CN107783734A - A kind of resource allocation methods, device and terminal based on super fusion storage system - Google Patents

A kind of resource allocation methods, device and terminal based on super fusion storage system Download PDF

Info

Publication number
CN107783734A
CN107783734A CN201711147145.4A CN201711147145A CN107783734A CN 107783734 A CN107783734 A CN 107783734A CN 201711147145 A CN201711147145 A CN 201711147145A CN 107783734 A CN107783734 A CN 107783734A
Authority
CN
China
Prior art keywords
storage resource
storage
data
stored
resource
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201711147145.4A
Other languages
Chinese (zh)
Other versions
CN107783734B (en
Inventor
彭江
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.)
Beijing Lenovo Super Fusion Technology Co Ltd
Original Assignee
Beijing Lenovo Super Fusion Technology 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 Beijing Lenovo Super Fusion Technology Co Ltd filed Critical Beijing Lenovo Super Fusion Technology Co Ltd
Priority to CN201711147145.4A priority Critical patent/CN107783734B/en
Publication of CN107783734A publication Critical patent/CN107783734A/en
Application granted granted Critical
Publication of CN107783734B publication Critical patent/CN107783734B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0629Configuration or reconfiguration of storage systems
    • G06F3/0631Configuration or reconfiguration of storage systems by allocating resources to storage systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0668Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/067Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

The invention discloses a kind of resource allocation methods, device and terminal based on super fusion storage system, it is related to technical field of memory, for solve prior art for data distribution storage resource when, the method of salary distribution is not reasonable, causes the storage resource allocative efficiency under super fusion storage system relatively low and invents.The method of the present invention includes:The storage demand of data to be stored is calculated according to the storage information of data to be stored;Obtain storage resource list;According to storage demand, corresponding storage resource in storage resource list is filtered using the tag along sort of presetting rule and storage resource, obtain available storage resource list, tag along sort is used for the label classified according to different dimensions to storage resource, and tag along sort comprises at least the one or more in location tags, performance label;The storage resource of storage resource list can be correspondingly used for data distribution to be stored.The present invention is suitably applied in the distribution of storage resource in super fusion storage system.

Description

A kind of resource allocation methods, device and terminal based on super fusion storage system
Technical field
The present invention relates to technical field of memory, more particularly to a kind of resource allocation methods based on super fusion storage system, Device and terminal.
Background technology
With the continuous progress of network technology, the continuous growth of data volume in network, surpass the data in fusion storage system Amount is also incrementally increasing.Wherein, it is a kind of distributed memory system of object-oriented to surpass fusion storage system, and the system is intended to reality Now by the integration of the resources such as the calculating in system, network, storage and virtualized server, multinode is gathered by network with reaching Altogether, realize modular seamless extending transversely, and form the effect of unified resource pool.In face of super fusion storage system In growing data volume when, the storage resource for data storage is often limited.Therefore, how reasonably to distribute and deposit Resource is stored up, is just particularly important for the data storage in super fusion storage system.
Generally, it is data distribution storage resource to be stored when the data in super fusion storage system store Mode is often what is be randomly assigned.However, in actual applications, the mode being randomly assigned can not effectively realize storage resource Reasonable utilization.Simultaneously as the access efficiency for the data that influence is stored therein by the storage resource of different attribute and position, Such as:Type for solid state hard disc (Solid State Disk, abbreviation SSD) storage resource in data access speed than Type is fast for the access speed of the data in the storage resource of mechanical hard disk (Hard Disk Drive, abbreviation HDD), therefore, if Storage resource is randomly assigned, then is likely to by the higher data storage of rate of people logging in the slower storage resource of response speed, and This is clearly inappropriate.Thus, prior art for data distribution storage resource when, the method for salary distribution is not reasonable, causes to surpass The storage resource allocative efficiency merged under storage system is relatively low.
The content of the invention
In view of the above problems, the present invention provides a kind of resource allocation methods, device and end based on super fusion storage system End, main purpose is to realize the reasonable distribution to storage resource, so as to improve the allocative efficiency of storage resource.
In order to solve the above technical problems, in a first aspect, the invention provides a kind of resource based on super fusion storage system Distribution method, this method include:
The storage demand of the data to be stored is calculated according to the storage information of data to be stored;
Storage resource list is obtained, the storage resource list is used to record the storage resource in super fusion storage system;
According to the storage demand, the storage resource is arranged using the tag along sort of presetting rule and the storage resource Corresponding storage resource is filtered in table, obtains available storage resource list, and the tag along sort is used for according to different dimensions The label classified to the storage resource, the tag along sort comprise at least location tags, one kind in performance label or It is a variety of;
For the storage resource of storage resource list can be used described in the data distribution correspondence to be stored.
Optionally, before the acquisition storage resource list, methods described also includes:
According to the Resource Properties of the storage resource, for the contingency table of the corresponding Resource Properties of storage resource addition Label, the Resource Properties of the storage resource include hardware attributes and position attribution.
Optionally, the Resource Properties according to the storage resource, for the corresponding resource of storage resource addition The tag along sort of attribute includes:
According to the hardware attributes of the storage resource, for the performance mark of the corresponding hardware attributes of storage resource addition Label, the performance label include HDD labels or SSD labels;And/or
According to positional information of the storage resource in the super fusion storage system, for storage resource addition pair The location tags of the position attribution are answered, the location tags include one in rack tag, main frame label and area label Kind is a variety of.
Optionally, the storage demand of the data to be stored is comprised at least in performance requirement, locational requirement, loading demand It is one or more;
The storage demand that the data to be stored are calculated according to the storage information of data to be stored includes:
The performance requirement of the data to be stored is determined according to the requirements for access of the data to be stored;And/or
The locational requirement of the data to be stored is determined according to the write-in end address information of the data to be stored;And/or
The loading demand of the data to be stored is determined according to the load condition of the storage resource.
Optionally, the storage demand according to data to be stored, utilizes the classification of presetting rule and the storage resource Label carries out filtering to corresponding storage resource in the storage resource list to be included:
According to the performance requirement of data to be stored and the performance label to corresponding storage in the storage resource list Resource is filtered;And/or
According to the locational requirement of data to be stored and the location tags to corresponding storage in the storage resource list Resource is filtered;And/or
According to the loading demand of data to be stored and the load condition of the storage resource in the storage resource list Corresponding storage resource is filtered.
Optionally, the performance requirement and the performance label according to data to be stored is in the storage resource list Corresponding storage resource, which carries out filtering, to be included:
The performance requirement of the data to be stored is extracted, the storage resource that the performance requirement is used to specify is solid state hard disc SSD or mechanical hard disk HDD;
When the performance requirement is that the storage resource specified is solid-state hard disk SSD, obtains and carried in the storage list The storage resource of SSD labels;
When it is mechanical hard disk HDD that the performance requirement, which is the storage resource specified, obtains and carried in the storage list The storage resource of HDD labels.
Optionally, the locational requirement and the location tags according to data to be stored is in the storage resource list Corresponding storage resource, which carries out filtering, to be included:
The locational requirement of the data to be stored is extracted, the locational requirement is used to specify the storage of different shelf positions to provide One or more in source, the storage resource of different position of host machine, the storage resource of different zones position;
When the locational requirement specifies the storage resource of different shelf positions, by rack tag to the storage resource Corresponding storage resource is filtered in list, to obtain the storage resource of different shelf positions;And/or
When the locational requirement specifies the storage resource of different position of host machine, by main frame label to the storage resource Corresponding storage resource is filtered in list, to obtain the storage resource of different position of host machine;And/or
When the locational requirement specifies the storage resource of different zones position, by area label to the storage resource Corresponding storage resource is filtered in list, to obtain the storage resource of different zones position.
Optionally, it is described according to the loading demand of data to be stored and the load condition of the storage resource to the storage Corresponding storage resource, which carries out filtering, in the Resources list includes:
Extract the load threshold in the loading demand;
Judge whether storage resource of the load more than the load threshold;
If in the presence of obtaining in the storage resource storage resource of the load less than the load threshold.
Optionally, described is the storage resource bag that can be used described in the data distribution to be stored corresponds to storage resource list Include:
Obtain the write-in end address information of the data to be stored;
According to said write end address information, from the address information of the storage resource with storage resource list It is determined that with the immediate storage resource in said write end address;
The data to be stored are preserved into the immediate storage resource.
Optionally, described preserve the data to be stored to the immediate storage resource includes:
The backup quantity specified according to the data to be stored, it is described can be with selecting identical quantity in storage resource list Storage resource, and the data to be stored are preserved into selected storage resource.
Optionally, the backup quantity specified according to the data to be stored, described with storage resource list Selecting the storage resource of identical quantity includes:
Judge whether the backup quantity is more than the quantity that can use storage resource in storage resource list;
If being more than, the storage resource of respective amount is selected to preserve the standby of excess amount in the storage resource list Part.
Second aspect, present invention also offers a kind of resource allocation device based on super fusion storage system, the device bag Include:
Computing unit, for calculating the storage demand of the data to be stored according to the storage information of data to be stored;
Acquiring unit, for obtaining storage resource list, the storage resource list is used to record super fusion storage system In storage resource;
Filter element, for the storage demand calculated according to the computing unit, utilize presetting rule and the storage Corresponding storage resource filters in the storage resource list that the tag along sort of resource obtains to the acquiring unit, and obtaining can With storage resource list, the tag along sort is used for the label classified according to different dimensions to the storage resource, described Tag along sort comprises at least the one or more in location tags, performance label;
Allocation unit, for corresponding to the available storage obtained after the filter element filtering for the data distribution to be stored The storage resource of the Resources list.
Optionally, described device also includes:
Adding device, for the Resource Properties according to the storage resource, for the corresponding money of storage resource addition The tag along sort of source attribute, the Resource Properties of the storage resource include hardware attributes and position attribution.
Optionally, the adding device is specifically used for, and is the storage resource according to the hardware attributes of the storage resource The performance label of the corresponding hardware attributes of addition, the performance label include HDD labels or SSD labels;And/or
The adding device is specifically used for, and is believed according to position of the storage resource in the super fusion storage system Breath, for the location tags of the corresponding position attribution of storage resource addition, the location tags include rack tag, main frame One or more in label and area label.
Optionally, the storage demand of the data to be stored is comprised at least in performance requirement, locational requirement, loading demand It is one or more;
The computing unit includes:
First determining module, for determining the performance of the data to be stored according to the requirements for access of the data to be stored Demand;And/or
Second determining module, the data to be stored are determined for the write-in end address information according to the data to be stored Locational requirement;And/or
3rd determining module, for determining the load need of the data to be stored according to the load condition of the storage resource Ask.
Optionally, the filter element includes:
First filtering module, for the performance requirement according to data to be stored and the performance label to the storage resource Corresponding storage resource is filtered in list;And/or
Second filtering module, for the locational requirement according to data to be stored and the location tags to the storage resource Corresponding storage resource is filtered in list;And/or
3rd filtering module, for the loading demand according to data to be stored and the load condition of the storage resource to institute Corresponding storage resource in storage resource list is stated to be filtered.
Optionally, first filtering module includes:
Extracting sub-module, for extracting the performance requirement of the data to be stored, what the performance requirement was used to specify deposits It is solid-state hard disk SSD or mechanical hard disk HDD to store up resource;
Acquisition submodule, the performance requirement for being extracted when the extracting sub-module are that the storage resource specified is that solid-state is hard During disk SSD, the storage resource that SSD labels are carried in the storage list is obtained;
The acquisition submodule, it is additionally operable to when the performance requirement of extracting sub-module extraction is that the storage resource specified is During mechanical hard disk HDD, the storage resource that HDD labels are carried in the storage list is obtained.
Optionally, second filtering module includes:
Extracting sub-module, for extracting the locational requirement of the data to be stored, the locational requirement is used to specify difference One kind or more in the storage resource of shelf position, the storage resource of different position of host machine, the storage resource of different zones position Kind;
First filter submodule, the locational requirement for being extracted when the extracting sub-module specify depositing for different shelf positions When storing up resource, corresponding storage resource in the storage resource list is filtered by rack tag, to obtain difference The storage resource of shelf position;And/or
Second filter submodule, the locational requirement for being extracted when the extracting sub-module specify depositing for different position of host machine When storing up resource, corresponding storage resource in the storage resource list is filtered by main frame label, to obtain difference The storage resource of position of host machine;And/or
3rd filter submodule, the locational requirement for being extracted when the extracting sub-module specify depositing for different zones position When storing up resource, corresponding storage resource in the storage resource list is filtered by area label, to obtain difference The storage resource of regional location.
Optionally, the 3rd filtering module includes:
Extracting sub-module, for extracting the load threshold in the loading demand;
Judging submodule, for judging whether that load exceedes the storage of the load threshold of extracting sub-module extraction Resource;
Acquisition submodule, if judging that load be present exceedes the negative of extracting sub-module extraction for the judging submodule The storage resource of threshold value is carried, then obtains the storage resource that load in the storage resource is less than the load threshold.
Optionally, the allocation unit includes:
Acquisition module, for obtaining the write-in end address information of the data to be stored;
Determining module, for the write-in end address information obtained according to the acquisition module, storage resource is used from described Determined and the immediate storage resource in said write end address in the address information of storage resource in list;
Preserving module, for the data to be stored to be preserved into the immediate storage resource determined to the determining module In.
Optionally, the preserving module is specifically used for, the backup quantity specified according to the data to be stored, it is described can With the storage resource that identical quantity is selected in storage resource list, and the data to be stored are preserved to selected storage resource In.
Optionally, the preserving module includes:
Judging submodule, for judging the backup quantity can use storage resource in storage resource list described in whether being more than Quantity;
Submodule is preserved, if judging that the backup quantity can be arranged described in being more than with storage resource for the judging submodule The quantity of storage resource in table, then the storage resource of respective amount is selected to preserve excess amount in the storage resource list Backup.
To achieve these goals, according to the third aspect of the invention we, there is provided a kind of based on super fusion storage system Resource allocation terminal, include processor and storage medium in the terminal, wherein, the processor and the storage medium are being entered During the distribution of row storage resource, the resource allocation methods described above based on super fusion storage system are able to carry out.
By above-mentioned technical proposal, it is provided by the invention based on the super fusion resource allocation methods of storage system, device and Terminal, for prior art for data distribution storage resource when, the method for salary distribution is not reasonable, cause it is super fusion storage system under Storage resource allocative efficiency it is relatively low the problem of, the present invention utilizes presetting rule and the storage resource by storage demand Tag along sort filters to the storage resource in system, and it is data distribution pair to be stored to realize according to different storage demands The effect of storage resource is answered, and then is embodied as the function of the different storage resource of different data distributions to be stored.Compared to existing There is technology, the present invention can filter when receiving data to be stored according to different tag along sorts to storage resource, can To realize the effect screened to storage resource.In addition, according to the storage information of data to be stored, data to be stored are calculated Storage demand, further according to the demand, storage resource is targetedly filtered using tag along sort, is directed to so as to realize Different storage demands, distribute the effect of different storage resources, so as to change prior art when treating data storage with Machine distribution storage resource causes the situation of unreasonable distribution, then improves the distribution effect of storage resource in super fusion storage system Rate.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention, And can be practiced according to the content of specification, and in order to allow above and other objects of the present invention, feature and advantage can Become apparent, below especially exemplified by the embodiment of the present invention.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, it is various other the advantages of and benefit it is common for this area Technical staff will be clear understanding.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention Limitation.And in whole accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings:
Fig. 1 shows a kind of resource allocation methods flow based on super fusion storage system provided in an embodiment of the present invention Figure;
Fig. 2 shows another resource allocation methods flow based on super fusion storage system provided in an embodiment of the present invention Figure;
Fig. 3 shows a kind of composition of resource allocation device based on super fusion storage system provided in an embodiment of the present invention Block diagram;
Fig. 4 shows the group of another resource allocation device based on super fusion storage system provided in an embodiment of the present invention Into block diagram.
Embodiment
The exemplary embodiment of the present invention is more fully described below with reference to accompanying drawings.Although the present invention is shown in accompanying drawing Exemplary embodiment, it being understood, however, that may be realized in various forms the present invention without should be by embodiments set forth here Limited.Conversely, there is provided these embodiments are to be able to be best understood from the present invention, and can be by the scope of the present invention Completely it is communicated to those skilled in the art.
In order to realize the reasonable distribution to storage resource, the allocative efficiency of storage resource is improved, the embodiment of the present invention provides A kind of resource allocation methods based on super fusion storage system, as shown in figure 1, this method includes:
101st, the storage demand of data to be stored is calculated according to the storage information of data to be stored.
In general, why to be stored data are, and its reason is to be obtained at any time as needed.Cause This, according to reasons such as the times of acquisition of data, the significance levels of data, different data need different storages to provide in storage Source, while user may also can have different requirements to storage resource in storage.Thus, it is to be stored described in this step The letter of the different contents such as the requirement of the attribute and user of data itself in storage is just contained in the storage information of data Breath.
Due to the storage information of different pieces of information be it is different, therefore, can be in the storage information according to data to be stored Particular content, the storage demand of corresponding data to be stored which kind of required storage resource when being stored calculated.Wherein, this step Rapid described storage demand can be understood as data in storage, be determined according to the storage information of itself for the data distribution , it is necessary to distribute the demand of which kind of storage resource during storage resource.
Thus, when data enter excess of imports fusion storage system, due to including different storage informations in different data, The then method according to this step, by calculating the storage information of the data to be stored, and according to result of calculation To determine the storage demand of data to be stored, specifically it can be the data in storage need which kind of form storage carried out Storage demand, or need the storage demand of which kind of storage resource.
It should be noted that the storage information of data to be stored described in this step can be stored in data to be stored During operation according to user's it needs to be determined that or data attribute in itself determines, or comprising above-mentioned two situations, It can specifically be determined according to real-time condition, do not limited herein.
For example, when being calculated by treating the storage information of data storage, it is that a needs are multiple to obtain the data During the frequently-used data of access, then it can determine the data when in face of multiple access afterwards, it is necessary to faster access speed.Cause This, it may be determined that the data need to be stored in the faster storage resource of response speed, and then can determine depositing for the data Storage demand is the faster storage resource of response speed.
102nd, storage resource list is obtained.
For any one storage system, for specification storage resource, generally storage resource can all be integrated In a list, in order to be managed to storage resource therein, similarly, for super fusion storage system, due to Wherein include substantial amounts of storage resource, it is also desirable to storage resource is integrated in a list, form storage resource list, so as to Storage resource in system is managed, wherein, the storage resource list described in this step can be understood as super for recording Merge the list of the storage resource in storage system.In addition, in the storage resource list, except the storage money can be included Outside the title in source, it can also include one or more in the much informations such as the sequence number, species, position of storage resource.
Thus according to the method described in this step, the storage resource list in the super emerging system can be obtained Take, to carry out the operation of subsequent step.Specifically, the mode for obtaining the storage resource list can be by the super fusion Obtained in the system record of internal system, can also by other means carry out, not limit herein, can enter as needed Row selection.
Method with reference to described in this step, when title, sequence number and storage money that storage resource is included in storage resource list During the type of hardware in source, the storage resource list got can be specific as shown in table 1 with as follows:
Table 1
1 Storage resource A SSD
2 Storage resource B HDD
3 Storage resource C HDD
4 Storage resource D SSD
5 Storage resource E HDD
6 Storage resource F HDD
7 Storage resource G SSD
8 Storage resource H SSD
In the storage resource list according to shown by table 1, digital " 1-8 " is the sequence number of storage resource, " storage resource A- storage resources H " is that the title of storage resource, " SSD " represent the type of hardware of storage resource with " HDD " respectively.
103rd, according to storage demand, using the tag along sort of presetting rule and storage resource to corresponding in storage resource list Storage resource filtered, obtain available storage resource list.
Due to substantial amounts of storage resource in super fusion storage system be present, the type of different storage resources is different , so that for the hardware attributes of storage resource two kinds of " SSD " and " HDD " can be included, and when with other dimensions to distinguish, such as Using location dimension, then may include other different zones or the storage resource of different main frames, therefore, in this step, it is necessary to According to the storage demand calculated in step 101, different types of storage resource is carried out using presetting rule and tag along sort Filtering, to be met the storage resource of different storage demands.
Need what is illustrated, in this step, described tag along sort be appreciated that for according to different dimensions to the storage The label that resource is classified, also, tag along sort can include such as location tags, at least the one of the different labels of performance label Kind label, specifically, not limiting herein, is determined according to actual conditions.
For example, when the storage demand of the data to be stored calculated is to specify the type of hardware of storage resource to be SSD, then Method according to this step, need to be to not being that the storage resource of SSD types filters in storage resource list, i.e., by type Screened out for HDD storage resource, storage class is SSD storage resource after being filtered, and then obtains available storage money Source list.Specifically, with reference to the table 1 described in step 102, the method according to this step, available storage money can be obtained Source list, it is specific as shown in table 2:
Table 2
1 Storage resource A SSD
4 Storage resource D SSD
7 Storage resource G SSD
8 Storage resource H SSD
104th, the storage resource of storage resource list can be correspondingly used for data distribution to be stored.
After step 103, available storage resource list can be obtained, therefore, in this step can be by the list Corresponding storage resource distributes to data to be stored, so that data to be stored are stored.
Specifically, with reference to according to the example in step 103, can be by the storage resource A in table 2, storage resource D, storage Resource G, storage resource H distribute to data to be stored, to meet that data needs to be stored are stored in SSD type storage resources Storage demand.It is used to store data to be stored wherein it is possible to select a storage resource at random in four storage resources, also may be used To choose by other means, do not limit herein.
Resource allocation methods provided in an embodiment of the present invention based on super fusion storage system, for prior art for number During according to distribution storage resource, the method for salary distribution is not reasonable, causes the storage resource allocative efficiency under super fusion storage system relatively low The problem of, the present invention is by storage demand, using the tag along sort of presetting rule and the storage resource to the storage in system Resource is filtered, and realizes the effect for corresponding to storage resource for data distribution to be stored according to different storage demands, and then It is embodied as the function of the different storage resource of different data distributions to be stored.Compared to prior art, the present invention is receiving During data to be stored, storage resource can be filtered according to different tag along sorts, it is possible to achieve storage resource is carried out The effect of screening.In addition, according to the storage information of data to be stored, the storage demand of data to be stored is calculated, further according to this Demand, storage resource is targetedly filtered using tag along sort, so as to realize for different storage demands, distribution The effect of different storage resources, storage resource is randomly assigned so as to changing prior art when treating data storage to be caused point With irrational situation, the super allocative efficiency for merging storage resource in storage system is then improved.
Further, another base is additionally provided as the refinement and extension to embodiment illustrated in fig. 1, the embodiment of the present invention In the resource allocation methods of super fusion storage system, as shown in Figure 2.Specifically, implementation process can be with as follows:
201st, according to the Resource Properties of storage resource, for the tag along sort of the corresponding Resource Properties of storage resource addition.
Due to substantial amounts of storage resource be present in super fusion storage system, for the ease of being managed to storage resource, , it is necessary to add the tag along sort of corresponding Resource Properties according to the Resource Properties of storage resource after storage resource is linked into system. Wherein, the Resource Properties described in the embodiment of the present invention can be understood as discriminative attributes the characteristics of for characterizing storage resource, The one or more in other attributes such as hardware attributes and position attribution can wherein be included.
Due to the difference that Resource Properties are, therefore the tag along sort of corresponding addition is also different.For example, when according to storage When the hardware attributes of resource are to add label, then what is added is the performance label of corresponding hardware attributes, such as HDD or SSD;Work as basis When the position attribution of resource is to add label, then what is added is the location tags of correspondence position attribute, such as rack tag, main frame mark One or more in the label such as label and area label.Meanwhile before tag along sort is added, to the Resource Properties of storage resource Acquisition storage resource can be tested to obtain by using prior art, or the storage resource in system journal Content in details table is obtained, and the mode of acquisition does not limit herein, can be chosen as needed.
Can be the performance label in the above-mentioned label of storage resource addition and position it should be noted that when adding label Label is put, or only adds one kind therein.In addition, when adding tag along sort for storage resource, can also further root According to the interface type of storage resource, the interface label of the type of the corresponding different interface of addition, such as PCI-E, NVME etc., herein not Limit, the species and quantity of tag along sort can be selected according to being actually needed.
Method according to this step, with reference to specific application scenarios, it is exemplified below:
When having accessed new storage resource A in super fusion storage system, test analysis is carried out to storage resource A, passed through Analysis result obtains storage resource A Resource Properties, and is added as needed on tag along sort, due to storage resource A hardware category Property for conventional mechanical hard disk, then according to the performance label that its hardware attributes is the corresponding mechanical hard disk of storage resource A additions " HDD ", when also needing to storage resource A point of addition labels, because storage resource A is stored in the main frame a in frame 1 In the XI of region, then " the shelf position 1- position of host machine a- regional locations of the position are corresponded to for its addition according to its storage location XI " location tags
Method according to this step, by the Resource Properties according to the storage resource in super fusion storage system, it is Tag along sort corresponding to its addition, and then the classification to storage resource can be realized, so as to be established for the reasonable distribution of storage resource Fixed basis.
202nd, the storage demand of data to be stored is calculated according to the storage information of data to be stored.
Wherein, the storage demand described in the embodiment of the present invention is comprised at least in performance requirement, locational requirement, loading demand It is one or more.Specifically, the storage information of the data to be stored and description and the previous embodiment of the storage demand Step 101 in description it is identical, will not be described here.
Meanwhile the method according to this step, because storage demand includes performance requirement, locational requirement, loading demand Three kinds, therefore, corresponding three kinds of storage demands, the calculation of corresponding three kinds of demands be present, specifically, including following three Individual aspect:
In a first aspect, the performance requirement of the data to be stored is determined according to the requirements for access of the data to be stored.By Different requirements for access be present in different data to be stored, therefore, it is necessary to determine that the data exist according to different requirements for access Different performance requirement during storage.For example, when the data that data to be stored are repeatedly accessed for a kind of needs, due to storing After need to face substantial amounts of access times, in order to improve access efficiency, it is necessary to by data storage to be stored response speed compared with In fast storage resource, because different hardware attributes in storage resource determine different performances, such as traditional mechanical hard disk HDD response speed will be slower than the response speed of solid-state hard disk SSD, therefore, when according to the requirements for access of resource to be stored for visit When asking the situation of speed, then according to the hardware attributes of storage resource, it may be determined that selection performance label depositing for SSD types It is performance requirement of the data to be stored in storage to store up resource.
Second aspect, the position need of the data to be stored are determined according to the write-in end address information of the data to be stored Ask.Because transmission speed of the data to be stored in different positions is different, therefore, it is necessary to consider the compatibility of data, its In, the position that the compatibilities of data can be understood as data storage is nearer from the access end of visitor, and the path of data transfer is got over Short, transmission speed is faster, better so as to the compatibility of data, and under normal conditions, the write-in end of data is data Access end.Therefore, for this angle, it is determined that the write-in end address of data actually just can determine the access end of data Address, so as to according to the address, determine the demand on storage location in data storage requirement to be stored.Certainly, from another For one angle, data are in storing process, from the nearer storage resource in write-in end address, the data of storage be it is faster, There is exception to ensure data not in transmitting procedure, and then cause data storage abnormal, can be by nearest away from write-in end The storage resource of storage location store data to be stored, so as to ensureing the security of data.It should be noted that according to treating The write-in end address of data storage determines that the locational requirement of data to be stored is only one kind in actual determination mode, can be with Determine, herein do not limit by other means.Thus, parsed by treating the write-in end address of data storage, energy Enough identify that the data to be stored need storage resource where, and then the storage location according to needed for data to be stored Determine the locational requirement of data to be stored.
The third aspect, the loading demand of the data to be stored is determined according to the load condition of the storage resource.Super Substantial amounts of storage resource be present in fusion storage system, and the load condition of storage resource and differ, therefore based on different Load condition a, it is thus necessary to determine that loading demand is to realize the load balance between storage resource.Wherein, institute of the embodiment of the present invention The loading demand stated, the load threshold actually determined according to the load condition of storage resource, the threshold value can be people A fixed value for setting or carrying out what is calculated during storage resource distribution according to super fusion storage system every time Float value, specifically, the setting of load threshold can be chosen according to being actually needed, do not limit herein, and loading demand It is determined that it can be determined with reference to the storage resource load condition under current super fusion storage system and the demand of load balance.Cause This, in the method described in this step, passes through the load condition of storage resource, it may be determined that goes out loading demand, will be stored with realizing The purpose of balancing resource load.Further, it is also possible to the determination according to loading demand, it is possible to achieve in data storage procedure, Storage resource in certain loading range is supplied to the effect of data to be stored.
Method according to this step, for storage information different in data to be stored, by using it is different really Mode is determined to determine different storage demands, it can be ensured that storage demand determines the accuracy of result, is then subsequent filter knot The accuracy of fruit provides safeguard, and ensures that the accuracy in the distribution of storage resource.
203rd, storage resource list is obtained.
Specifically, storage resource list and its acquisition modes described in the embodiment of the present invention and step in previous embodiment 102 description is identical, herein without repeating.
204th, according to storage demand, using the tag along sort of presetting rule and storage resource to corresponding in storage resource list Storage resource filtered, obtain available storage resource list.
According to the calculating in abovementioned steps 202, it may be determined that the species of the storage demand is different, therefore, is corresponded to Different storage demands, it is also different to the filter type of storage resource in this step.
In a first aspect, when the storage demand is performance requirement, then by performance label to the storage resource list In corresponding storage resource filtered.Specifically, the performance requirement of data to be stored can be extracted first, then according to the property Energy demand, is filtered using the performance label of storage resource to the storage resource in the storage resource table, to obtain band There is the storage resource of the label of corresponding performance demand.Wherein, described performance requirement can be that the storage resource for specifying is Any of two kinds of situations of solid-state hard disk SSD or mechanical hard disk HDD.
Therefore, based on two kinds of situations in performance requirement, when performance requirement is that the storage resource specified is solid-state hard disk SSD When, then filtered in the storage resource list, and obtain the storage resource with SSD labels;And when performance requirement is specified Storage resource when being mechanical hard disk HDD, then filtered in the storage resource list, and obtain the storage with HDD labels Resource.
Second aspect, when the storage demand is locational requirement, then by location tags to the storage resource list In corresponding storage resource filtered.Specifically, the locational requirement of data to be stored can be extracted first, then according to specific Locational requirement, the storage resource for meeting the locational requirement is inquired using location tags.Wherein, the money of the locational requirement is met Source can be the storage for specifying the storage resource of different shelf positions, the storage resource of different position of host machine, different zones position One or more in resource.The storage resource of several level positioning is specifically needed, is determined according to actual conditions.
Thus, a variety of situations based on diverse location demand, when the locational requirement is that data to be stored need to be stored in During different shelf position, then need by rack tag, corresponding storage resource in storage resource list is filtered, with Just the storage resource for obtaining different shelf positions selects for data to be stored;When the locational requirement is that data to be stored need to deposit Storage then needs to be filtered by the main frame label of storage resource, obtains depositing for different position of host machine in different position of host machine Storage resource supplies resource selection to be stored;And when the locational requirement is that data to be stored need to be stored in different zones position, Then need to be filtered by the area label of storage resource, the storage resource for obtaining different position of host machine is selected for resource to be stored Select.It should be noted that there may be hierarchical relationship between shelf position, position of host machine and regional location described in this step, Different main frames are set under frame, different zones are set under main frame, therefore, the filter type described in present aspect can also be by difference Rack tag, main frame label, the filter type of area label integrated, to meet the storage demand of data to be stored as not Situation with frame during different zones in different main frames.
The third aspect, when the storage demand is loading demand, then deposited by the load condition of storage resource to described Corresponding storage resource is filtered in storage the Resources list.Wherein, the method described in present aspect is specifically as follows:First, from negative Load threshold is extracted in load demand, then judges whether is storage resource in the storage resource list using the load threshold In the presence of the storage resource more than load threshold, and when it is determined that the storage resource more than load threshold be present, the part is stored Resource is rejected from list and filters out storage resource of the load less than load threshold.Wherein, the load threshold described in this step, Can currently it be visited according in the current read-write data volume of storage resource, storage resource capacity accounting, the access queue of storage resource Any one for asking the various ways such as quantity determines, does not limit, can be chosen as needed herein.
It should be noted that the method for three aspects described in this step, may not be isolated presence in practical operation , it is likely that cause the specific filter type described in this step to carry out integrating and one due to a variety of demands of the presence of storage demand With use.
Thus, it is specific as follows by taking specific application scenarios as an example based on the method described in this step:
Data 1 to be stored need to be stored in the storage resource of HDD types, while in order to ensure the safety of the data storage Property is needed in the different zones that are stored under the different main frames of different shelf positions, and the data need to be stored in load condition When being the storage resource that storage resource capacity accounting amount is less than 20%, then the method according to this step, the specific side of filtering Formula can be with as follows:
When the method according to step 203 gets storage resource list as shown in table 3:
Table 3
Sequence number Title Performance Frame Main frame Region Load
1 Storage resource A HDD 1 a V 20%
2 Storage resource B SSD 1 b V 33%
3 Storage resource C SSD 2 a III 15%
4 Storage resource D HDD 2 c II 44%
5 Storage resource E SSD 3 f I 23%
6 Storage resource F HDD 4 g XI 34%
7 Storage resource G SSD 5 a VI 15%
8 Storage resource H SSD 5 c VII 8%
Wherein, table 3 includes the title of storage resource, sequence number, performance label, rack tag, main frame label, area label And the plurality of kinds of contents such as load condition, storage resource A is corresponded to respectively to storage resource H.
The then method according to this step, first, table 3 is filtered by performance label, due to data to be stored Performance requirement be to be stored in the storage resource of SSD types, therefore obtained list is as shown in table 4 after filtering, specific as follows:
Table 4
Sequence number Title Performance Frame Main frame Region Load
2 Storage resource B SSD 1 b V 33%
3 Storage resource C SSD 2 a III 15%
5 Storage resource E SSD 3 f I 23%
7 Storage resource G SSD 5 a VI 15%
8 Storage resource H SSD 5 c VII 8%
Then, table 4 is filtered according to location tags, because data demand to be stored is stored under different frames not With the different zones in main frame, therefore, after filtering to obtain list as shown in table 5, it is specific as follows:
Table 5
Sequence number Title Performance Frame Main frame Region Load
2 Storage resource B SSD 1 b V 33%
3 Storage resource C SSD 2 a III 15%
5 Storage resource E SSD 3 f I 23%
8 Storage resource H SSD 5 c VII 8%
Finally, table 5 is filtered according to load condition, be less than because data to be stored need to be stored in capacity accounting 20% storage resource, thus may determine that load threshold in loading demand is 20%, the storage resource more than 20% will be by Filtering is gone out, and therefore, the list after filtering can be as shown in table 6, specific as follows:
Table 6
Sequence number Title Performance Frame Main frame Region Load
3 Storage resource C SSD 2 a III 15%
8 Storage resource H SSD 5 c VII 8%
It is the available storage resource list described in the embodiment of the present invention in the table 6 after above-mentioned filtering, drawn, its In, storage resource C and storage resource H in the table are the storage resource for the storage demand for meeting data 1 to be stored.
It should be noted that in filter type under above-mentioned concrete application scene, based on performance label, location tags and The order of the different filter types of load condition can randomly select, can also by storage demand set order priority come It is determined that not limiting herein, can be determined as needed.
Thus, by the method described in this step, by different according to performance requirement, locational requirement and loading demand etc. Storage demand, using corresponding performance label, location tags and load condition to the storage resource in storage resource list Carry out the screening and filtering of different modes, it is possible to achieve the effect for filtering different storage resources.Also, due to different storages Demand is filtered using different tag along sorts, and the storage resource that is obtained after filtering can be made more accurate, so as to be this hair The accuracy of storage resource distribution described in bright embodiment provides guarantee.
205th, the storage resource of storage resource list can be correspondingly used for data distribution to be stored.
After available storage resource list is filtrated to get by step 204, treated due to meeting in available storage resource list The storage resource of data storage storage demand is probably multiple, and considers that data preferably store in storage in this step In the storage resource nearer with data write-in end, to ensure the compatibility of data.Thus, the method according to this step, Specifically it by obtaining the address information at data write-in end to be stored can also determine that the storage in storage resource list can be used to provide The write-in end of which more close data to be stored in source, then the storage resource closest to write-in end address is extracted, And treat data storage using the storage resource and stored.
Simultaneously as data to be stored are when stored, it is sometimes desirable to are backed up according to redundancy strategy, wherein redundancy plan Slightly it is a kind of needs based on data storage security, it is in other storage resources while former data are stored to preserve several parts of pairs more The strategy of notebook data.Therefore, based on redundancy strategy, need to back up quantity to it when data to be stored are in storage and count, Then judge whether the storage resource in currently available storage resource list meets backup quantity according to the result after statistics.When can When being more than or equal to the backup quantity of data to be stored with the storage resource quantity in storage resource list, then it can be deposited available Storage the Resources list extracts sufficient amount of storage resource and supplies data storage to be stored.And when the storage resource in available resources school timetable When quantity is less than the backup quantity of data to be stored, the part that quantity exceeds in data to be stored then can be in the embodiment of the present invention In the storage resource list in storage resource stored.Wherein, storage resource is obtained from storage resource list for treating When data storage stores, the mode of acquisition can be random acquisition, can also be filtered again by setting corresponding step, It can specifically select as needed, then this is not limited.
Specifically, by the example " data 1 " to be stored and " exemplified by table 3- tables 6 ", in step described in abovementioned steps 204 Comprising storage resource C and storage resource H in available storage resource list after 204 filterings, lead to by the method described in this step Cross and address information is parsed, it is the storage resource that distance writes end closest to data 1 to be stored to judge storage resource H, Storage resource H can be then supplied to data 1 to be stored, so that it is stored;And when data 1 to be stored are in storage, it is determined that needing When backing up 3 numbers and meeting redundancy strategy according to this, due to can use the storage resource included in storage resource list only to be deposited comprising two Resource is stored up, quantity is less than the backup quantity of data to be stored, and it is 1 part that wherein quantity, which exceeds,.Therefore, it is necessary to from storage resource list It is middle distribution storage resource with meet data bulk to be stored exceed part storage needs, i.e., from storage resource A, storage resource B, A storage resource is picked out in storage resource D, storage resource E, storage resource F and storage resource G and supplies data storage to be stored.
Method according to this step, can be from available storage money by obtaining the write-in end address of data to be stored The length of transmission path when storage resource nearest therewith is obtained in the list of source, and then storage can be reduced, so as to reduce storage The time that data are spent, at the same also ensure data read when can from nearby storage resource obtain data, improve number According to access efficiency, and then realize the reasonable distribution of storage resource.In addition, when storage resource in available storage resource list When quantity can not meet the backup quantity of data to be stored, by being obtained from storage resource list with being deposited beyond quantity identical Storage resource supplies data storage to be stored, can most meet that it stores the storage resource of needs being embodied as data offer to be stored Simultaneously, moreover it is possible to meet requirement of the data to be stored to stored number under redundancy strategy, so as to reach super fusion storage system The reasonable distribution of storage resource under system, and then improve allocative efficiency.
Further, as the realization to method shown in above-mentioned Fig. 1, the embodiment of the present invention is additionally provided one kind and melted based on super The resource allocation device of storage system is closed, for being realized to the method shown in above-mentioned Fig. 1.The device embodiment and foregoing side Method embodiment is corresponding, and for ease of reading, present apparatus embodiment is no longer carried out one by one to the detail content in preceding method embodiment Repeat, it should be understood that the device in the present embodiment can correspond to the full content realized in preceding method embodiment.Such as Fig. 3 Shown, the device includes:Computing unit 31, acquiring unit 32, filter element 33, allocation unit 34, wherein
Computing unit 31, it can be used for the storage need that the data to be stored are calculated according to the storage information of data to be stored Ask.
Acquiring unit 32, it can be used for obtaining storage resource list, the storage resource list is deposited for recording super fusion Storage resource in storage system.
Filter element 33, can be used for the storage demand calculated according to the computing unit 31, using presetting rule and Corresponding storage resource is carried out in the storage resource list that the tag along sort of the storage resource obtains to the acquiring unit 32 Filtering, obtains available storage resource list, and the tag along sort is used to classify to the storage resource according to different dimensions Label, the tag along sort comprises at least location tags, the one or more in performance label.
Allocation unit 34, it can be used for obtaining after the data distribution to be stored corresponds to the filtering of filter element 33 The storage resource of storage resource list can be used.
Further, as the realization to method shown in above-mentioned Fig. 2, the embodiment of the present invention additionally provides another based on super The resource allocation device of storage system is merged, for being realized to the method shown in above-mentioned Fig. 2.The device embodiment with it is foregoing Embodiment of the method is corresponding, for ease of reading, present apparatus embodiment no longer the detail content in preceding method embodiment is carried out by One repeats, it should be understood that the device in the present embodiment can correspond to the full content realized in preceding method embodiment.Such as figure Shown in 4, the device includes:Computing unit 41, acquiring unit 42, filter element 43, allocation unit 44, wherein
Computing unit 41, it can be used for the storage need that the data to be stored are calculated according to the storage information of data to be stored Ask.
Acquiring unit 42, it can be used for obtaining storage resource list, the storage resource list is deposited for recording super fusion Storage resource in storage system.
Filter element 43, can be used for the storage demand calculated according to the computing unit 41, using presetting rule and Corresponding storage resource is carried out in the storage resource list that the tag along sort of the storage resource obtains to the acquiring unit 42 Filtering, obtains available storage resource list, and the tag along sort is used to classify to the storage resource according to different dimensions Label, the tag along sort comprises at least location tags, the one or more in performance label.
Allocation unit 44, it can be used for obtaining after the data distribution to be stored corresponds to the filtering of filter element 43 The storage resource of storage resource list can be used.
Further, described device also includes:
Adding device 45, it can be used for the Resource Properties according to the storage resource, it is corresponding for storage resource addition The tag along sort of the Resource Properties, so that the label that the filter element 43 adds according to the adding device 45 was carried out Filter, the Resource Properties of the storage resource include hardware attributes and position attribution.
Further, the adding device 45 can be specifically used for, and be described according to the hardware attributes of the storage resource The performance label of the corresponding hardware attributes of storage resource addition, the performance label is HDD labels or SSD labels.And/or
The adding device 45 can be specifically used for, according to position of the storage resource in the super fusion storage system Confidence ceases, and for the location tags of the corresponding position attribution of storage resource addition, the location tags are rack tag, master One or more in machine label and area label.
Further, the storage demand of the data to be stored is comprised at least in performance requirement, locational requirement, loading demand One or more;
The computing unit 41 includes:
First determining module 411, it can be used for determining the number to be stored according to the requirements for access of the data to be stored According to performance requirement.And/or
Second determining module 412, it can be used for treating according to determining the write-in end address information of the data to be stored The locational requirement of data storage.And/or
3rd determining module 413, it can be used for determining the data to be stored according to the load condition of the storage resource Loading demand.
Further, the filter element 43 includes:
First filtering module 431, it can be used for performance requirement and the performance label according to data to be stored to described Corresponding storage resource is filtered in storage resource list.And/or
Second filtering module 432, it can be used for locational requirement and the location tags according to data to be stored to described Corresponding storage resource is filtered in storage resource list.And/or
3rd filtering module 433, it can be used for the load of the loading demand and the storage resource according to data to be stored State filters to corresponding storage resource in the storage resource list.
Further, first filtering module 431 includes:
Extracting sub-module 4311, can be used for the performance requirement for extracting the data to be stored, and the performance requirement is used for The storage resource specified is solid-state hard disk SSD or mechanical hard disk HDD.
Acquisition submodule 4312, it can be used for when the performance requirement of the extracting sub-module 4311 extraction is the storage specified When resource is solid-state hard disk SSD, the storage resource that SSD labels are carried in the storage list is obtained.
The acquisition submodule 4312, it can be additionally operable to when the performance requirement of the extracting sub-module 4311 extraction is specified Storage resource when being mechanical hard disk HDD, obtain the storage resource that HDD labels are carried in the storage list.
Further, second filtering module 432 includes:
Extracting sub-module 4321, can be used for the locational requirement for extracting the data to be stored, and the locational requirement is used for Specify in the storage resource of different shelf positions, the storage resource of different position of host machine, the storage resource of different zones position It is one or more.
First filter submodule 4322, it can be used for when the locational requirement of the extracting sub-module 4321 extraction specifies difference During the storage resource of shelf position, corresponding storage resource in the storage resource list is filtered by rack tag, To obtain the storage resource of different shelf positions.And/or
Second filter submodule 4323, it can be used for when the locational requirement of the extracting sub-module 4321 extraction specifies difference During the storage resource of position of host machine, corresponding storage resource in the storage resource list is filtered by main frame label, To obtain the storage resource of different position of host machine.And/or
3rd filter submodule 4324, it can be used for when the locational requirement of the extracting sub-module 4321 extraction specifies difference During the storage resource of regional location, corresponding storage resource in the storage resource list is filtered by area label, To obtain the storage resource of different zones position.
Further, the 3rd filtering module 433 includes:
Extracting sub-module 4331, it can be used for extracting the load threshold in the loading demand.
Judging submodule 4332, it can be used for judging whether that load exceedes the negative of the extracting sub-module 4331 extraction Carry the storage resource of threshold value.
Acquisition submodule 4333, if can be used for the judging submodule 4332 judges that load be present exceedes extraction The storage resource of the load threshold of module extraction, then obtain storage of the load less than the load threshold in the storage resource and provide Source.
Further, the allocation unit 44 includes:
Acquisition module 441, it can be used for the write-in end address information for obtaining the data to be stored.
Determining module 442, can be used for according to the acquisition module 441 obtain write-in end address information, from it is described can Determined and the immediate storage resource in said write end address with the address information of the storage resource in storage resource list.
Preserving module 443, can be used for preserving the data to be stored to the determining module 442 determine it is closest Storage resource in.
Further, the preserving module 443 can be specifically used for, the backup number specified according to the data to be stored Amount, can use the storage resource that identical quantity is selected in storage resource list, and the data to be stored are preserved to institute described In the storage resource of choosing.
Further, the preserving module 443 includes:
Judging submodule 4431, it can be used for judging the backup quantity can be used in storage resource list described in whether being more than The quantity of storage resource;
Submodule 4432 is preserved, if can be used for the judging submodule 4431 judges that the backup quantity can described in being more than With the quantity of storage resource in storage resource list, then the storage resource of respective amount is selected to protect in the storage resource list Deposit the backup of excess amount.
Further, the embodiment of the present invention additionally provides a kind of resource allocation terminal based on super fusion storage system, institute Stating terminal includes above-mentioned whole devices based in the super resource allocation device for merging storage system, in actual applications, the end Hold as intellectual computing device, terminal is server in this in embodiments of the present invention, is comprised at least in the server:Processor and Storage medium, the specific batch operation of storage resource is carried out by the processor and storage medium, performs institute in above-described embodiment The resource allocation methods based on super fusion storage system stated.
By such scheme, the embodiments of the invention provide a kind of resource allocation methods based on super fusion storage system, Device and terminal.For prior art for data distribution storage resource when, the method for salary distribution is not reasonable, cause it is super fusion storage The problem of storage resource allocative efficiency under system is relatively low, the present invention utilize presetting rule and the storage by storage demand The tag along sort of resource filters to the storage resource in system, and it is data to be stored to realize according to different storage demands The effect of the corresponding storage resource of distribution, and then it is embodied as the function of the different storage resource of different data distributions to be stored.Phase Than in prior art, the present invention can be carried out when receiving data to be stored according to different tag along sorts to storage resource Filtering, it is possible to achieve the effect screened to storage resource.In addition, according to the storage information of data to be stored, calculate and treat The storage demand of data storage, further according to the demand, storage resource is targetedly filtered using tag along sort, so as to Realization is directed to different storage demands, the effect of different storage resources is distributed, so as to change prior art to be stored Storage resource is randomly assigned during data causes the situation of unreasonable distribution, then improves storage resource in super fusion storage system Allocative efficiency.Meanwhile by storage information different in data to be stored, determined using different determination modes different Storage demand, it can be ensured that storage demand determines the accuracy of result, can then be sure that and filters storage money according to storage demand During source, the accuracy of filter result, the accuracy in the distribution of storage resource is ensured that.Further, pass through Using the tag along sort of corresponding different storage demands, such as performance label, location tags, the storage in storage resource list is provided Source carries out the filtering of different modes, it is possible to achieve the effect for filtering different storage resources.Also, due to different storage demands Filtered using corresponding filter type, the storage resource that is obtained after filtering can be made more accurate, so as to be this hair The accuracy of storage resource distribution described in bright embodiment provides guarantee.In addition, the write-in end by obtaining data to be stored Road is transmitted when address, storage resource nearest therewith can be obtained from available storage resource list, and then storage can be reduced The length in footpath, so as to reduce the time that data storage is spent, while also ensure data can be from depositing nearby when reading Resource acquisition data are stored up, improve the access efficiency of data, and then realize the reasonable distribution of storage resource.Also, deposited when available When the quantity of storage resource can not meet the backup quantity of data to be stored in storage the Resources list, by from storage resource list Obtain with supplying data storage to be stored beyond quantity identical storage resource, can most meet being embodied as data offer to be stored While it stores the storage resource needed, moreover it is possible to meet requirement of the data to be stored to stored number under redundancy strategy, from And reached the effect of the reasonable distribution of the storage resource under super fusion storage system, and then improve allocative efficiency.In addition, According to the Resource Properties of the storage resource in super fusion storage system, for tag along sort corresponding to its addition, it is possible to achieve to depositing The classification of resource is stored up, so as to be laid the foundation for the reasonable distribution of storage resource.
The text processing apparatus includes processor and memory, above-mentioned computing unit, acquiring unit, filter element, point Stored in memory as program unit with unit etc., by the said procedure unit of computing device storage in memory To realize corresponding function.
Kernel is included in processor, is gone in memory to transfer corresponding program unit by kernel.Kernel can set one Or more, the reasonable distribution of storage resource is realized by adjusting kernel parameter, improves allocative efficiency.
Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/ Or the form such as Nonvolatile memory, such as read-only storage (ROM) or flash memory (flashRAM), memory includes at least one storage Chip.
The embodiments of the invention provide a kind of equipment, equipment includes processor, memory and storage on a memory and can The program run on a processor, following steps are realized during computing device program:According to the storage information meter of data to be stored Calculate the storage demand of the data to be stored;Storage resource list is obtained, the storage resource list is deposited for recording super fusion Storage resource in storage system;According to the storage demand, using the tag along sort of presetting rule and the storage resource to institute State corresponding storage resource in storage resource list to be filtered, obtain available storage resource list, the tag along sort is used for The label classified according to different dimensions to the storage resource, the tag along sort comprise at least location tags, performance mark One or more in label;For the storage resource of storage resource list can be used described in the data distribution correspondence to be stored.
The embodiment of the present invention additionally provides a kind of computer program product, when being performed on data processing equipment, is suitable to Perform the program of initialization there are as below methods step:Depositing for the data to be stored is calculated according to the storage information of data to be stored Storage demand;Storage resource list is obtained, the storage resource list is used to record the storage resource in super fusion storage system;Root According to the storage demand, using the tag along sort of presetting rule and the storage resource to corresponding in the storage resource list Storage resource is filtered, and obtains available storage resource list, and the tag along sort is used for according to different dimensions to the storage The label that resource is classified, the tag along sort comprise at least the one or more in location tags, performance label;To be described The storage resource of storage resource list can be used described in data distribution to be stored is corresponding.
It should be understood by those skilled in the art that, embodiments herein can be provided as method, system or computer program Product.Therefore, the application can use the reality in terms of complete hardware embodiment, complete software embodiment or combination software and hardware Apply the form of example.Moreover, the application can use the computer for wherein including computer usable program code in one or more The computer program production that usable storage medium is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.) The form of product.
The application is with reference to the flow according to the method for the embodiment of the present application, equipment (system) and computer program product Figure and/or block diagram describe.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided The processors of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce A raw machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real The device for the function of being specified in present one flow of flow chart or one square frame of multiple flows and/or block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which produces, to be included referring to Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that counted Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, so as in computer or The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one The step of function of being specified in individual square frame or multiple square frames.
In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net Network interface and internal memory.
Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/ Or the form such as Nonvolatile memory, such as read-only storage (ROM) or flash memory (flashRAM).Memory is computer-readable medium Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method Or technology come realize information store.Information can be computer-readable instruction, data structure, the module of program or other data. The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moved State random access memory
(DRAM), other kinds of random access memory (RAM), read-only storage (ROM), electrically erasable Read memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only storage (CD-ROM), the more work(of numeral Can CD (DVD) or other optical storages, magnetic cassette tape, the storage of tape magnetic rigid disk or other magnetic storage apparatus or it is any its His non-transmission medium, the information that can be accessed by a computing device available for storage.Defined according to herein, computer-readable Jie Matter does not include temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.
It should also be noted that, term " comprising ", "comprising" or its any other variant are intended to nonexcludability Comprising so that process, method, commodity or equipment including a series of elements not only include those key elements, but also wrapping Include the other element being not expressly set out, or also include for this process, method, commodity or equipment intrinsic want Element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including key element Other identical element in process, method, commodity or equipment also be present.
It will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program product. Therefore, the application can be using the embodiment in terms of complete hardware embodiment, complete software embodiment or combination software and hardware Form.Deposited moreover, the application can use to can use in one or more computers for wherein including computer usable program code The shape for the computer program product that storage media is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.) Formula.
Embodiments herein is these are only, is not limited to the application.To those skilled in the art, The application can have various modifications and variations.All any modifications made within spirit herein and principle, equivalent substitution, Improve etc., it should be included within the scope of claims hereof.

Claims (13)

1. a kind of resource allocation methods based on super fusion storage system, it is characterised in that methods described includes:
The storage demand of the data to be stored is calculated according to the storage information of data to be stored;
Storage resource list is obtained, the storage resource list is used to record the storage resource in super fusion storage system;
According to the storage demand, using the tag along sort of presetting rule and the storage resource in the storage resource list Corresponding storage resource is filtered, and obtains available storage resource list, and the tag along sort is used for according to different dimensions to institute The label that storage resource is classified is stated, the tag along sort comprises at least the one or more in location tags, performance label;
For the storage resource of storage resource list can be used described in the data distribution correspondence to be stored.
2. according to the method for claim 1, it is characterised in that before the acquisition storage resource list, methods described Also include:
According to the Resource Properties of the storage resource, the tag along sort for corresponding to the Resource Properties is added for the storage resource, The Resource Properties of the storage resource include hardware attributes and position attribution.
3. according to the method for claim 2, it is characterised in that the Resource Properties according to the storage resource, for institute Stating the tag along sort of the corresponding Resource Properties of storage resource addition includes:
According to the hardware attributes of the storage resource, the performance label for corresponding to the hardware attributes is added for the storage resource, The performance label includes HDD labels or SSD labels;And/or
According to positional information of the storage resource in the super fusion storage system, corresponding institute is added for the storage resource State the location tags of position attribution, the location tags include rack tag, main frame label and one kind in area label or It is a variety of.
4. according to the method for claim 1, it is characterised in that the storage demand of the data to be stored comprises at least performance One or more in demand, locational requirement, loading demand;
The storage demand that the data to be stored are calculated according to the storage information of data to be stored includes:
The performance requirement of the data to be stored is determined according to the requirements for access of the data to be stored;And/or
The locational requirement of the data to be stored is determined according to the write-in end address information of the data to be stored;And/or
The loading demand of the data to be stored is determined according to the load condition of the storage resource.
5. according to the method for claim 4, it is characterised in that it is described according to storage demand, utilize presetting rule and described The tag along sort of storage resource carries out filtering to corresponding storage resource in the storage resource list to be included:
According to the performance requirement of data to be stored and the performance label to corresponding storage resource in the storage resource list Filtered;And/or
According to the locational requirement of data to be stored and the location tags to corresponding storage resource in the storage resource list Filtered;And/or
According to the loading demand of data to be stored and the load condition of the storage resource to corresponding in the storage resource list Storage resource filtered.
6. according to the method for claim 5, it is characterised in that the performance requirement according to data to be stored and the property Energy label carries out filtering to corresponding storage resource in the storage resource list to be included:
The performance requirement of the data to be stored is extracted, the storage resource that the performance requirement is used to specify is solid-state hard disk SSD Or mechanical hard disk HDD;
When the performance requirement is that the storage resource specified is solid-state hard disk SSD, obtains and marked in the storage list with SSD The storage resource of label;
When it is mechanical hard disk HDD that the performance requirement, which is the storage resource specified, obtains and marked in the storage list with HDD The storage resource of label.
7. according to the method for claim 5, it is characterised in that the locational requirement and institute's rheme according to data to be stored Putting label and carrying out filtering to corresponding storage resource in the storage resource list includes:
Extract the locational requirement of the data to be stored, the locational requirement be used to specifying the storage resource of different shelf positions, One or more in the storage resource of different position of host machine, the storage resource of different zones position;
When the locational requirement specifies the storage resource of different shelf positions, by rack tag to the storage resource list In corresponding storage resource filtered, to obtain the storage resource of different shelf positions;And/or
When the locational requirement specifies the storage resource of different position of host machine, by main frame label to the storage resource list In corresponding storage resource filtered, to obtain the storage resource of different position of host machine;And/or
When the locational requirement specifies the storage resource of different zones position, by area label to the storage resource list In corresponding storage resource filtered, to obtain the storage resource of different zones position.
8. according to the method for claim 5, it is characterised in that the loading demand according to data to be stored and described deposit The load condition of storage resource carries out filtering to corresponding storage resource in the storage resource list to be included:
Extract the load threshold in the loading demand;
Judge whether storage resource of the load more than the load threshold;
If in the presence of obtaining in the storage resource storage resource of the load less than the load threshold.
9. according to the method described in claim any one of 1-8, it is characterised in that described to be corresponded to for the data distribution to be stored It is described to be included with the storage resource of storage resource list:
Obtain the write-in end address information of the data to be stored;
According to said write end address information, can be determined from described with the address information of the storage resource in storage resource list With the immediate storage resource in said write end address;
The data to be stored are preserved into the immediate storage resource.
10. according to the method for claim 9, it is characterised in that described preserve the data to be stored to described most connects Near storage resource includes:
The backup quantity specified according to the data to be stored, it is described can be with selecting depositing for identical quantity in storage resource list Resource is stored up, and the data to be stored are preserved into selected storage resource.
11. according to the method for claim 10, it is characterised in that the backup number specified according to the data to be stored Amount, can select the storage resource of identical quantity to include described with storage resource list:
Judge whether the backup quantity is more than the quantity that can use storage resource in storage resource list;
If being more than, the storage resource of respective amount is selected to preserve the backup of excess amount in the storage resource list.
12. it is a kind of based on it is super fusion storage system resource allocation terminal, it is characterised in that the terminal include processor with Storage medium, the processor and the storage medium carry out storage resource distribution when, be able to carry out as claim 1 to The resource allocation methods based on super fusion storage system any one of claim 11.
13. a kind of resource allocation device based on super fusion storage system, it is characterised in that described device includes:
Computing unit, for calculating the storage demand of the data to be stored according to the storage information of data to be stored;
Acquiring unit, for obtaining storage resource list, the storage resource list is used to record in super fusion storage system Storage resource;
Filter element, for the storage demand calculated according to the computing unit, utilize presetting rule and the storage resource The storage resource list that is obtained to the acquiring unit of tag along sort in corresponding storage resource filter, obtain available deposit The Resources list is stored up, the tag along sort is used for the label classified according to different dimensions to the storage resource, the classification Label comprises at least the one or more in location tags, performance label;
Allocation unit, for corresponding to the available storage resource obtained after the filter element filtering for the data distribution to be stored The storage resource of list.
CN201711147145.4A 2017-11-17 2017-11-17 Resource allocation method, device and terminal based on super-fusion storage system Active CN107783734B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711147145.4A CN107783734B (en) 2017-11-17 2017-11-17 Resource allocation method, device and terminal based on super-fusion storage system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711147145.4A CN107783734B (en) 2017-11-17 2017-11-17 Resource allocation method, device and terminal based on super-fusion storage system

Publications (2)

Publication Number Publication Date
CN107783734A true CN107783734A (en) 2018-03-09
CN107783734B CN107783734B (en) 2021-03-16

Family

ID=61433302

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711147145.4A Active CN107783734B (en) 2017-11-17 2017-11-17 Resource allocation method, device and terminal based on super-fusion storage system

Country Status (1)

Country Link
CN (1) CN107783734B (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108881942A (en) * 2018-06-06 2018-11-23 西安交通大学 A kind of super fusion normality recording and broadcasting system based on distributed objects storage
CN108874316A (en) * 2018-06-06 2018-11-23 郑州云海信息技术有限公司 Method, apparatus, equipment and the readable storage medium storing program for executing of storage resource centralized management
CN109564567A (en) * 2018-10-17 2019-04-02 算丰科技(北京)有限公司 Date storage method, device, electronic equipment and computer readable storage medium
CN110858122A (en) * 2018-08-23 2020-03-03 杭州海康威视系统技术有限公司 Method and device for storing data
CN110928484A (en) * 2018-09-19 2020-03-27 上海仪电(集团)有限公司中央研究院 Hybrid cloud storage method based on software defined storage
CN111221644A (en) * 2018-11-27 2020-06-02 阿里巴巴集团控股有限公司 Resource scheduling method, device and equipment
CN112612660A (en) * 2020-12-16 2021-04-06 海光信息技术股份有限公司 Specification information database creating method, chip selecting method, device and system
US11010071B2 (en) 2018-07-06 2021-05-18 Samsung Electronics Co., Ltd. Solid state drive that allocates stream data to super blocks based on stream information and a memory allocation method thereof
CN112905718A (en) * 2021-03-09 2021-06-04 东莞中国科学院云计算产业技术创新与育成中心 Data management method, system, electronic device and medium based on super-fusion architecture
CN117555493A (en) * 2024-01-11 2024-02-13 北京星辰天合科技股份有限公司 Data processing method, system, device, storage medium and electronic equipment
CN117555493B (en) * 2024-01-11 2024-05-10 北京星辰天合科技股份有限公司 Data processing method, system, device, storage medium and electronic equipment

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1952941A (en) * 2005-10-21 2007-04-25 中国科学院计算技术研究所 A resource allocation method and system for shared memory
CN105446813A (en) * 2014-05-30 2016-03-30 华为技术有限公司 Resource distribution method and device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1952941A (en) * 2005-10-21 2007-04-25 中国科学院计算技术研究所 A resource allocation method and system for shared memory
CN105446813A (en) * 2014-05-30 2016-03-30 华为技术有限公司 Resource distribution method and device

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108881942B (en) * 2018-06-06 2020-11-06 西安交通大学 Super-fusion normal state recorded broadcast system based on distributed object storage
CN108874316A (en) * 2018-06-06 2018-11-23 郑州云海信息技术有限公司 Method, apparatus, equipment and the readable storage medium storing program for executing of storage resource centralized management
CN108881942A (en) * 2018-06-06 2018-11-23 西安交通大学 A kind of super fusion normality recording and broadcasting system based on distributed objects storage
US11010071B2 (en) 2018-07-06 2021-05-18 Samsung Electronics Co., Ltd. Solid state drive that allocates stream data to super blocks based on stream information and a memory allocation method thereof
CN110858122B (en) * 2018-08-23 2023-10-20 杭州海康威视系统技术有限公司 Method and device for storing data
CN110858122A (en) * 2018-08-23 2020-03-03 杭州海康威视系统技术有限公司 Method and device for storing data
CN110928484A (en) * 2018-09-19 2020-03-27 上海仪电(集团)有限公司中央研究院 Hybrid cloud storage method based on software defined storage
CN110928484B (en) * 2018-09-19 2023-12-22 上海仪电(集团)有限公司中央研究院 Hybrid cloud storage method based on software defined storage
CN109564567B (en) * 2018-10-17 2023-07-25 北京算能科技有限公司 Data storage method, device, electronic equipment and computer readable storage medium
CN109564567A (en) * 2018-10-17 2019-04-02 算丰科技(北京)有限公司 Date storage method, device, electronic equipment and computer readable storage medium
CN111221644A (en) * 2018-11-27 2020-06-02 阿里巴巴集团控股有限公司 Resource scheduling method, device and equipment
CN111221644B (en) * 2018-11-27 2023-06-13 阿里巴巴集团控股有限公司 Resource scheduling method, device and equipment
CN112612660A (en) * 2020-12-16 2021-04-06 海光信息技术股份有限公司 Specification information database creating method, chip selecting method, device and system
CN112612660B (en) * 2020-12-16 2024-02-13 海光信息技术股份有限公司 Specification information database creation method, chip selection method, device and system
CN112905718A (en) * 2021-03-09 2021-06-04 东莞中国科学院云计算产业技术创新与育成中心 Data management method, system, electronic device and medium based on super-fusion architecture
CN117555493A (en) * 2024-01-11 2024-02-13 北京星辰天合科技股份有限公司 Data processing method, system, device, storage medium and electronic equipment
CN117555493B (en) * 2024-01-11 2024-05-10 北京星辰天合科技股份有限公司 Data processing method, system, device, storage medium and electronic equipment

Also Published As

Publication number Publication date
CN107783734B (en) 2021-03-16

Similar Documents

Publication Publication Date Title
CN107783734A (en) A kind of resource allocation methods, device and terminal based on super fusion storage system
CN102725753B (en) Method and apparatus for optimizing data access, method and apparatus for optimizing data storage
CN108510311A (en) A kind of method, apparatus and electronic equipment of determining marketing program
CN107391532B (en) Data filtering method and device
CN103309984B (en) The method and apparatus that data process
CN107633184A (en) A kind of database and method and apparatus for being used to manage user right
CN107622061A (en) A kind of method, apparatus and system for determining address uniqueness
CN108153776A (en) Data query method and device
CN107592296A (en) The recognition methods of rubbish account and device
CN109657427A (en) The method and device of permission control
CN110264332A (en) The method, apparatus and electronic equipment that account is entered an item of expenditure in the accounts
CN103500226A (en) Method and device for removing sensitivity of sensitive data
CN107729330A (en) The method and apparatus for obtaining data set
CN109218211A (en) The method of adjustment of threshold value, device and equipment in the control strategy of data flow
CN109426439A (en) The method and device of dilatation is carried out to distributed memory system
CN108109028A (en) A kind of method and device for generating cloud computing commodity, electronic equipment
CN108460673A (en) A kind of processing method and processing device of training data
CN106570005A (en) Database cleaning method and device
CN110020954B (en) Revenue distribution method and device and computer equipment
CN109426905A (en) A kind of determination method and device that the criminal document measurement of penalty deviates
CN107391163A (en) page data loading method, device and application in mobile terminal
CN108108444B (en) Enterprise business unit self-adaptive system and implementation method thereof
CN110457182A (en) A kind of load balancing cluster example operating index monitoring system
CN109582834A (en) Data Risk Forecast Method and device
KR101948603B1 (en) Anonymization Device for Preserving Utility of Data and Method thereof

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant