CN112925472A - Request processing method and device, electronic equipment and computer storage medium - Google Patents
Request processing method and device, electronic equipment and computer storage medium Download PDFInfo
- Publication number
- CN112925472A CN112925472A CN201911242658.2A CN201911242658A CN112925472A CN 112925472 A CN112925472 A CN 112925472A CN 201911242658 A CN201911242658 A CN 201911242658A CN 112925472 A CN112925472 A CN 112925472A
- Authority
- CN
- China
- Prior art keywords
- cloud disk
- read
- write
- storage
- disk storage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0604—Improving or facilitating administration, e.g. storage management
- G06F3/0607—Improving or facilitating administration, e.g. storage management by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0653—Monitoring storage devices or systems
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input 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
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a request processing method, a request processing device, electronic equipment and a computer storage medium, wherein the request processing method comprises the following steps: determining a corresponding cloud disk storage strategy according to the received read-write request, and executing cloud disk operation corresponding to the read-write request according to the determined cloud disk storage strategy; and collecting sample data influencing the read-write performance of a service end according to the operation result of the cloud disk, and adjusting the storage strategy of the cloud disk according to the sample data. According to the scheme provided by the embodiment, the cloud disk storage strategy can be adjusted based on a closed-loop feedback mechanism, so that the cloud disk storage strategy dynamically changes according to the change of the corresponding read-write request, and particularly when the change rule of the read-write request received by the cloud disk is diversified, the scheduling efficiency of cloud disk storage resources can be improved, the generation of long tail effect can be reduced, and the cloud disk storage performance is improved.
Description
Technical Field
The embodiment of the invention relates to the technical field of data processing, in particular to a request processing method and device, electronic equipment and a computer storage medium.
Background
At present, most of various service providers provide services to users through one cloud disk server cluster, and all users of cloud disks share the storage system of one cloud disk server cluster. However, the storage policy in the existing cloud disk storage system is mostly set uniformly based on the rule that all data in the cloud disk server cluster is accessed.
However, when different users access the cloud disk through the service end, the storage access mode is different and changes according to the needs of the users. The storage access mode may be random access to the content stored in the cloud disk, or targeted access to a part of the content stored in the cloud disk.
When a large amount of concurrent storage access exists, the storage access mode of the service end is diversified, and the storage strategy of the existing cloud disk server cluster is single, so that the time consumption for processing the concurrent storage access by the cloud disk server cluster is long, the local long tail effect is easily caused, and the overall performance of the cloud disk is reduced.
Disclosure of Invention
Embodiments of the present invention provide a request processing method, apparatus, electronic device and computer storage medium to solve at least some of the above problems.
According to a first aspect of the embodiments of the present invention, there is provided a request processing method, including: determining a corresponding cloud disk storage strategy according to the received read-write request, and executing cloud disk operation corresponding to the read-write request according to the determined cloud disk storage strategy; and collecting sample data influencing the read-write performance of a service end according to the operation result of the cloud disk, and adjusting the storage strategy of the cloud disk according to the sample data.
According to a second aspect of the embodiments of the present invention, there is provided a request processing apparatus including: the determining module is used for determining a corresponding cloud disk storage strategy according to the received read-write request and executing cloud disk operation corresponding to the read-write request according to the determined cloud disk storage strategy; and the adjusting module is used for acquiring sample data influencing the read-write performance according to the operation result of the cloud disk and adjusting the storage strategy of the cloud disk according to the sample data.
According to a third aspect of embodiments of the present invention, there is provided an electronic apparatus, including: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus; the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the request processing method.
According to a fourth aspect of embodiments of the present invention, there is provided a computer storage medium having stored thereon a computer program which, when executed by a processor, implements the request processing method as described above.
According to the scheme provided by the embodiment of the invention, the corresponding cloud disk storage strategy is determined according to the received read-write request, and the cloud disk operation corresponding to the read-write request is executed according to the determined cloud disk storage strategy; and acquiring sample data influencing the read-write performance of the service end according to the operation result of the cloud disk, and adjusting the cloud disk storage strategy through the sample data, so that the cloud disk storage strategy can be adjusted based on a closed-loop feedback mechanism, the cloud disk storage strategy is dynamically changed according to the change of the corresponding read-write request, namely the cloud disk storage strategy is dynamically adapted to the corresponding read-write request, and especially when the change rule of the read-write request received by the cloud disk is diversified due to the diversification of the storage access mode of the service end, the scheduling efficiency of cloud disk storage resources can be improved through the cloud disk storage strategy dynamically adapted to the read-write request of the service end, the generation of the long tail effect can be further reduced, and the cloud disk storage performance is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present invention, and it is also possible for a person skilled in the art to obtain other drawings based on the drawings.
FIG. 1a is a flowchart illustrating a request processing method according to a first embodiment of the present invention;
FIG. 1b is a schematic view of a usage scenario according to a first embodiment of the present invention;
FIG. 1c is a flowchart of data processing according to a first embodiment of the present invention;
FIG. 2 is a flowchart illustrating a request processing method according to a second embodiment of the present invention;
FIG. 3 is a flowchart illustrating steps of a method for processing a query request according to a second embodiment of the present invention;
fig. 4 is a block diagram of a request processing apparatus according to a third embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments of the present invention shall fall within the scope of the protection of the embodiments of the present invention.
The following further describes specific implementation of the embodiments of the present invention with reference to the drawings.
Example one
Referring to fig. 1a, a flowchart illustrating steps of a request processing method according to a first embodiment of the present invention is shown.
The request processing method of the embodiment comprises the following steps:
s102, determining a corresponding cloud disk storage strategy according to the received read-write request, and executing cloud disk operation corresponding to the read-write request according to the determined cloud disk storage strategy.
In this embodiment, the read-write request may be a read-write request sent by the service end to the cloud disk. The service end is a service end corresponding to the cloud disk server cluster and used for accessing the cloud disk. The service end includes but is not limited to: and accessing a user client of the cloud disk, or a certain server which requests to read and write data in the cloud disk server cluster. The cloud disk can be used as a storage unit of a plurality of service ends and used for storing data of the plurality of service ends respectively.
The service end may generate a read-write request according to an access requirement, and then send the read-write request to the cloud disk, where the read-write request may specifically include, but is not limited to: a request to read data in the cloud disk, or a request to write data to the cloud disk.
As shown in fig. 1b, the number of the service ends may be multiple, for example, the service end 1, the service end 2, the service end 3, and the like shown in fig. 1b, where the multiple service ends all send the read-write request to the cloud disk, the cloud disk returns the processing result to the corresponding service end, and the returned processing result may include: the cloud disk operation corresponding to the read-write request is successfully or unsuccessfully executed, the data returned after the cloud disk operation is executed, and the like. For example, the service end 1 may send a read-write request 1 to the cloud disk, and the cloud disk determines an operation result 1 after executing a cloud disk operation corresponding to the read-write request 1, and returns the operation result 1 to the service end 1; the flows corresponding to the service end 2 and the service end 3 are similar to the flow of the service end 1, and are not described herein again.
In this embodiment, after receiving the read-write request, the cloud disk may determine a corresponding cloud disk storage policy according to the read-write request. And the cloud disk storage strategy is used for scheduling the cloud disk storage resources required by the read-write request so as to execute the cloud disk operation corresponding to the read-write request.
Specifically, multiple groups of cloud disk storage strategies may be stored in advance in a cloud disk, one group of cloud disk storage strategies may include multiple different cloud disk storage strategies, and cloud disk read-write requests from the same service end may share one group of cloud disk storage strategies; when a read-write request is received, the cloud disk can determine a cloud disk storage strategy corresponding to the read-write request according to a service end sending the read-write request; of course, the read-write requests of multiple service ends may also share one set of cloud disk storage policy, which is not limited in this embodiment.
Specifically, the cloud disk storage policy may be used to determine whether to execute a cloud disk operation corresponding to the read-write request, which specific resources are allocated to the cloud disk operation, and the like, so that cloud disk storage resources required for processing the read-write request may be scheduled according to the cloud disk storage policy.
In this embodiment, the read-write request may specifically include: the cloud disk operation method comprises the steps of inquiring requests, writing requests, modifying requests, deleting requests and the like, wherein the cloud disk operation corresponding to different reading and writing requests is different. For example, if the read-write request is an inquiry request, the cloud disk operation may be to read data requested by the inquiry request from a database, write the data into the cloud disk storage, and return the data to the service end through the cloud disk storage; if the read-write request is a write-in request, the cloud disk operation may be to write data corresponding to the write-in request into a cloud disk storage, and write the data into a database through the cloud disk storage.
And S104, collecting sample data influencing the read-write performance of the service end according to the operation result of the cloud disk, and adjusting the storage strategy of the cloud disk according to the sample data.
In this embodiment, the result of the cloud disk operation may affect the read-write performance of the service end. For example, after the cloud disk operation is completed, the completion time may affect the response time corresponding to the read-write request of the service end.
In this embodiment, because the cloud disk generally provides cloud disk services to multiple service ends at the same time, the read-write request received by the cloud disk comes from multiple service ends, and the sample data acquired according to the result of the cloud disk operation may be: sample data affecting the read-write performance of multiple service ends.
The sample data may be, for example, sample data within a preset time length, a preset number of sample data, or the like. Sample data may include, but is not limited to: the processing duration of the read-write request, the hit rate of the query request in the cloud disk storage and the like when the read-write request is the query request.
In this embodiment, since the sample data is data that affects the read-write performance of the service end, the read-write performance of the service end can be evaluated through the collected sample data, and it can be determined which parameters have a larger influence on the read-write performance according to the sample data, so that the cloud disk storage policy can be adjusted according to the sample data to adjust the parameters (such as the hit rate) that have a larger influence on the read-write performance, thereby adjusting the read-write performance of the service end.
The read-write requests from different service ends can also correspond to different cloud disk storage strategies, and when the cloud disk storage strategy is adjusted according to sample data, the cloud disk storage strategy corresponding to a certain service end can be adjusted according to the sample data corresponding to the service end, or a plurality of cloud disk storage strategies can be adjusted while comprehensively considering the sample data corresponding to a plurality of service ends.
In addition, after the cloud disk storage strategy is adjusted this time, the adjusted cloud disk storage strategy may be determined according to a subsequently received read-write request, so that a cloud disk operation may be performed according to the adjusted cloud disk storage strategy, and the cloud disk storage strategy may be adjusted again through step S104, so that a cycle of using the cloud disk storage strategy, adjusting according to a result, and continuing to use the adjusted cloud disk storage strategy is formed, that is, a closed-loop feedback mechanism is formed.
For example, as shown in fig. 1c, a data processing flow chart of this embodiment is shown, and as shown in fig. 1c, after the read-write request is received, cloud disk operation corresponding to the read-write request may be executed according to a cloud disk storage policy, and a cloud disk operation result may be determined, where the cloud disk operation result may be used to determine an operation result returned to the service end. Sampling the cloud disk operation result to obtain sample data, and returning and adjusting the cloud disk storage strategy according to the sample data to form a feedback loop required by a closed-loop feedback mechanism, wherein the feedback loop can be shown as a dotted line in fig. 1 c.
The cloud disk storage strategy is adjusted through a closed-loop feedback mechanism, so that the cloud disk storage strategy dynamically changes according to the change of the corresponding read-write request, and particularly when the storage access mode of a service end is diversified, so that the rules of the read-write request received by the cloud disk are diversified, the cloud disk storage strategy is controlled to be dynamically adaptive to the read-write requests of a plurality of service ends, the scheduling efficiency of cloud disk storage resources can be improved, the generation of long tail effect can be reduced, and the performance of the cloud disk is improved.
In the solution provided in this embodiment, a corresponding cloud disk storage policy is determined according to a received read-write request, and a cloud disk operation corresponding to the read-write request is executed according to the determined cloud disk storage policy; and acquiring sample data influencing the read-write performance of the service end according to the operation result of the cloud disk, and adjusting the cloud disk storage strategy through the sample data, so that the cloud disk storage strategy can be adjusted based on a closed-loop feedback mechanism, the cloud disk storage strategy is dynamically changed according to the change of the corresponding read-write request, namely the cloud disk storage strategy is dynamically adapted to the corresponding read-write request, and especially when the storage access mode of the service end is diversified and the change rule of the read-write request received by the cloud disk is diversified, the scheduling efficiency of cloud disk storage resources can be improved through the cloud disk storage strategy dynamically adapted to the read-write request of the service end, the generation of long tail effect can be reduced, and the performance of cloud disk storage is improved.
The request processing method of the present embodiment may be executed by any suitable electronic device having data processing capability, including but not limited to: a server, etc.
Example two
Referring to fig. 2, a flowchart illustrating steps of a request processing method according to a second embodiment of the present invention is shown.
The request processing method of the embodiment comprises the following steps:
s202, determining a corresponding cloud disk storage strategy for the read-write request from the service end.
In this embodiment, the cloud disk storage policy may include a cloud disk write policy S1 and/or a storage deletion policy S2. The embodiment is exemplified by including a cloud disk write policy S1 and a storage deletion policy S2; in a specific implementation, the cloud disk storage policy may only include the cloud disk write policy S1 or only include the storage deletion policy S2, which is not limited in this embodiment.
In this embodiment, each service end may correspond to one cloud disk write strategy S1, and after receiving the read-write request, the cloud disk write strategy S1 corresponding to the read-write request is determined according to the service end that sent the read-write request.
In this embodiment, part or all of the service terminals may share one storage deletion policy S2. If part of the service ends share one storage deletion strategy S2, after receiving the read-write request, determining a corresponding storage deletion strategy according to the service end sending the read-write request S2; if all the service terminals share one storage deletion policy S2, the storage deletion policy S2 may be directly set as a default execution policy of the cloud disk storage.
And S204, determining whether to write the data requested by the read-write request into the cloud disk storage according to a cloud disk write strategy in the cloud disk storage strategy.
If so, executing step S206; and if the data is not written in, directly rejecting the read-write request, and ending the read-write request processing flow, wherein the operation result returned to the service end by the cloud disk can be processing failure.
In this embodiment, the cloud disk write policy S1 is used to evaluate whether to write the data requested by the read/write request to the cloud disk storage.
Specifically, if the read-write request is an inquiry request, after receiving the inquiry request, the cloud disk may first determine whether the inquiry request has hit data in the cloud disk storage; if yes, directly returning the hit data serving as an operation result to the service end; if not, determining whether to write the data requested by the query request to the cloud disk storage through the cloud disk write policy S1; if the data is determined to be written, writing the data requested by the query request into the cloud disk storage through steps S206-S210, and returning the written data serving as an operation result to the service end through the cloud disk storage; and if the data is determined not to be written, directly refusing to process the query request, and returning the processing failure as an operation result to the service end.
In this embodiment, since the cloud disk write policies S1 may correspond to the service end one to one, each cloud disk write policy S1 may be dynamically adapted to the storage access mode of the service end, so that all the cloud disk storage policies in the cloud disk may be compatible with a plurality of different storage access modes at the same time, and different storage access modes may be applicable to different data processing scenarios.
Specifically, in this embodiment, step S202 specifically includes: and determining whether to write the data requested by the read-write request into the cloud disk storage according to the writing probability in the cloud disk writing strategy S1 and the random generated number.
In this embodiment, the cloud disk write strategy S1 may correspond to the service end one to one, so that it is determined whether to write the data requested by the read/write request into the cloud disk storage according to the write probability and the random generation number, and after receiving the read/write request from the service end, the data requested by the read/write request may be written into the cloud disk storage at random, and it may be ensured that, at a large data level, the proportion of the read/write request corresponding to the successful writing of the requested data into the cloud disk storage in all the read/write requests of the corresponding service end corresponds to the write probability. The set write probability may directly affect the read-write performance of the service end, for example, the higher the write probability is, the fewer the number of rejected read-write requests is, so that the higher the read-write performance of the service end is.
In this embodiment, after the write probability is adjusted each time, the adjusted write probability may be used as a write probability corresponding to a subsequent read-write request, and the purpose of adjusting the write probability is to make the read-write performance of the service end closer to the expected performance; if the writing probability is not adjusted, the value of the writing probability may be a preset initial value, the preset initial value may be determined by analyzing the cloud disk operation executed after the cloud disk historically receives the read-write request, or may be a randomly generated initial value, and a plurality of writing probabilities may share the same preset initial value; the random number generation may be generated by a preset random number generation algorithm or device.
And S206, judging whether the cloud disk storage is full.
If the cloud disk storage is full, executing step S208; if the cloud disk storage is not full, step S212 is executed.
In this embodiment, the cloud disk storage serves as an intermediate layer between the service end and the database, and may serve as a buffer of the database. The purpose of the read-write operation is to read data in the database or write data into the database; when data is read, the data is read into a cloud disk for storage, and then the data is returned to a service end through the cloud disk storage, so that the data reading is completed; when data is written in, the data is written in the cloud disk for storage, and then the data is written in the database through the cloud disk storage. For example, if the read-write request may specifically include (time, w/r, device, block), where time is read-write time, w/r is read or write, the read-write request only includes one of w and r, device is a cloud disk corresponding to the service end, and block is a storage address of the requested data in the cloud disk corresponding to the service end.
When a request arrives, if w/r is w (namely writing), determining that data needs to be written into the block, and when the request arrives, firstly writing the data into the cloud disk storage, and then writing the data into an address indicated by the block through the cloud disk storage; if the w/r is specifically r (namely reading), whether hit data exists in the cloud disk storage is determined, if not, the data in the address indicated by the block is written into the cloud disk storage, and then the data is returned to the service end through the cloud disk storage.
Therefore, after determining to write the data requested by the read-write request into the cloud disk storage, it is required to first determine whether there is a free location in the cloud disk storage where the requested data can be stored through step S206; if the cloud disk storage is full, it indicates that there is no free location for storing the requested data, step S208 is executed, and if the cloud disk storage is not full, it indicates that there is a free location for storing the requested data, step S212 is executed.
And S208, determining the data to be deleted stored in the cloud disk according to the storage deletion strategy.
In this embodiment, the storage deletion policy S2 is used to evaluate which data in the cloud disk storage can be preferentially deleted.
In this embodiment, when the read-write request is an inquiry request, as shown in fig. 3, it is determined whether there is data hit by the inquiry request in the cloud disk storage, that is, whether the inquiry request hits, in the process of processing the inquiry request. If so, returning the hit data as query data; if the data is not written into the cloud disk storage according to the cloud disk write strategy S1, the data to be deleted in the cloud disk storage may be determined according to the storage deletion strategy S2, and the query data and the data to be deleted are replaced, so that the query data is written into the cloud disk storage, and the query data is returned through the cloud disk storage. Through the storage deletion strategy S2, data with low hit probability in the cloud disk storage can be determined as data to be deleted, so that the hit rate of the query request in the cloud disk storage is improved, and the read-write performance of the service end is improved.
Optionally, in this embodiment, the storage deletion policy S2 includes a plurality of basic deletion policies, and step S208 specifically includes: and determining the data to be deleted in the cloud disk storage according to the plurality of basic deletion strategies and the weights corresponding to the basic deletion strategies. By setting a plurality of basic deletion strategies and mixing the plurality of basic deletion strategies through the weight, a new storage deletion strategy S2 can be obtained by adjusting the weight of the basic deletion strategy, so that the obtained storage deletion strategy S2 is more diversified, and the adjustment of the storage deletion strategy S2 is more flexible.
In this embodiment, the basic deletion policy may include two or more of the following: least frequently used algorithm (LFU), least recently used algorithm (LRU), adaptive cache replacement Algorithm (ARC), first-in-first-out algorithm (FIFO), and the like.
Specifically, the purpose of adjusting the weight of the basic deletion policy is to make the cloud disk performance better. In actual use, if the weight has been adjusted, the adjusted weight of the basic deletion policy may be used as the storage deletion policy S2 corresponding to the subsequent read-write request; if the weights have not been adjusted, the weights of the plurality of basic deletion policies may be preset initial values or randomly generated initial values. Multiple basic deletion policies may share a same initial value, e.g., each 100, etc. The weight adjustment method for the first time of adjusting the weight may be randomly generated.
And S210, replacing the determined data to be deleted by using the data requested by the read-write request.
And after the data to be deleted is determined, directly replacing the data to be deleted with the data requested by the read-write request, and writing the requested data into the cloud disk storage.
And S212, writing the data requested by the read-write request into a cloud disk storage.
In this embodiment, if the cloud disk storage is not full, it indicates that there is a free location in the cloud disk storage where the requested data of this time can be stored, and the data may be directly written into the free location in the cloud disk storage in step S212.
S214, collecting sample data influencing the read-write performance of the service end according to the operation result of the cloud disk, and adjusting the storage strategy of the cloud disk according to the sample data.
In this embodiment, the steps of replacing the data to be deleted with the requested data, writing the requested data into the spare location in the cloud disk storage, and the like all belong to cloud disk operations; in addition, when the read-write request is an inquiry request, whether hit data corresponding to the inquiry request exists in the cloud disk storage or not can also belong to cloud disk operation.
In this embodiment, the sample data acquired according to the result of the cloud disk operation may include: the response result of the read-write request, the response time of the read-write request, the hit rate of the query request in the cloud disk storage, and the like.
In this embodiment, when the cloud disk storage policy includes the write probability, adjusting the cloud disk write policy S1 according to the sample data includes: determining the read-write performance of the service end through the sample data, and comparing the read-write performance of the service end with the expected performance; if the read-write performance of the service end is lower than the expected performance, adjusting the cloud disk write strategy S1 through the sample data to increase the write probability corresponding to the read-write request; or if the read-write performance of the service end is higher than the expected performance, the cloud disk write strategy S1 is adjusted through the sample data to reduce the write probability corresponding to the read-write request, so that the actual read-write performance of the service end is as close as possible to the expected read-write performance, and the utilization rate of cloud disk resources can be maximized by controlling the expected performance of all service ends of the cloud disk.
In the embodiment, the writing probability can be more adaptive to the read-write request of the corresponding service end by adjusting the value of the writing probability; and when the storage access mode of the service end changes, the read-write request sent by the service end also changes, if the cloud disk operation corresponding to the read-write request is continuously executed according to the original write probability, the obtained deviation between the read-write performance and the expected performance of the service end changes, and at the moment, the write probability can be continuously adjusted, so that the write probability changes along with the change of the storage access mode of the service end.
Specifically, since the collected sample data is data that affects the read-write performance of the service end, the read-write performance of the service end can be determined through the sample data, for example, the read-write performance of the service end is determined according to the number of rejected read-write requests, the response time of the read-write requests, and the like in the sample data.
In this embodiment, the expected performance of the service end may be determined according to a service end performance expectation standard preset in the cloud disk. For example, if the preset service end performance expectation criterion is load balancing of multiple service ends of the cloud disk, the performance criterion during the load balancing of the service ends may be directly determined as the expected performance of the service ends, and the cloud disk write strategy S1 may be adjusted according to the performance criterion.
In this embodiment, the writing probability is positively correlated with the read-write performance of the service end, and by increasing the writing probability, the read-write performance of the service end can be improved, whereas the read-write performance of the service end can be reduced.
However, other factors besides the writing probability may affect the read-write performance, for example, if the query request in the read-write request of the service end occupies a large proportion, the hit rate of the query request in the cloud disk storage may also affect the read-write performance of the service end, and the higher the hit rate is, the better the read-write performance is. Therefore, in actual use, it cannot be guaranteed that the read-write performance of the service end approaches the expected performance after each adjustment, and the read-write performance may possibly be far away from the expected performance.
In addition, when the cloud disk storage policy includes a storage deletion policy S2, said adjusting the cloud disk storage policy by the sample data further includes: determining current cloud disk parameter data influencing the read-write performance of the service end according to the sample data; determining the performance change trend of the cloud disk according to the current cloud disk parameter data and the historical cloud disk parameter data; if the trend indicates that the performance of the cloud disk is improved, the weights corresponding to the basic deletion strategies are adjusted again by using a previous weight adjustment mode; if the trend indicates that the performance of the cloud disk is reduced, adjusting the weights corresponding to the basic deletion strategies in a way different from the previous weight adjustment way so as to maximize the performance of the cloud disk.
In this embodiment, the determining, according to the sample data, the current cloud disk parameter data may include: hit rate of current cloud disk storage, data volume deleted by the cloud disk in the current stage, data volume written into the cloud disk again after deletion in the current stage and the like; correspondingly, the determined change trend of the cloud disk performance can comprise that the hit rate of the cloud disk storage is high or low and the like, and the hit rate of the cloud disk storage is high, so that when the read-write request is an inquiry request, the probability that the data requested by the inquiry request is directly hit in the cloud disk storage is high, the data volume needing to be written into the cloud disk storage is reduced, and the read-write performance of the service end of the cloud disk is improved integrally; in addition, compared with the method that one storage deletion policy is selected from the basic deletion policies and used as the cloud disk fixation, in the embodiment, the storage deletion policy most suitable for the cloud disk can be obtained by adjusting the weights of the plurality of basic deletion policies, so that the performance of the cloud disk is greatly improved.
In this embodiment, when the performance of the cloud disk is improved, it is stated that the previous weight adjustment mode contributes to the improvement of the performance of the cloud disk, and the previous weight adjustment mode may continue to be used, that is, the previous weight adjustment mode continues to be used to adjust the weights corresponding to the multiple basic deletion policies again; otherwise, the weights corresponding to the multiple basic deletion strategies need to be adjusted in a manner different from the previous weight adjustment manner. Assuming that three basic elimination strategies are included, the weight adjustment manner may be, for example: the weight of the first basic elimination strategy is increased by 0.1, the weight of the second basic elimination strategy is increased by 0.2, and the weight of the third basic elimination strategy is reduced by 0.3.
Specifically, when determining a manner different from the previous weight adjustment manner, a new manner different from the previous weight adjustment manner may be randomly generated, or may be modified based on the previous weight adjustment manner to obtain a new manner, which is not limited in this embodiment.
In addition, according to the above, part or all of the service terminals may correspond to one storage deletion policy. If a part of the service ends correspond to a storage deletion strategy, the cloud disk comprises a plurality of storage deletion strategies S2, and when adjustment is performed, the storage deletion strategies S2 corresponding to the part of the service ends can be adjusted according to sample data of the part of the service ends; if all the service terminals correspond to one storage deletion policy, the storage deletion policy S2 may be adjusted according to all the sample data.
Optionally, when the service end corresponds to one cloud disk storage policy, step S214 may further include: and respectively collecting sample data influencing the read-write performance of a plurality of service ends according to the operation result of the cloud disk, and adjusting the cloud disk storage strategies corresponding to the service ends according to the sample data and the available cloud disk resource information. By adjusting the cloud disk resource strategy according to the sample data corresponding to the plurality of service terminals, the available cloud disk resource information and the like, the adjusted cloud disk resource strategy can be more adaptive to the cloud disk and the plurality of service terminals.
In this embodiment, the available cloud disk resource information may include an available cloud disk storage size and the like. During specific adjustment, cloud disk resources can be distributed to a plurality of service ends according to the available cloud disk resource information and sample data of a plurality of service ends, and the expected performance of the service ends and the expected performance of the cloud disks are determined according to the distribution results, so that the cloud disk storage strategies corresponding to the service ends are adjusted according to the expected performance of the service ends and the cloud disks.
Optionally, in this embodiment, the cloud disk includes multiple stages, and correspondingly, each stage of storage corresponds to one group of the cloud disk storage policies, and one group of the cloud disk storage policies includes a cloud disk write policy and/or a storage delete policy.
Specifically, the multi-level storage can be arranged in parallel, also can be arranged in series, or arranged in a series-parallel mixed mode; when the multi-level storage is arranged in parallel, the first-level storage corresponding to the read-write request can be directly determined from the multi-level storage, and the cloud disk storage strategy corresponding to the read-write request is determined from a group of cloud disk storage strategies corresponding to the first-level storage; when the multi-level storage serial is set, the read-write request can be executed to complete the cloud disk operation only through the cloud disk storage of the multi-level serial setting, and for each level of storage through which the read-write request passes, the cloud disk storage strategy corresponding to the read-write request can be determined from a group of cloud disk storage strategies corresponding to the level of storage. Compared with the same group of cloud disk storage strategies shared by the multiple levels of cloud disks, the cloud disk storage strategy which can be adjusted adaptively is set for each level of storage, so that the utilization rate of the cloud disk storage can be improved, and the performance of the cloud disk is further improved.
Further, when the cloud disk includes multiple stages, determining a corresponding cloud disk storage policy for the read-write request from the service end, and executing the cloud disk operation corresponding to the read-write request according to the determined cloud disk storage policy includes: determining a storage grade corresponding to the read-write request of the service end, and determining a corresponding cloud disk storage strategy for the read-write request from a group of cloud disk storage strategies corresponding to the grade storage; determining whether to write the data requested by the read-write request into the level of storage according to the determined cloud disk storage strategy; and if the writing is determined and the storage of the level is full, determining the data to be deleted from the storage of the level according to the determined cloud disk storage strategy, and replacing the determined data to be deleted by using the data requested by the reading and writing request.
Determining whether to write the data requested by the read-write request into the level of storage according to the determined cloud disk storage strategy, so that the performance of a service end where the read-write request is located can be controlled by adjusting the cloud disk storage strategy; determining data to be deleted from the storage according to the determined cloud disk storage strategy, and adjusting the cloud disk storage strategy to adjust the data to be deleted so as to improve the hit rate of cloud disk storage; by combining the two, the performance of multi-level storage can be improved, the multi-level storage can be adapted to a plurality of service ends, and the overall performance of the cloud disk is further improved.
According to the scheme provided by the embodiment, the value of the writing probability is adjusted, so that the writing probability is more adaptive to the read-write request of the corresponding service end, all the cloud disk storage strategies in the cloud disk can be compatible with different read-write behavior modes of a plurality of service ends at the same time, and the use scene of the cloud disk is expanded; when the read-write request of the service end changes, the cloud disk operation corresponding to the read-write request is continuously executed according to the write-in probability, the obtained deviation between the read-write performance of the service end and the expected performance changes, and at the moment, the write-in probability can be continuously adjusted so that the write-in probability changes along with the change of the read-write request of the service end; in addition, the multiple basic deletion strategies are mixed through the weight, so that the mixed storage deletion strategy is more diversified, the adjustment of the storage deletion strategy is more flexible, and the storage deletion strategy most suitable for the cloud disk can be obtained through adjusting the weight of the multiple basic deletion strategies, so that the performance of the cloud disk is improved to the maximum extent.
The request processing method of the present embodiment may be executed by any suitable electronic device having data processing capability, including but not limited to: a server, etc.
EXAMPLE III
Referring to fig. 4, a block diagram of a request processing apparatus according to a third embodiment of the present invention is shown.
As shown in fig. 4, the request processing apparatus according to the embodiment of the present invention includes: a determination module 302 and an adjustment module 304.
The determining module 302 is configured to determine a corresponding cloud disk storage policy according to the received read-write request, and execute a cloud disk operation corresponding to the read-write request according to the determined cloud disk storage policy.
And the adjusting module 304 is configured to collect sample data that affects the read-write performance of the service end according to the result of the cloud disk operation, and adjust the cloud disk storage policy according to the sample data.
In an optional implementation manner, the cloud disk storage policy includes a cloud disk write policy, and correspondingly, the determining module 302 includes: and the writing determining module is used for determining whether to write the data requested by the reading and writing request into the cloud disk storage according to the cloud disk writing strategy.
In an alternative embodiment, the write determination module includes: and the random determining module is used for determining whether to write the data requested by the read-write request into the cloud disk storage according to the write probability in the cloud disk write strategy and the random generated number.
In an alternative embodiment, the adjusting module 304 includes: the comparison module is used for determining the read-write performance of the service end according to the sample data and comparing the read-write performance of the service end with the expected performance; the device further comprises: an adding module or a reducing module, wherein the adding module is used for adjusting the cloud disk write strategy through the sample data to increase the write probability corresponding to the read-write request if the read-write performance of the service end is lower than the expected performance; and the reducing module is used for adjusting the cloud disk write strategy through the sample data to reduce the write probability corresponding to the read-write request if the read-write performance of the service end is higher than the expected performance.
In an optional implementation manner, the cloud disk storage policy includes a storage deletion policy, and correspondingly, the determining module 302 includes: the data to be deleted determining module is used for determining the data to be deleted stored in the cloud disk according to the storage deletion strategy; and the replacing module is used for replacing the determined data to be deleted by using the data requested by the read-write request.
In an optional implementation manner, the storage deletion policy includes a plurality of basic deletion policies, and correspondingly, the to-be-deleted data determining module is specifically configured to: and determining the data to be deleted stored in the cloud disk according to the plurality of basic deletion strategies and the weights corresponding to the basic deletion strategies.
In an alternative embodiment, the adjusting module 304 includes: the current parameter determining module is used for determining current cloud disk parameter data influencing the read-write performance of the service end according to the sample data; the trend determining module is used for determining the cloud disk performance change trend according to the current cloud disk parameter data and the historical cloud disk parameter data; the weight adjusting module is used for adjusting the weights corresponding to the basic deletion strategies again by using a previous weight adjusting mode if the trend indicates that the performance of the cloud disk is improved; or if the trend indicates that the performance of the cloud disk is reduced, adjusting the weights corresponding to the basic deletion strategies in a way different from the previous weight adjustment way.
In an optional implementation manner, the cloud disk includes multiple levels of storage, and correspondingly, each level of storage corresponds to one set of the cloud disk storage policies, and one set of the cloud disk storage policies includes a cloud disk write policy and/or a storage delete policy.
In an alternative embodiment, the determining module 302 includes: the system comprises a stage number determining module, a read-write module and a storage module, wherein the stage number determining module is used for determining the storage stage number corresponding to the read-write request and determining a corresponding cloud disk storage strategy for the read-write request from a group of cloud disk storage strategies corresponding to the stage storage; the writing determining module is used for determining whether to write the data requested by the reading and writing request into the level storage according to a cloud disk reading strategy in the determined cloud disk storage strategies; if the writing is determined and the storage of the level is full, the device further comprises a to-be-deleted data determining module, which is used for determining to-be-deleted data from the storage of the level according to a storage elimination strategy in the determined cloud disk storage strategies; and the replacing module is used for replacing the determined data to be deleted by using the data requested by the read-write request.
In an optional implementation manner, the adjusting module 304 is specifically configured to: and respectively collecting sample data influencing the read-write performance of a plurality of service ends according to the operation result of the cloud disk, and adjusting the cloud disk storage strategies corresponding to the service ends according to the sample data and the available cloud disk resource information.
In the request processing scheme provided by this embodiment, a corresponding cloud disk storage policy is determined according to a received read-write request, and a cloud disk operation corresponding to the read-write request is executed according to the determined cloud disk storage policy; and acquiring sample data influencing the read-write performance of the service end according to the operation result of the cloud disk, and adjusting the cloud disk storage strategy through the sample data, so that the cloud disk storage strategy can be adjusted based on a closed-loop feedback mechanism, the cloud disk storage strategy is dynamically changed according to the change of the corresponding read-write request, namely the cloud disk storage strategy is dynamically adapted to the corresponding read-write request, and especially when the change rule of the read-write request received by the cloud disk is diversified due to the diversification of the storage access mode of the service end, the scheduling efficiency of cloud disk storage resources can be improved through the cloud disk storage strategy dynamically adapted to the read-write request of the service end, the generation of the long tail effect can be further reduced, and the cloud disk storage performance is improved.
Example four
An electronic device, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus; the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the request processing method.
Specifically, referring to fig. 5, a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention is shown, and the specific embodiment of the present invention does not limit the specific implementation of the electronic device.
As shown in fig. 5, the electronic device may include: a processor (processor)402, a Communications Interface 404, a memory 406, and a Communications bus 408.
Wherein:
the processor 402, communication interface 404, and memory 406 communicate with each other via a communication bus 408.
A communication interface 404 for communicating with other electronic devices or servers.
The processor 402 is configured to execute the program 410, and may specifically perform relevant steps in the foregoing request processing method embodiment.
In particular, program 410 may include program code comprising computer operating instructions.
Processor 402 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement an embodiment of the present invention; the processor may also be a Programmable Gate Array (FPGA), a graphics processor GPU, an embedded neural network processor NPU, or the like. The electronic device comprises one or more processors, which can be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.
And a memory 406 for storing a program 410. Memory 406 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 410 may specifically be configured to cause the processor 402 to perform the following operations: determining a corresponding cloud disk storage strategy according to the received read-write request, and executing cloud disk operation corresponding to the read-write request according to the determined cloud disk storage strategy; and collecting sample data influencing the read-write performance of a service end according to the operation result of the cloud disk, and adjusting the storage strategy of the cloud disk according to the sample data.
In an optional implementation manner, the cloud disk storage policy includes a cloud disk write policy, and correspondingly, the executing, according to the determined cloud disk storage policy, a cloud disk operation corresponding to the read-write request includes: and determining whether to write the data requested by the read-write request into the cloud disk storage according to the cloud disk write strategy.
In an optional implementation manner, the determining whether to write the data requested by the read-write request to a cloud disk storage according to the cloud disk write policy includes: and determining whether to write the data requested by the read-write request into the cloud disk storage according to the write probability in the cloud disk write strategy and the random generated number.
In an optional embodiment, the adjusting the cloud disk storage policy by the sample data includes: determining the read-write performance of the service end through the sample data, and comparing the read-write performance of the service end with the expected performance; if the read-write performance of the service end is lower than the expected performance, the cloud disk write strategy is adjusted through the sample data to increase the write probability corresponding to the read-write request; or if the read-write performance of the service end is higher than the expected performance, the cloud disk write strategy is adjusted through the sample data to reduce the write probability corresponding to the read-write request.
In an optional implementation manner, the cloud disk storage policy includes a storage deletion policy, and correspondingly, the executing, according to the determined cloud disk storage policy, the cloud disk operation corresponding to the read-write request further includes: determining data to be deleted stored in the cloud disk according to the storage deletion strategy; and replacing the determined data to be deleted by using the data requested by the read-write request.
In an optional implementation manner, the storage deletion policy includes a plurality of basic deletion policies, and correspondingly, the determining, according to the storage deletion policy, data to be deleted stored in the cloud disk includes: and determining the data to be deleted stored in the cloud disk according to the plurality of basic deletion strategies and the weights corresponding to the basic deletion strategies.
In an optional implementation manner, the adjusting the cloud disk storage policy by the sample data further includes: determining current cloud disk parameter data influencing the read-write performance of the service end according to the sample data; determining the performance change trend of the cloud disk according to the current cloud disk parameter data and the historical cloud disk parameter data; if the trend indicates that the performance of the cloud disk is improved, the weights corresponding to the basic deletion strategies are adjusted again by using a previous weight adjustment mode; and if the trend indicates that the performance of the cloud disk is reduced, adjusting the weights corresponding to the basic deletion strategies in a way different from the previous weight adjustment way.
In an optional implementation manner, the cloud disk includes multiple levels of storage, and correspondingly, each level of storage corresponds to one set of the cloud disk storage policies, and one set of the cloud disk storage policies includes a cloud disk write policy and/or a storage delete policy.
In an optional implementation manner, the determining, according to the received read-write request, a corresponding cloud disk storage policy, and executing, according to the determined cloud disk storage policy, a cloud disk operation corresponding to the read-write request includes: determining the storage grade number corresponding to the read-write request, and determining a corresponding cloud disk storage strategy for the read-write request from a group of cloud disk storage strategies corresponding to the storage grade; determining whether to write the data requested by the read-write request into the level of storage according to a cloud disk read-in strategy in the determined cloud disk storage strategies; and if the writing is determined and the storage of the level is full, determining the data to be deleted from the storage of the level according to a storage elimination strategy in the determined cloud disk storage strategy, and replacing the determined data to be deleted by using the data requested by the reading and writing request.
In an optional implementation manner, the collecting, according to the result of the cloud disk operation, sample data that affects the read-write performance of a service end, and adjusting the cloud disk storage policy by using the sample data includes: and respectively collecting sample data influencing the read-write performance of a plurality of service ends according to the operation result of the cloud disk, and adjusting the cloud disk storage strategies corresponding to the service ends according to the sample data and the available cloud disk resource information.
For specific implementation of each step in the program 410, reference may be made to corresponding steps and corresponding descriptions in units in the foregoing request processing method embodiments, which are not described herein again. It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described devices and modules may refer to the corresponding process descriptions in the foregoing method embodiments, and are not described herein again.
In the electronic device of this embodiment, a corresponding cloud disk storage policy is determined according to a received read-write request, and a cloud disk operation corresponding to the read-write request is executed according to the determined cloud disk storage policy; and acquiring sample data influencing the read-write performance of the service end according to the operation result of the cloud disk, and adjusting the cloud disk storage strategy through the sample data, so that the cloud disk storage strategy can be adjusted based on a closed-loop feedback mechanism, the cloud disk storage strategy is dynamically changed according to the change of the corresponding read-write request, namely the cloud disk storage strategy is dynamically adapted to the corresponding read-write request, and especially when the change rule of the read-write request received by the cloud disk is diversified due to the diversification of the storage access mode of the service end, the scheduling efficiency of cloud disk storage resources can be improved through the cloud disk storage strategy dynamically adapted to the read-write request of the service end, the generation of the long tail effect can be further reduced, and the cloud disk storage performance is improved.
It should be noted that, according to the implementation requirement, each component/step described in the embodiment of the present invention may be divided into more components/steps, and two or more components/steps or partial operations of the components/steps may also be combined into a new component/step to achieve the purpose of the embodiment of the present invention.
The above-described method according to an embodiment of the present invention may be implemented in hardware, firmware, or as software or computer code storable in a recording medium such as a CD ROM, a RAM, a floppy disk, a hard disk, or a magneto-optical disk, or as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium downloaded through a network and to be stored in a local recording medium, so that the method described herein may be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware such as an ASIC or FPGA. It will be appreciated that a computer, processor, microprocessor controller, or programmable hardware includes memory components (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code that, when accessed and executed by a computer, processor, or hardware, implements the request processing methods described herein. Further, when a general-purpose computer accesses code for implementing the request processing method illustrated herein, execution of the code transforms the general-purpose computer into a special-purpose computer for executing the request processing method illustrated herein.
Those of ordinary skill in the art will appreciate that the various illustrative elements and method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present embodiments.
The above embodiments are only for illustrating the embodiments of the present invention and not for limiting the embodiments of the present invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of the present invention, so that all equivalent technical solutions also belong to the scope of the embodiments of the present invention, and the scope of patent protection of the embodiments of the present invention should be defined by the claims.
Claims (13)
1. A method for processing a request, comprising:
determining a corresponding cloud disk storage strategy according to the received read-write request, and executing cloud disk operation corresponding to the read-write request according to the determined cloud disk storage strategy;
and collecting sample data influencing the read-write performance of a service end according to the operation result of the cloud disk, and adjusting the storage strategy of the cloud disk according to the sample data.
2. The method of claim 1, wherein the cloud disk storage policy comprises a cloud disk write policy, and correspondingly, the performing, according to the determined cloud disk storage policy, a cloud disk operation corresponding to the read-write request comprises:
and determining whether to write the data requested by the read-write request into the cloud disk storage according to the cloud disk write strategy.
3. The method of claim 2, wherein the determining whether to write the data requested by the read/write request to cloud disk storage according to the cloud disk write policy comprises:
and determining whether to write the data requested by the read-write request into the cloud disk storage according to the write probability in the cloud disk write strategy and the random generated number.
4. The method of claim 2, wherein said adjusting the cloud disk storage policy by the sample data comprises:
determining the read-write performance of the service end through the sample data, and comparing the read-write performance of the service end with the expected performance;
if the read-write performance of the service end is lower than the expected performance, the cloud disk write strategy is adjusted through the sample data to increase the write probability corresponding to the read-write request;
or if the read-write performance of the service end is higher than the expected performance, the cloud disk write strategy is adjusted through the sample data to reduce the write probability corresponding to the read-write request.
5. The method according to claim 1, wherein the cloud disk storage policy includes a storage deletion policy, and correspondingly, the performing, according to the determined cloud disk storage policy, a cloud disk operation corresponding to the read-write request further includes:
determining data to be deleted stored in the cloud disk according to the storage deletion strategy;
and replacing the determined data to be deleted by using the data requested by the read-write request.
6. The method according to claim 5, wherein the storage deletion policy includes a plurality of basic deletion policies, and correspondingly, the determining data to be deleted stored in the cloud disk according to the storage deletion policy includes:
and determining the data to be deleted stored in the cloud disk according to the plurality of basic deletion strategies and the weights corresponding to the basic deletion strategies.
7. The method of claim 6, wherein said adjusting the cloud disk storage policy with the sample data further comprises:
determining current cloud disk parameter data influencing the read-write performance of the service end according to the sample data;
determining the performance change trend of the cloud disk according to the current cloud disk parameter data and the historical cloud disk parameter data;
if the trend indicates that the performance of the cloud disk is improved, the weights corresponding to the basic deletion strategies are adjusted again by using a previous weight adjustment mode;
and if the trend indicates that the performance of the cloud disk is reduced, adjusting the weights corresponding to the basic deletion strategies in a way different from the previous weight adjustment way.
8. The method of claim 1, wherein the cloud disk comprises multiple levels of storage, and correspondingly, each level of storage corresponds to one set of the cloud disk storage policies, and one set of the cloud disk storage policies comprises a cloud disk write policy and/or a storage delete policy.
9. The method according to claim 8, wherein the determining a corresponding cloud disk storage policy according to the received read-write request, and the executing the cloud disk operation corresponding to the read-write request according to the determined cloud disk storage policy comprises:
determining the storage grade number corresponding to the read-write request, and determining a corresponding cloud disk storage strategy for the read-write request from a group of cloud disk storage strategies corresponding to the storage grade;
determining whether to write the data requested by the read-write request into the level of storage according to a cloud disk read-in strategy in the determined cloud disk storage strategies;
and if the writing is determined and the storage of the level is full, determining the data to be deleted from the storage of the level according to a storage elimination strategy in the determined cloud disk storage strategy, and replacing the determined data to be deleted by using the data requested by the reading and writing request.
10. The method according to claim 1, wherein the collecting sample data that affects read-write performance of a service end according to the result of the cloud disk operation, and the adjusting the cloud disk storage policy by the sample data includes:
and respectively collecting sample data influencing the read-write performance of a plurality of service ends according to the operation result of the cloud disk, and adjusting the cloud disk storage strategies corresponding to the service ends according to the sample data and the available cloud disk resource information.
11. A request processing apparatus, comprising:
the determining module is used for determining a corresponding cloud disk storage strategy according to the received read-write request and executing cloud disk operation corresponding to the read-write request according to the determined cloud disk storage strategy;
and the adjusting module is used for acquiring sample data influencing the read-write performance according to the operation result of the cloud disk and adjusting the storage strategy of the cloud disk according to the sample data.
12. An electronic device, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the operation corresponding to the request processing method according to any one of claims 1-10.
13. A computer storage medium having stored thereon a computer program which, when executed by a processor, implements a request processing method according to any one of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911242658.2A CN112925472A (en) | 2019-12-06 | 2019-12-06 | Request processing method and device, electronic equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911242658.2A CN112925472A (en) | 2019-12-06 | 2019-12-06 | Request processing method and device, electronic equipment and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112925472A true CN112925472A (en) | 2021-06-08 |
Family
ID=76161716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911242658.2A Pending CN112925472A (en) | 2019-12-06 | 2019-12-06 | Request processing method and device, electronic equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112925472A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114048106A (en) * | 2021-11-26 | 2022-02-15 | 北京志凌海纳科技有限公司 | Disk state detection method, system, medium and storage device |
| CN114546279A (en) * | 2022-02-24 | 2022-05-27 | 重庆紫光华山智安科技有限公司 | IO request prediction method and device, storage node and readable storage medium |
| CN118605809A (en) * | 2024-06-12 | 2024-09-06 | 山东爱特云翔信息技术有限公司 | A method and device for switching cloud disk mode |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101635841A (en) * | 2009-09-04 | 2010-01-27 | 杭州华三通信技术有限公司 | Method and equipment for adjusting read-write performance of video service information |
| CN102857560A (en) * | 2012-08-15 | 2013-01-02 | 华数传媒网络有限公司 | Multi-service application orientated cloud storage data distribution method |
| US20190042089A1 (en) * | 2018-03-02 | 2019-02-07 | Intel Corporation | Method of improved data distribution among storage devices |
| CN109491789A (en) * | 2018-11-02 | 2019-03-19 | 浪潮电子信息产业股份有限公司 | Distributed storage system service equalization processing method, device and equipment |
| WO2019149261A1 (en) * | 2018-02-01 | 2019-08-08 | 中兴通讯股份有限公司 | File storage method for distributed file system and distributed file system |
| CN110278229A (en) * | 2018-03-15 | 2019-09-24 | 阿里巴巴集团控股有限公司 | Load-balancing method, device and the electronic equipment of Distributed Services cluster |
| CN110389841A (en) * | 2019-07-25 | 2019-10-29 | 中南民族大学 | A kind of server load balancing method, apparatus and storage medium |
-
2019
- 2019-12-06 CN CN201911242658.2A patent/CN112925472A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101635841A (en) * | 2009-09-04 | 2010-01-27 | 杭州华三通信技术有限公司 | Method and equipment for adjusting read-write performance of video service information |
| CN102857560A (en) * | 2012-08-15 | 2013-01-02 | 华数传媒网络有限公司 | Multi-service application orientated cloud storage data distribution method |
| WO2019149261A1 (en) * | 2018-02-01 | 2019-08-08 | 中兴通讯股份有限公司 | File storage method for distributed file system and distributed file system |
| US20190042089A1 (en) * | 2018-03-02 | 2019-02-07 | Intel Corporation | Method of improved data distribution among storage devices |
| CN110278229A (en) * | 2018-03-15 | 2019-09-24 | 阿里巴巴集团控股有限公司 | Load-balancing method, device and the electronic equipment of Distributed Services cluster |
| CN109491789A (en) * | 2018-11-02 | 2019-03-19 | 浪潮电子信息产业股份有限公司 | Distributed storage system service equalization processing method, device and equipment |
| CN110389841A (en) * | 2019-07-25 | 2019-10-29 | 中南民族大学 | A kind of server load balancing method, apparatus and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 徐敏;李明;郑建忠;孙强;管建超;罗华永;张辉;: "基于OpenStack的Swift负载均衡算法", 计算机系统应用, no. 01 * |
| 王辉;李小亮;洪波;: "马尔科夫决策过程在移动端云存储策略中的应用", 工业仪表与自动化装置, no. 06 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114048106A (en) * | 2021-11-26 | 2022-02-15 | 北京志凌海纳科技有限公司 | Disk state detection method, system, medium and storage device |
| CN114546279A (en) * | 2022-02-24 | 2022-05-27 | 重庆紫光华山智安科技有限公司 | IO request prediction method and device, storage node and readable storage medium |
| CN114546279B (en) * | 2022-02-24 | 2023-11-14 | 重庆紫光华山智安科技有限公司 | IO request prediction method and device, storage node and readable storage medium |
| CN118605809A (en) * | 2024-06-12 | 2024-09-06 | 山东爱特云翔信息技术有限公司 | A method and device for switching cloud disk mode |
| CN118605809B (en) * | 2024-06-12 | 2025-05-16 | 山东爱特云翔信息技术有限公司 | Method and equipment for Yun Panmo type switching |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7583501B2 (en) | Data query method, apparatus and device | |
| US10621110B2 (en) | Deterministic multifactor cache replacement | |
| US12360706B2 (en) | Data processing method and system based on multi-level cache | |
| US10430338B2 (en) | Selectively reading data from cache and primary storage based on whether cache is overloaded | |
| CN116467353B (en) | A cache method and system based on LRU differentiation adaptive adjustment | |
| CN108495195A (en) | A kind of network direct broadcasting ranking list generation method, device, equipment and storage medium | |
| CN109981702B (en) | File storage method and system | |
| US20120303905A1 (en) | Method and apparatus for implementing cache | |
| CN112925472A (en) | Request processing method and device, electronic equipment and computer storage medium | |
| CN104866339A (en) | Distributed persistent management method, system and device of FOTA data | |
| CN116166181A (en) | Cloud monitoring method and cloud management platform | |
| CN110090436B (en) | H5 mini game resource caching method | |
| WO2023165543A1 (en) | Shared cache management method and apparatus, and storage medium | |
| CN109582233A (en) | A kind of caching method and device of data | |
| US20210097049A1 (en) | Method, device and computer program product for managing index tables | |
| CN113297106A (en) | Data replacement method based on hybrid storage, related method, device and system | |
| CN111666045A (en) | Data processing method, system, equipment and storage medium based on Git system | |
| CN118626171B (en) | A resource file loading method, device and system | |
| CN119397166A (en) | A data request processing method, device, equipment and storage medium | |
| WO2025039849A1 (en) | Online evolution method and apparatus for execution plans, device and storage medium | |
| CN118051449A (en) | Method and device for generating data cache queue, computer equipment and storage medium | |
| US10992743B1 (en) | Dynamic cache fleet management | |
| CN118210739A (en) | Multi-level cache management method, device and computer-readable storage medium | |
| CN116248699B (en) | Data reading method, device, equipment and storage medium in multi-copy scene | |
| CN113742304B (en) | Data storage method of hybrid cloud |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210608 |
|
| RJ01 | Rejection of invention patent application after publication |