23 2015-11-23 Storage resource distribution method and storage resource distribution system

Country Status (2)

Country Link
CN (1) CN105516242A (en)
WO (1) WO2017088717A1 (en)

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017088717A1 (en) * 2015-11-23 2017-06-01 华为技术有限公司 Storage resource allocation method and system
CN107463338A (en) * 2017-08-15 2017-12-12 郑州云海信息技术有限公司 A kind for the treatment of method and apparatus of logical volume
CN107688492A (en) * 2016-08-05 2018-02-13 阿里巴巴集团控股有限公司 Control method, device and the cluster resource management system of resource
CN107770236A (en) * 2017-07-04 2018-03-06 中国电子科技集团公司电子科学研究院 A kind of storage resource distribution method, device and computer-readable medium
WO2018099311A1 (en) * 2016-12-02 2018-06-07 中兴通讯股份有限公司 Resource allocation method, apparatus, system and computer storage medium
CN108268421A (en) * 2016-12-31 2018-07-10 英特尔公司 For providing the mechanism of reconfigurable data Layer in rack scale environment
CN109034381A (en) * 2017-06-09 2018-12-18 宏达国际电子股份有限公司 Training mission optimization system, method and its non-transient computer readable media
CN109144403A (en) * 2017-06-19 2019-01-04 阿里巴巴集团控股有限公司 A kind of method and apparatus for cloud disk pattern switching
CN109582234A (en) * 2018-11-23 2019-04-05 金色熊猫有限公司 Storage resources distribution method, device, electronic equipment and computer-readable medium
CN110611578A (en) * 2018-06-15 2019-12-24 华为技术有限公司 Method and device for allocating network resources
CN110830391A (en) * 2018-08-10 2020-02-21 阿里巴巴集团控股有限公司 Resource allocation method and device and cluster system
CN111176823A (en) * 2018-11-09 2020-05-19 深圳市中兴微电子技术有限公司 Method and device for allocating storage resources and computer readable storage medium
CN111443870A (en) * 2020-03-26 2020-07-24 腾讯科技（深圳）有限公司 Data processing method, device and storage medium
CN111918314A (en) * 2020-08-20 2020-11-10 浪潮思科网络科技有限公司 Mobile network optimization method and system based on distributed SDN
CN112637299A (en) * 2020-12-15 2021-04-09 中国联合网络通信集团有限公司 Cloud resource allocation method, device, equipment, medium and product
CN112866386A (en) * 2021-01-19 2021-05-28 青岛越超传媒有限公司 Data storage data construction method based on cloud computing
CN113452586A (en) * 2021-06-11 2021-09-28 青岛海尔科技有限公司 Method and device for registering edge computing node and intelligent home system
CN113721855A (en) * 2021-09-01 2021-11-30 中国建设银行股份有限公司 Storage method and device of storage resources, electronic equipment and computer storage medium
CN117076093A (en) * 2023-10-16 2023-11-17 北京乐讯科技有限公司 Storage resource scheduling method and device based on machine learning and storage medium

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112288123B (en) * 2020-06-16 2023-01-31 北京沃东天骏信息技术有限公司 Resource attribute prediction method and system, training method and related equipment
CN111930299B (en) * 2020-06-22 2024-01-26 中国建设银行股份有限公司 Method for distributing storage units and related equipment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102023818A (en) * 2010-12-06 2011-04-20 成都市华为赛门铁克科技有限公司 Method and device for regulating capacity of storage equipment, and storage equipment
CN103605484A (en) * 2013-11-26 2014-02-26 华为技术有限公司 Data storage method and storage server
CN103685517A (en) * 2013-12-13 2014-03-26 浪潮(北京)电子信息产业有限公司 Storage hierarchical scheduling method and system based on service class characteristics
CN104717247A (en) * 2013-12-13 2015-06-17 中国电信股份有限公司 Method and system for dynamically scheduling storage resources in cloud storage system
CN104735095A (en) * 2013-12-18 2015-06-24 中国移动通信集团四川有限公司 Method and device for job scheduling of cloud computing platform

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103179048B (en) * 2011-12-21 2016-04-13 中国电信股份有限公司 Main frame qos policy transform method and the system of cloud data center
US9251115B2 (en) * 2013-03-07 2016-02-02 Citrix Systems, Inc. Dynamic configuration in cloud computing environments
CN103581339A (en) * 2013-11-25 2014-02-12 广东电网公司汕头供电局 Storage resource allocation monitoring and processing method based on cloud computing
CN104580524A (en) * 2015-01-30 2015-04-29 华为技术有限公司 Resource scaling method and cloud platform with same
CN105516242A (en) * 2015-11-23 2016-04-20 华为技术有限公司 Storage resource distribution method and storage resource distribution system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102023818A (en) * 2010-12-06 2011-04-20 成都市华为赛门铁克科技有限公司 Method and device for regulating capacity of storage equipment, and storage equipment
CN103605484A (en) * 2013-11-26 2014-02-26 华为技术有限公司 Data storage method and storage server
CN103685517A (en) * 2013-12-13 2014-03-26 浪潮(北京)电子信息产业有限公司 Storage hierarchical scheduling method and system based on service class characteristics
CN104717247A (en) * 2013-12-13 2015-06-17 中国电信股份有限公司 Method and system for dynamically scheduling storage resources in cloud storage system
CN104735095A (en) * 2013-12-18 2015-06-24 中国移动通信集团四川有限公司 Method and device for job scheduling of cloud computing platform

Cited By (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017088717A1 (en) * 2015-11-23 2017-06-01 华为技术有限公司 Storage resource allocation method and system
CN107688492A (en) * 2016-08-05 2018-02-13 阿里巴巴集团控股有限公司 Control method, device and the cluster resource management system of resource
CN107688492B (en) * 2016-08-05 2021-11-09 阿里巴巴集团控股有限公司 Resource control method and device and cluster resource management system
WO2018099311A1 (en) * 2016-12-02 2018-06-07 中兴通讯股份有限公司 Resource allocation method, apparatus, system and computer storage medium
CN108156628A (en) * 2016-12-02 2018-06-12 中兴通讯股份有限公司 A kind of method, apparatus and system of resource allocation
CN108268421B (en) * 2016-12-31 2024-03-29 英特尔公司 Mechanism for providing a reconfigurable data layer in a rack scale environment
CN108268421A (en) * 2016-12-31 2018-07-10 英特尔公司 For providing the mechanism of reconfigurable data Layer in rack scale environment
US11144828B2 (en) 2017-06-09 2021-10-12 Htc Corporation Training task optimization system, training task optimization method and non-transitory computer readable medium for operating the same
CN109034381A (en) * 2017-06-09 2018-12-18 宏达国际电子股份有限公司 Training mission optimization system, method and its non-transient computer readable media
CN109144403B (en) * 2017-06-19 2021-10-01 阿里巴巴集团控股有限公司 Method and equipment for switching cloud disk modes
CN109144403A (en) * 2017-06-19 2019-01-04 阿里巴巴集团控股有限公司 A kind of method and apparatus for cloud disk pattern switching
CN107770236A (en) * 2017-07-04 2018-03-06 中国电子科技集团公司电子科学研究院 A kind of storage resource distribution method, device and computer-readable medium
CN107463338A (en) * 2017-08-15 2017-12-12 郑州云海信息技术有限公司 A kind for the treatment of method and apparatus of logical volume
CN110611578A (en) * 2018-06-15 2019-12-24 华为技术有限公司 Method and device for allocating network resources
US11265221B2 (en) 2018-06-15 2022-03-01 Huawei Technologies Co., Ltd. Network resource allocation method and apparatus
CN110611578B (en) * 2018-06-15 2021-08-27 华为技术有限公司 Method and device for allocating network resources
CN110830391A (en) * 2018-08-10 2020-02-21 阿里巴巴集团控股有限公司 Resource allocation method and device and cluster system
CN111176823A (en) * 2018-11-09 2020-05-19 深圳市中兴微电子技术有限公司 Method and device for allocating storage resources and computer readable storage medium
CN109582234A (en) * 2018-11-23 2019-04-05 金色熊猫有限公司 Storage resources distribution method, device, electronic equipment and computer-readable medium
CN111443870A (en) * 2020-03-26 2020-07-24 腾讯科技（深圳）有限公司 Data processing method, device and storage medium
CN111443870B (en) * 2020-03-26 2021-08-03 腾讯科技（深圳）有限公司 Data processing method, device and storage medium
CN111918314B (en) * 2020-08-20 2024-03-29 浪潮思科网络科技有限公司 Mobile network optimization method and system based on distributed SDN
CN111918314A (en) * 2020-08-20 2020-11-10 浪潮思科网络科技有限公司 Mobile network optimization method and system based on distributed SDN
CN112637299A (en) * 2020-12-15 2021-04-09 中国联合网络通信集团有限公司 Cloud resource allocation method, device, equipment, medium and product
CN112637299B (en) * 2020-12-15 2022-04-26 中国联合网络通信集团有限公司 Cloud resource allocation method, device, equipment, medium and product
CN112866386A (en) * 2021-01-19 2021-05-28 青岛越超传媒有限公司 Data storage data construction method based on cloud computing
CN113452586A (en) * 2021-06-11 2021-09-28 青岛海尔科技有限公司 Method and device for registering edge computing node and intelligent home system
CN113721855A (en) * 2021-09-01 2021-11-30 中国建设银行股份有限公司 Storage method and device of storage resources, electronic equipment and computer storage medium
CN117076093A (en) * 2023-10-16 2023-11-17 北京乐讯科技有限公司 Storage resource scheduling method and device based on machine learning and storage medium
CN117076093B (en) * 2023-10-16 2024-01-05 北京乐讯科技有限公司 Storage resource scheduling method and device based on machine learning and storage medium

Also Published As

Publication number Publication date
WO2017088717A1 (en) 2017-06-01

Similar Documents

Publication Publication Date Title
CN105516242A (en) Storage resource distribution method and storage resource distribution system
US7761629B2 (en) Method for using host and storage controller port information to configure paths between a host and storage controller
US10303365B1 (en) Data fingerprint distribution on a data storage system
US9805140B2 (en) Striping of directed graphs and nodes with improved functionality
CN102939594B (en) The method and apparatus that migration with the virtual resource in customer resources to data center environment is relevant
US20080301333A1 (en) System and article of manufacture for using host and storage controller port information to configure paths between a host and storage controller
US9626116B1 (en) Distributed service level objective management in active-active environments
US8838849B1 (en) Link sharing for multiple replication modes
CN103229487B (en) Partition balancing method, device and server in distributed memory system
US20140189236A1 (en) Data storage method and storage device
US20150355947A1 (en) Resource provisioning based on logical profiles and piecewise objective functions
US20150269000A1 (en) Resource provisioning based on logical profiles and objective functions
US20190317682A1 (en) Metrics driven expansion of capacity in solid state storage systems
CN106776146A (en) A kind of data verification method, apparatus and system
CN103608784B (en) Method for creating network volumes, data storage method, storage device and storage system
CN103562881A (en) Defragmentation of data storage pools
CN107977167B (en) Erasure code based degeneration reading optimization method for distributed storage system
US9569569B1 (en) Data storage system modeling
WO2008056670A1 (en) Resource information providing system, method, resource information providing apparatus, and program
CN103617136A (en) SCSI drive side and I/O request control method
JP2015191324A (en) Storage management device, performance adjustment method, and performance adjustment program
CN109196458B (en) Storage system available capacity calculation method and device
US11579982B2 (en) Calculating bandwidth requirements for a specified recovery point objective
US20180246763A1 (en) Computer system management method and computer system
US9858147B2 (en) Storage apparatus and method of controlling storage apparatus

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20160420