CN109697018A - The method and apparatus for adjusting memory node copy amount - Google Patents
The method and apparatus for adjusting memory node copy amount Download PDFInfo
- Publication number
- CN109697018A CN109697018A CN201710982861.8A CN201710982861A CN109697018A CN 109697018 A CN109697018 A CN 109697018A CN 201710982861 A CN201710982861 A CN 201710982861A CN 109697018 A CN109697018 A CN 109697018A
- Authority
- CN
- China
- Prior art keywords
- prediction
- memory node
- amount
- access
- copy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F3/061—Improving I/O performance
- G06F3/0611—Improving I/O performance in relation to response time
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F3/0629—Configuration or reconfiguration of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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 invention discloses the method and apparatus of adjustment memory node copy amount, are related to field of computer technology.One specific embodiment of this method includes: to predict storage node accesses amount based on prediction model；The quantity of memory node copy is adjusted according to the storage node accesses amount of prediction；Wherein, original series are constructed according to historical data, the cumulative sequence of Accumulating generation is carried out to original series, differential equation of first order then is established to obtain prediction model to cumulative sequence.The embodiment can save memory space；Meanwhile the recovery for the memory node data that fail, need to only download a memory node based on complete copy redundancy can complete recovery process, reduce overhead.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of methods and dress for adjusting memory node copy amount
It sets.
Background technique
With the development of computer technology, the various service class systems such as e-commerce system, financial system, ticket-booking system are answered
It transports and gives birth to, there is data volumes for these systems greatly, the non-uniform feature of access, such as e-commerce system and financial system are at certain
A little big rush days or certain specific time amount of access can increase suddenly；Ticket-booking system amount of access of a period of time before and after holiday also can
Suddenly it increases.The data of these systems are stored in each memory node of distributed system, and distributed system refers to that foundation exists
Software systems on network, in a distributed system, the imperceptible data of user are distributions, i.e., user is not necessary to know relationship
Whether divide, which memory node is stored in whether there is or not copy, data and is executed on which memory node etc..If occurred simultaneously
A large amount of user access request (i.e. amount of access increases suddenly), can make the load pressure for storing the memory node of these system datas
Power becomes larger, and it is elongated in turn result in the response time.
Currently, the storage mode of distributed system has complete copy redundancy scheme and correcting and eleting codes technology redundancy scheme.Completely
Copy redundancy scheme, which refers to, creates multiple memory node copies for same memory node within the storage system to improve data resource
Availability, and all memory nodes are respectively provided with the copy of identical quantity, and each source data is stored in a memory node.
Correcting and eleting codes technology redundancy scheme is that f source data is encoded to n (n > f) a subdata, and n subdata is stored respectively in n
Memory node, so that f original data can be reconstructed with f coded data any in this n data.
In realizing process of the present invention, at least there are the following problems in the prior art for inventor's discovery:
1. for complete copy redundancy scheme, the quantity of memory node copy is difficult to determine, if the number of memory node copy
Amount setting is smaller, then accesses and load higher, throughput of system reduction；If the quantity setting of memory node copy is larger, consume
More memory space；
2. after memory node failure, the process for restoring data needs biggish system for correcting and eleting codes technology redundancy scheme
System expense, needs to decode initial data according to coded data.
Summary of the invention
In view of this, the embodiment of the present invention provides a kind of method and apparatus for adjusting memory node copy amount, Neng Goujie
Save memory space；Reduce overhead.
To achieve the above object, according to an aspect of an embodiment of the present invention, a kind of adjustment memory node copy is provided
The method of quantity.
A kind of method of adjustment memory node copy amount of the embodiment of the present invention includes: to predict to store based on prediction model
Node visit amount；The quantity of memory node copy is adjusted according to the storage node accesses amount of prediction；Wherein, according to history number
According to building original series, the cumulative sequence of Accumulating generation is carried out to the original series, single order then is established to the cumulative sequence
The differential equation is to obtain prediction model.
It optionally, include: that the history access data of memory node are defeated based on prediction model prediction storage node accesses amount
Enter the prediction model and obtains memory node prediction array；Inverse accumulated generating prediction is carried out to the memory node prediction array
The storage node accesses amount.
Optionally, adjusting the quantity of memory node copy according to the storage node accesses amount of prediction includes: based on institute
State prediction model forecasting system amount of access；Meter is accessed according to the storage node accesses amount of prediction and the system of prediction
It calculates memory node and predicts temperature；The quantity of the memory node copy is adjusted based on memory node prediction temperature.
It optionally, include: by the history access data input institute of system based on the prediction model forecasting system amount of access
It states prediction model and obtains system prediction array；The system for carrying out an inverse accumulated generating prediction to the system prediction array is visited
The amount of asking.
Optionally, the memory node prediction temperature determines according to the following formula:Wherein, HiIt indicates
The memory node in i-th of period predicts temperature, HjIndicate the averaged historical temperature of memory node, miIndicate i-th of period of prediction
Storage node accesses amount, NiIndicate the system amount of access in i-th of period of prediction, 0≤p≤1,0≤q≤1, and p+q=1.
Optionally, the method also includes: when user requests access to system, access path based on user and described deposit
The memory node copy for the response time Response to selection user request for storing up node copy.
To achieve the above object, according to another aspect of an embodiment of the present invention, a kind of adjustment memory node copy is provided
The device of quantity.
A kind of device of adjustment memory node copy amount of the embodiment of the present invention includes: prediction module, for based on pre-
Survey model prediction storage node accesses amount；Module is adjusted, for adjusting storage section according to the storage node accesses amount of prediction
The quantity of point copy；Wherein, original series are constructed according to historical data, the cumulative sequence of Accumulating generation is carried out to the original series
Column, then establish differential equation of first order to the cumulative sequence to obtain prediction model.
Optionally, the prediction module is also used to: the history access data of memory node being inputted the prediction model and are obtained
To memory node prediction array；The memory node for carrying out an inverse accumulated generating prediction to the memory node prediction array is visited
The amount of asking.
Optionally, the adjustment module is also used to: being based on the prediction model forecasting system amount of access；According to the institute of prediction
The system amount of access for stating storage node accesses amount and prediction calculates memory node prediction temperature；It is pre- based on the memory node
Calorimetric degree adjusts the quantity of the memory node copy.
Optionally, the adjustment module is further used for: the history access data of system being inputted the prediction model and are obtained
To system prediction array；The system amount of access of inverse accumulated generating prediction is carried out to the system prediction array.
Optionally, the memory node prediction temperature determines according to the following formula:Wherein, HiIt indicates
The memory node in i-th of period predicts temperature, HjIndicate the averaged historical temperature of memory node, miIndicate i-th of period of prediction
Storage node accesses amount, NiIndicate the system amount of access in i-th of period of prediction, 0≤p≤1,0≤q≤1, and p+q=1.
Optionally, described device further include: selecting module, for when user requests access to system, the visit based on user
Ask the way diameter and the memory node copy response time Response to selection user request memory node copy.
To achieve the above object, another aspect according to an embodiment of the present invention provides a kind of adjustment memory node copy
The electronic equipment of quantity.
A kind of electronic equipment of adjustment memory node copy amount of the embodiment of the present invention includes: one or more processing
Device；Storage device, for storing one or more programs, when one or more of programs are by one or more of processors
It executes, so that one or more of processors realize a kind of side of adjustment memory node copy amount of the embodiment of the present invention
Method.
To achieve the above object, according to an embodiment of the present invention in another aspect, providing a kind of computer-readable storage medium
Matter.
A kind of computer readable storage medium of the embodiment of the present invention is stored thereon with computer program, described program quilt
A kind of method of adjustment memory node copy amount of the embodiment of the present invention is realized when processor executes.
One embodiment in foregoing invention has the following advantages that or the utility model has the advantages that because using prediction model is based on come pre-
The amount of access of some memory node is surveyed, and adjusts according to the storage node accesses amount of prediction the skill of the copy amount of the memory node
Art means, thus the quantity for overcoming memory node copy be difficult to determine and memory node failure after, restore the process of data
The larger technical problem of middle overhead, and then reach and additions and deletions are carried out to memory node copy in advance, if increasing storage in advance
The quantity of node copy can be avoided the problem of memory node copy replication lags, effectively alleviate node load；If in advance
The quantity for reducing memory node copy, then can save memory space；Meanwhile the recovery for the memory node data that fail, base
A memory node, which need to only be downloaded, in complete copy redundancy can complete recovery process, reduce the technology of overhead
Effect.
Further effect possessed by above-mentioned non-usual optional way adds hereinafter in conjunction with specific embodiment
With explanation.
Detailed description of the invention
Attached drawing for a better understanding of the present invention, does not constitute an undue limitation on the present invention.Wherein:
Fig. 1 is the schematic diagram of the key step of the method for adjustment memory node copy amount according to an embodiment of the present invention；
Fig. 2 is the schematic diagram of the main modular of the device of adjustment memory node copy amount according to an embodiment of the present invention；
Fig. 3 is the development trend schematic diagram of initial data；
Fig. 4 is the development trend schematic diagram for the data sequence that accumulation operations generate；
Fig. 5 is that the embodiment of the present invention can be applied to exemplary system architecture figure therein；
Fig. 6 is adapted for the structural representation of the computer system for the terminal device or server of realizing the embodiment of the present invention
Figure.
Specific embodiment
Below in conjunction with attached drawing, an exemplary embodiment of the present invention will be described, including the various of the embodiment of the present invention
Details should think them only exemplary to help understanding.Therefore, those of ordinary skill in the art should recognize
It arrives, it can be with various changes and modifications are made to the embodiments described herein, without departing from scope and spirit of the present invention.Together
Sample, for clarity and conciseness, descriptions of well-known functions and structures are omitted from the following description.
Fig. 1 is the key step schematic diagram of the method for adjustment memory node copy amount according to an embodiment of the present invention.
As shown in Figure 1, a kind of method of adjustment memory node copy amount of the embodiment of the present invention mainly includes following step
It is rapid:
Step S101: storage node accesses amount is predicted based on prediction model.
Since that there is data volumes is big, access is uneven for the service class system such as e-commerce system, financial system, ticket-booking system
Even feature can make the load for storing the memory node of these system datas if the amount of access of these systems increases suddenly
Pressure becomes larger, and it is elongated in turn result in the response time.
Currently, the mode that distributed system stores these system datas has complete copy redundancy scheme and correcting and eleting codes technology superfluous
Remaining mechanism, still, for complete copy redundancy scheme, the quantity of memory node copy is difficult to determine in distributed system；For
Correcting and eleting codes technology redundancy scheme, after memory node failure, the process for restoring data needs biggish overhead, needs basis
Coded data decodes initial data.Wherein, in computer network architecture, other than useful data, there are also many controls
Information, such as committed memory, occupancy CPU etc., these control information are referred to as overhead.
The amount of access of usual system has the characteristics that cyclically-varying, likewise, in system each memory node access
Amount be also it is periodically variable, be based on this, the embodiment of the present invention proposes that prediction model is based on using historical data to be predicted each to deposit
Node visit amount is stored up, and adjusts the number of copies of each memory node according to the amount of access of each memory node of prediction, that is, is improved
The load capacity of memory node, and memory space has been saved, and solve the problems, such as the copy adjustment lag of memory node.
In embodiments of the present invention, original series are constructed according to historical data, it is cumulative to carry out Accumulating generation to original series
Then sequence establishes differential equation of first order to cumulative sequence to obtain prediction model.By the way that discrete data are carried out cumulative place
Reason, to show the regularity of discrete data, due to data information can sequence variation at any time, but total trend have it is potential
Rule, discrete irregular data can be changed into regular data sequence according to potential rule, therefore, can be with
Discrete data are considered as continuous variable discrete value acquired during variation.It is i.e. each based on the prediction of partial history data
The realization of the amount of access of the amount of access and system of a memory node is that discrete data are considered as to continuous variable in the process of variation
In acquired discrete value, and data are handled using the differential equation, partial history data are added up to obtain formation sequence,
Formation sequence is constructed into differential equation of first order, to construct prediction model, part random error can be offset in this way, show from
Dissipate the regularity of data.
Predict that amount of access needs complete historical data generally according to historical data, but in practical applications, history number
It is low so as to cause the accuracy of prediction according to often and imperfect.The prediction model that the embodiment of the present invention proposes can be based on part
Historical data predicts the amount of access of each memory node and the amount of access of system, so that in the incomplete situation of historical data,
Still it is capable of the variation tendency of accurate prediction data, thus the amount of access of the amount of access of each memory node and system.
It in embodiments of the present invention, include: by the history of memory node based on prediction model prediction storage node accesses amount
Access data input prediction model obtains memory node prediction array；It is pre- that inverse accumulated generating is carried out to memory node prediction array
The storage node accesses amount of survey.
When predicting the amount of access of some memory node, the history of the memory node can be accessed into data and be input in advance
In the prediction model of foundation, memory node prediction array can be obtained by the prediction model, deposited by what prediction model obtained
Storing up node prediction array is one-accumulate amount, therefore, it is also desirable to which carrying out a regressive to the memory node prediction array can give birth to
At the storage node accesses amount of prediction.
Step S102: the quantity of the storage node accesses amount adjustment memory node copy based on prediction.
In order to efficiently utilize each memory node, each storage that the embodiment of the present invention is predicted based on step S101
Node visit amount adjusts the quantity of each memory node copy, to alleviate each memory node load, meets requirements for access, together
When solve the problems, such as duplication lag.It should be noted that how to adjust can be according to storage for the quantity of each memory node copy
The capacity of node, actual use situation or demand etc. determine.
In addition, the storage mode of data has complete copy redundancy scheme and correcting and eleting codes technology redundancy scheme etc..In order to guarantee
Reliability, the availability of data can determine storage mode, the quantity of memory node copy according to the quantity of memory node copy
When greater than preset value, storage mode can choose complete copy redundancy scheme, and the quantity of memory node copy is not more than preset value
When, storage mode can choose correcting and eleting codes technology redundancy scheme.Such as the quantity of memory node copy be three when, storage mode
It can choose correcting and eleting codes technology redundancy scheme.
Since memory node each in system has different amount of access, each memory node has different deposit
Node temperature is stored up, for memory node load balancing each in guarantee system, saves memory space, it can be according further to each
The memory node temperature of a memory node determines the quantity of the memory node copy of each memory node.In the embodiment of the present invention
In, the quantity of the storage node accesses amount adjustment memory node copy based on prediction includes: to be visited based on prediction model forecasting system
The amount of asking；Memory node, which is calculated, according to the storage node accesses amount of prediction and the system amount of access of prediction predicts temperature；Based on storage
Node predicts the quantity of temperature adjustment memory node copy.
The storage node accesses amount of prediction can embody the potential access amount of memory node, but not embody memory node
Potential access amount ratio in the entire system memory node can also be further therefore predicted in the embodiment of the present invention
Predict temperature, the quantity based on memory node prediction temperature adjustment memory node copy.Since memory node temperature and storage save
Point amount of access is related to system amount of access, and the system amount of access in next period can be predicted by the history amount of access of system,
To calculate memory node prediction temperature according to the storage node accesses amount of prediction and the system amount of access of prediction.Utilize history
Data predict following storage node accesses amount and system amount of access, and the storage node accesses amount based on prediction and pre-
The system amount of access of survey can predict the higher data resource of potential temperature, the i.e. higher memory node of temperature in advance, to mention
The preceding duplication for carrying out data to the higher memory node of amount of access stores, and the access load of memory node is effectively reduced；Likewise,
The system amount of access of storage node accesses amount and prediction based on prediction can predict the lower data resource of potential temperature in advance,
That is the lower memory node of temperature carries out the deletion of memory node copy to the lower memory node of temperature in advance, to reduce system
Storage overhead.
In addition, being lower than the memory node of certain numerical value, for the quantity of memory node copy in order to guarantee the reliable of data
Property, availability, can using correcting and eleting codes technology come to data resource carry out code storage.
Predict some system amount of access when, can by the history of the system access data be input to pre-establish it is pre-
It surveys in model, system prediction array can be obtained by the prediction model, the system prediction array obtained by prediction model is
One-accumulate amount, therefore, it is also desirable to which carrying out a regressive to the system prediction array is to produce the system amount of access of prediction.?
It include: that the history of system is accessed into data input prediction mould based on prediction model forecasting system amount of access in the embodiment of the present invention
Type obtains system prediction array；The system amount of access of inverse accumulated generating prediction is carried out to system prediction array.
In embodiments of the present invention, memory node prediction temperature can determine according to the following formula:
Wherein, HiIndicate the memory node prediction temperature in i-th of period, HjIndicate the averaged historical temperature of memory node, miIndicate prediction
I-th of period storage node accesses amount, NiThe system amount of access in i-th of period of expression prediction, 0≤p≤1,0≤q≤
1, and p+q=1.
It should be noted that the value of parameter a and b can according to need in above-mentioned formula or actual conditions adjust, Hi-1It indicates
The history temperature of memory node, memory node history temperature are bigger to this predicted impact, then the value of parameter a is bigger；Storage section
Point history temperature is smaller to this predicted impact, then the value of parameter a is smaller.
In order to improve resource utilization, and the usage experience of user is improved, is each storage in system based on above-mentioned steps
Node is arranged after a certain number of memory node copies, when user accesses system, can according to the access path of user from
The shortest memory node copy of several access path is selected in memory node copy as candidate, then to candidate storage section
The response time of point copy predicted, the Response to selection time shortest copy memory node from candidate memory node copy
Respond user's request.Wherein, access path refers to the path by station address to memory node copy address；Response time it is pre-
Survey is what the storage node accesses amount based on prediction carried out, different for the memory node of the memory node copy of identical quantity
Storage node accesses amount correspond to the different response times, so, in the storage node accesses amount predicted and determine storage
After the quantity of node copy, the corresponding response time can be obtained.In embodiments of the present invention, when user requests access to system
When, the memory node copy of the response time Response to selection user request of access path and memory node copy based on user.
The method of adjustment memory node copy amount according to an embodiment of the present invention can be seen that because using based on prediction
Model predicts the amount of access of some memory node, and adjusts the copy of the memory node according to the storage node accesses amount of prediction
The technological means of quantity, thus the quantity for overcoming memory node copy be difficult to determine and memory node failure after, restore number
The larger technical problem of overhead during, and then reach and additions and deletions are carried out to memory node copy in advance, if in advance
The quantity for increasing memory node copy can be avoided the problem of memory node copy replication lags, effectively alleviate node load；
If reducing the quantity of memory node copy in advance, memory space can be saved；Meanwhile for failure memory node data
Restore, need to only download a memory node based on complete copy redundancy can complete recovery process, and the system of reducing is opened
The technical effect of pin.
Fig. 2 is the schematic diagram of the main modular of the device of adjustment memory node copy amount according to an embodiment of the present invention.
As shown in Fig. 2, a kind of device 200 of adjustment memory node copy amount of the embodiment of the present invention specifically include that it is pre-
Survey module 201 and adjustment module 202.
Wherein,
Prediction module 201, for predicting storage node accesses amount based on prediction model；
Module 202 is adjusted, for adjusting the quantity of memory node copy according to the storage node accesses amount of prediction；
Wherein, original series are constructed according to historical data, the cumulative sequence of Accumulating generation is carried out to the original series, then
Differential equation of first order is established to obtain prediction model to the cumulative sequence.
In embodiments of the present invention, the prediction module 201 is also used to: the history access data of memory node are inputted institute
It states prediction model and obtains memory node prediction array；The institute of inverse accumulated generating prediction is carried out to the memory node prediction array
State storage node accesses amount.
In embodiments of the present invention, the adjustment module 202 is also used to: being accessed based on the prediction model forecasting system
Amount；Memory node, which is calculated, according to the storage node accesses amount of prediction and the system amount of access of prediction predicts temperature；Base
The quantity of the memory node copy is adjusted in memory node prediction temperature.
In embodiments of the present invention, the adjustment module 202 is further used for: by the history access data input institute of system
It states prediction model and obtains system prediction array；The system for carrying out an inverse accumulated generating prediction to the system prediction array is visited
The amount of asking.
In addition, the memory node prediction temperature determines according to the following formula:Wherein, HiIndicate the
The memory node in i period predicts temperature, HjIndicate the averaged historical temperature of memory node, miIndicate i-th of period of prediction
Storage node accesses amount, NiIndicate the system amount of access in i-th of period of prediction, 0≤p≤1,0≤q≤1, and p+q=1.
In embodiments of the present invention, described device further include: selecting module, for when user requests access to system, base
In the memory node copy of the response time Response to selection user request of the access path and memory node copy of user.
The device of adjustment memory node copy amount according to an embodiment of the present invention can be seen that because using based on prediction
Model predicts the amount of access of some memory node, and adjusts the copy of the memory node according to the storage node accesses amount of prediction
The technological means of quantity, thus the quantity for overcoming memory node copy be difficult to determine and memory node failure after, restore number
The larger technical problem of overhead during, and then reach and additions and deletions are carried out to memory node copy in advance, if in advance
The quantity for increasing memory node copy can be avoided the problem of memory node copy replication lags, effectively alleviate node load；
If reducing the quantity of memory node copy in advance, memory space can be saved；Meanwhile for failure memory node data
Restore, need to only download a memory node based on complete copy redundancy can complete recovery process, and the system of reducing is opened
The technical effect of pin.
The method of the adjustment memory node copy amount of embodiment in order to enable those skilled in the art to better understand the present invention
And device, it is introduced now in conjunction with the prediction model of Fig. 3, Fig. 4 to the embodiment of the present invention:
Usually complete data information includes element information, boundary information, structural information, operation action information.It is completely bright
True data information is white information；Complete indefinite data information is black information；The indefinite letter in the clear part in part
Breath is referred to as grey information.Wherein, information it is indefinite be broadly divided into element information is incomplete, boundary information not exclusively, structure letter
Cease incomplete, operation action INFORMATION OF INCOMPLETE.
Gray theory thinks that all information be exactly discrepant, INFORMATION OF INCOMPLETE solution be non-unique, information is cognition
Foundation, new information old information is greater than to the effect of cognition, " information is indefinite " be it is absolute, the characteristics of gray theory is exactly
Existing " minimum information " fully is utilized, the embodiment of the present invention is based on gray theory and proposes prediction model, passes through the prediction mould
The amount of access of the amount of access that each memory node is predicted based on partial history data and system may be implemented in type.
In the embodiment of the present invention, the building of prediction model is that discrete data are considered as continuous variable during variation
Acquired discrete value, and data are handled using the differential equation, then initial data is added up to obtain formation sequence, most
Formation sequence is constructed into differential equation of first order afterwards, to obtain prediction model, and the prediction model can offset part at random
Error shows the regularity of discrete data.
Now it is illustrated by principle of the following embodiment to the prediction model of the embodiment of the present invention:
Assuming that initial data, there are four data, this four data values are as shown in table 1.
Table 1
Firstly, drawing the available Fig. 3 of tendency chart according to initial data, the development trend that Fig. 3 show initial data is shown
It is intended to, for initial data without apparent rule, development trend is fluctuation as seen from Figure 3.
Then, initial data is subjected to accumulation operations, generates one group of new data sequence, can indicates are as follows:
X(1)(1)=X(0)(1)=2；
X(1)(2)=X(0)(1)+X(0)(2)=2+3=5；
X(1)(3)=X(0)(1)+X(0)(2)+X(0)(3)=2+3+2.5=7.5；
X(1)(4)=X(0)(1)+X(0)(2)+X(0)(3)+X(0)(4)=2+3+2.5+5=12.5；
The data sequence generated by accumulation operations is as shown in table 2.
Table 2
The available Fig. 4 of tendency chart is drawn according to the data sequence that accumulation operations generate, Fig. 4 show accumulation operations generation
Data sequence development trend schematic diagram, as seen from Figure 4 accumulation operations generate data sequence be monotonic increase number
Column, are provided with apparent regularity.Gray theory is exactly that discrete data are carried out accumulation process, to show discrete data
Regularity, data sequence variation at any time, but total trend have potential rule, can will be discrete according to potential rule
Irregular data be changed into regular data sequence.
Now the foundation of the prediction model of the embodiment of the present invention is illustrated:
The embodiment of the present invention constructs prediction model according to given data information and date and the variation relation of time series, into
The prediction of row data information, meanwhile, prediction model is based on first differential GM (1,1) model construction, utilizes single order according to known ordered series of numbers
The differential equation embodies the regularity that data are showed after the time is cumulative.
Firstly, constructing original predictive sequence: X according to given data(0)={ X(0)(i), i=1,2 ... K }.
Then, original predictive sequence is made into one-accumulate and generates (1-AGO), obtain formation sequence X(1)={ X(1)(K), K=
1,2……}。
Wherein,
Finally, to X(1)Differential equation of first order is established, GM (1,1) model is obtained:
Wherein, a is development system；U is grey actuating quantity.
Prediction model can be obtained by solving above-mentioned formula:
Prediction model can be denoted as argument sequence It is solved with following formula:
Wherein, B is data matrix, YKFor data column；
YK=[X(0)(2), X(0)(3) ... ... X(0)(K)]T。
It, must will be obtained by GM (1,1) model it should be noted that is obtained due to GM (1,1) model is one-accumulate amount
DataIt is reduced to by regressiveThat is:
Cause are as follows:
So:
ThenValue when for K+1.
After the adjustment of former sequence, new sequence X (0)={ X (0) (i), i=2,3 ..., K+1 } is constituted.New sequence is used for
Next prediction thus constitutes dynamic and updates forecast sample sequence, and realizes dynamic prediction.
The above content is the introduction for the amount of access for how predicting using prediction model each memory node, based on measuring in advance
To each memory node amount of access can the memory node copy amount to each memory node be adjusted.In addition, this
Inventive embodiments can also predict the amount of access of each memory node and the amount of access of system using prediction model, according to pre-
The amount of access of each memory node and the amount of access of system surveyed calculate the memory node prediction temperature of each memory node, are based on
Memory node prediction temperature is adjusted the memory node copy amount of each memory node.
Since the temperature of memory node is mainly related with the amount of access of memory node, but storage is saved based on temperature threshold values
Point copy is adjusted, and when amount of access fluctuation is larger, there are problems that duplication lag.Therefore, proposition pair of the embodiment of the present invention
The temperature of memory node is predicted that the memory node based on prediction predicts temperature to the memory node copy of each memory node
Quantity is adjusted.Assuming that the memory node prediction temperature of some memory node is predicted according to the historical data in three periods,
Then prediction model forecasting system amount of access and storage node accesses amount can be based on according to the historical data in three periods, thus
Calculate the memory node prediction temperature of each memory node.Specifically:
As shown in table 3, it is the historical data in three periods, is the storage node accesses in each three periods respectively in table
Amount, system amount of access, temperature rank, memory node number of copies, wherein temperature rank can be by memory node temperature and each temperature
The corresponding heating range of rank is compared to determine, the corresponding heating range of each temperature rank can root in the embodiment of the present invention
It is preset according to actual conditions or demand.
Table 3
Historical data based on three periods predicts the system amount of access and memory node in next period using prediction model
Amount of access, and temperature is predicted according to system amount of access and the calculated memory node of storage node accesses amount, wherein memory node
Temperature can be according to formulaIt calculates.Obtain file each time cycle amount of access and each time cycle
Total amount of access, and then according to the amount of access of history amount of access prediction next cycle.
System amount of access and storage node accesses amount are predicted by prediction model, obtain the storage of next cycle
Node visit amount is 1063, system amount of access is 4808, then calculates memory node prediction temperature and determine temperature rank.It will prediction
System amount of access, storage node accesses amount and temperature grade renewal table 4, as shown in table 4, the 1st, 2 are obtained into historical data
The corresponding period is historical data, and corresponding the 3rd period is prediction data, and the quantity of memory node copy is according to temperature rank tune
It is whole.
Table 4
Assuming that the actual amount of access of the actual amount of access of the memory node of next period and the system is as shown in table 5, lead to
Although crossing comparison sheet 4 and table 5 it can be found that the variation of storage node accesses amount is smaller, system amount of access is significantly reduced, because
This, the temperature of the memory node is actually increasing, so, in order to make the load balancing of each memory node, at the 3rd
Between the period, temperature rank becomes 2 from 3, and the quantity of memory node copy increases to 4 by 3, due to system amount of access, storage
Node visit amount is to predict to obtain according to history amount of access in advance, so temperature rank and number of copies are provided in advance, thus pole
The big Provisioning Policy for optimizing number of copies, is effectively relieved node load.
Table 5
Fig. 5 is shown can be using the method or adjustment storage section of the adjustment memory node copy amount of the embodiment of the present invention
The exemplary system architecture 500 of the device of point copy amount.
As shown in figure 5, system architecture 500 may include terminal device 501,502,503, network 504 and server 505.
Network 504 between terminal device 501,502,503 and server 505 to provide the medium of communication link.Network 504 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 501,502,503 and be interacted by network 504 with server 505, to receive or send out
Send message etc..Various telecommunication customer end applications, such as the application of shopping class, net can be installed on terminal device 501,502,503
The application of page browsing device, searching class application, instant messaging tools, mailbox client, social platform software etc..
Terminal device 501,502,503 can be the various electronic equipments with display screen and supported web page browsing, packet
Include but be not limited to smart phone, tablet computer, pocket computer on knee and desktop computer etc..
Server 505 can be to provide the server of various services, such as utilize terminal device 501,502,503 to user
The shopping class website browsed provides the back-stage management server (merely illustrative) supported.Back-stage management server can be to reception
To the data such as information query request analyze etc. processing, and by processing result (such as target push information, product letter
Breath -- merely illustrative) feed back to terminal device.
It should be noted that the method for adjustment memory node copy amount is generally by servicing provided by the embodiment of the present invention
Device 505 executes, and correspondingly, the device of adjustment memory node copy amount is generally positioned in server 505.
It should be understood that the number of terminal device, network and server in Fig. 5 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
Below with reference to Fig. 6, it illustrates the computer systems 600 for the terminal device for being suitable for being used to realize the embodiment of the present invention
Structural schematic diagram.Terminal device shown in Fig. 6 is only an example, function to the embodiment of the present invention and should not use model
Shroud carrys out any restrictions.
As shown in fig. 6, computer system 600 includes central processing unit (CPU) 601, it can be read-only according to being stored in
Program in memory (ROM) 602 or be loaded into the program in random access storage device (RAM) 603 from storage section 608 and
Execute various movements appropriate and processing.In RAM 603, also it is stored with system 600 and operates required various programs and data.
CPU 601, ROM 602 and RAM 603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to always
Line 604.
I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 608 including hard disk etc.；
And the communications portion 609 of the network interface card including LAN card, modem etc..Communications portion 609 via such as because
The network of spy's net executes communication process.Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to read from thereon
Computer program be mounted into storage section 608 as needed.
Particularly, disclosed embodiment, the process described above with reference to flow chart may be implemented as counting according to the present invention
Calculation machine software program.For example, embodiment disclosed by the invention includes a kind of computer program product comprising be carried on computer
Computer program on readable medium, the computer program include the program code for method shown in execution flow chart.?
In such embodiment, which can be downloaded and installed from network by communications portion 609, and/or from can
Medium 611 is dismantled to be mounted.When the computer program is executed by central processing unit (CPU) 601, system of the invention is executed
The above-mentioned function of middle restriction.
It should be noted that computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the present invention, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In invention, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule
The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
Being described in module involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part is realized.Described module also can be set in the processor, for example, can be described as: a kind of processor packet
Include prediction module and adjustment module.Wherein, the title of these modules does not constitute the limit to the module itself under certain conditions
It is fixed, for example, prediction module is also described as " module based on prediction model prediction storage node accesses amount ".
As on the other hand, the present invention also provides a kind of computer-readable medium, which be can be
Included in equipment described in above-described embodiment；It is also possible to individualism, and without in the supplying equipment.Above-mentioned calculating
Machine readable medium carries one or more program, when said one or multiple programs are executed by the equipment, makes
Obtaining the equipment includes: step S101: predicting storage node accesses amount based on prediction model；Step S102: the storage based on prediction
The quantity of node visit amount adjustment memory node copy.
Technical solution according to an embodiment of the present invention, because using the visit for predicting some memory node based on prediction model
The amount of asking, and the technological means of the copy amount of the memory node is adjusted according to the storage node accesses amount of prediction, so overcoming
The quantity of memory node copy be difficult to determine and memory node failure after, overhead is biggish during restoring data
Technical problem, and then reach and additions and deletions are carried out to memory node copy in advance, if increasing the quantity of memory node copy, energy in advance
It enough avoids the problem that memory node copy replication lags, effectively alleviates node load；If reducing memory node copy in advance
Quantity, then can save memory space；Meanwhile the recovery for the memory node data that fail, it is based on complete copy redundancy skill
Art, which need to only download a memory node, can complete recovery process, reduce the technical effect of overhead.
Above-mentioned specific embodiment, does not constitute a limitation on the scope of protection of the present invention.Those skilled in the art should be bright
It is white, design requirement and other factors are depended on, various modifications, combination, sub-portfolio and substitution can occur.It is any
Made modifications, equivalent substitutions and improvements etc. within the spirit and principles in the present invention, should be included in the scope of the present invention
Within.
Claims (14)
1. a kind of method for adjusting memory node copy amount characterized by comprising
Storage node accesses amount is predicted based on prediction model；
The quantity of memory node copy is adjusted according to the storage node accesses amount of prediction；
Wherein, original series are constructed according to historical data, the cumulative sequence of Accumulating generation is carried out to the original series, then to institute
It states cumulative sequence and establishes differential equation of first order to obtain prediction model.
2. the method according to claim 1, wherein including: based on prediction model prediction storage node accesses amount
The history access data of memory node are inputted into the prediction model and obtain memory node prediction array；
The storage node accesses amount of inverse accumulated generating prediction is carried out to the memory node prediction array.
3. being stored the method according to claim 1, wherein being adjusted according to the storage node accesses amount of prediction
The quantity of node copy includes:
Based on the prediction model forecasting system amount of access；
Memory node, which is calculated, according to the storage node accesses amount of prediction and the system amount of access of prediction predicts temperature；
The quantity of the memory node copy is adjusted based on memory node prediction temperature.
4. according to the method described in claim 3, it is characterized in that, including: based on the prediction model forecasting system amount of access
The history access data of system are inputted into the prediction model and obtain system prediction array；
The system amount of access of inverse accumulated generating prediction is carried out to the system prediction array.
5. according to the method described in claim 3, it is characterized in that, memory node prediction temperature is true according to the following formula
It is fixed:
Wherein, HiIndicate the memory node prediction temperature in i-th of period, HjIndicate the averaged historical temperature of memory node, miIt indicates
The storage node accesses amount in i-th of period of prediction, NiThe system amount of access in i-th of period of expression prediction, 0≤p≤1,0≤
Q≤1, and p+q=1.
6. the method according to claim 1, wherein the method also includes:
When user requests access to system, the response time of access path and the memory node copy based on user selects to ring
The memory node copy requested using family.
7. a kind of device for adjusting memory node copy amount characterized by comprising
Prediction module, for predicting storage node accesses amount based on prediction model；
Module is adjusted, for adjusting the quantity of memory node copy according to the storage node accesses amount of prediction；
Wherein, original series are constructed according to historical data, the cumulative sequence of Accumulating generation is carried out to the original series, then to institute
It states cumulative sequence and establishes differential equation of first order to obtain prediction model.
8. device according to claim 7, which is characterized in that the prediction module is also used to:
The history access data of memory node are inputted into the prediction model and obtain memory node prediction array；
The storage node accesses amount of inverse accumulated generating prediction is carried out to the memory node prediction array.
9. device according to claim 7, which is characterized in that the adjustment module is also used to:
Based on the prediction model forecasting system amount of access；
Memory node, which is calculated, according to the storage node accesses amount of prediction and the system amount of access of prediction predicts temperature；
The quantity of the memory node copy is adjusted based on memory node prediction temperature.
10. device according to claim 9, which is characterized in that the adjustment module is further used for:
The history access data of system are inputted into the prediction model and obtain system prediction array；
The system amount of access of inverse accumulated generating prediction is carried out to the system prediction array.
11. device according to claim 9, which is characterized in that the memory node prediction temperature is true according to the following formula
It is fixed:
Wherein, HiIndicate the memory node prediction temperature in i-th of period, HjIndicate the averaged historical temperature of memory node, miIt indicates
The storage node accesses amount in i-th of period of prediction, NiThe system amount of access in i-th of period of expression prediction, 0≤p≤1,0≤
Q≤1, and p+q=1.
12. device according to claim 7, which is characterized in that described device further include:
Selecting module, for when user requests access to system, access path and the memory node copy based on user
The memory node copy of response time Response to selection user request.
13. a kind of electronic equipment for adjusting memory node copy amount characterized by comprising
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as method as claimed in any one of claims 1 to 6.
14. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is held by processor
Such as method as claimed in any one of claims 1 to 6 is realized when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710982861.8A CN109697018A (en) | 2017-10-20 | 2017-10-20 | The method and apparatus for adjusting memory node copy amount |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710982861.8A CN109697018A (en) | 2017-10-20 | 2017-10-20 | The method and apparatus for adjusting memory node copy amount |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109697018A true CN109697018A (en) | 2019-04-30 |
Family
ID=66225235
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710982861.8A Pending CN109697018A (en) | 2017-10-20 | 2017-10-20 | The method and apparatus for adjusting memory node copy amount |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109697018A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110209693A (en) * | 2019-05-10 | 2019-09-06 | 苏宁云计算有限公司 | High concurrent data query method, apparatus, system, equipment and readable storage medium storing program for executing |
| CN110266525A (en) * | 2019-06-04 | 2019-09-20 | 湖北五五互联科技有限公司 | CDN server quantity configuration method, equipment and computer readable storage medium |
| CN110968745A (en) * | 2019-11-13 | 2020-04-07 | 泰康保险集团股份有限公司 | Data processing method and device, electronic equipment and computer readable medium |
| CN111475108A (en) * | 2020-03-20 | 2020-07-31 | 平安国际智慧城市科技股份有限公司 | Distributed storage method, computer equipment and computer readable storage medium |
| CN112734103A (en) * | 2021-01-05 | 2021-04-30 | 烽火通信科技股份有限公司 | Video cold picture prediction method and device based on space-time sequence |
| CN114205416A (en) * | 2021-10-27 | 2022-03-18 | 北京旷视科技有限公司 | Resource caching method and device, electronic equipment and computer readable medium |
| CN114691033A (en) * | 2022-02-17 | 2022-07-01 | 阿里巴巴（中国）有限公司 | Data copying method, data storage system control method, device, equipment and medium |
| CN115185996A (en) * | 2022-07-19 | 2022-10-14 | 广州凯园软件科技有限公司 | Education training comprehensive evaluation system based on point system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1658560A (en) * | 2005-02-28 | 2005-08-24 | 中国人民解放军国防科学技术大学 | Quickly data copy method based on prediction |
| CN106331001A (en) * | 2015-06-23 | 2017-01-11 | 华中科技大学 | Cloud storage method and system for access of mobile device |
| US20170060957A1 (en) * | 2015-09-01 | 2017-03-02 | Dell Software, Inc. | Optimized and consistent replication of file overwrites |
| CN106651023A (en) * | 2016-12-17 | 2017-05-10 | 三峡大学 | Grey correlation analysis-based improved fireworks algorithm mid-long term load prediction method |
| CN106796545A (en) * | 2014-12-28 | 2017-05-31 | 斯特拉托斯卡莱有限公司 | Recovery synchronization in distributed memory system |
| CN106886376A (en) * | 2017-03-30 | 2017-06-23 | 上海海洋大学 | A kind of marine monitoring data trnascription management method optimized based on many attributes |
-
2017
- 2017-10-20 CN CN201710982861.8A patent/CN109697018A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1658560A (en) * | 2005-02-28 | 2005-08-24 | 中国人民解放军国防科学技术大学 | Quickly data copy method based on prediction |
| CN106796545A (en) * | 2014-12-28 | 2017-05-31 | 斯特拉托斯卡莱有限公司 | Recovery synchronization in distributed memory system |
| CN106331001A (en) * | 2015-06-23 | 2017-01-11 | 华中科技大学 | Cloud storage method and system for access of mobile device |
| US20170060957A1 (en) * | 2015-09-01 | 2017-03-02 | Dell Software, Inc. | Optimized and consistent replication of file overwrites |
| CN106651023A (en) * | 2016-12-17 | 2017-05-10 | 三峡大学 | Grey correlation analysis-based improved fireworks algorithm mid-long term load prediction method |
| CN106886376A (en) * | 2017-03-30 | 2017-06-23 | 上海海洋大学 | A kind of marine monitoring data trnascription management method optimized based on many attributes |
Non-Patent Citations (1)
| Title |
|---|
| 李希亮: "基于RS纠删码的数据冗余策略", 《道客巴巴，HTTPS://WWW.DOC88.COM/P-6002345239131.HTML》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110209693A (en) * | 2019-05-10 | 2019-09-06 | 苏宁云计算有限公司 | High concurrent data query method, apparatus, system, equipment and readable storage medium storing program for executing |
| CN110266525A (en) * | 2019-06-04 | 2019-09-20 | 湖北五五互联科技有限公司 | CDN server quantity configuration method, equipment and computer readable storage medium |
| CN110266525B (en) * | 2019-06-04 | 2022-02-22 | 湖北五五互联科技有限公司 | CDN server number configuration method, equipment and computer readable storage medium |
| CN110968745A (en) * | 2019-11-13 | 2020-04-07 | 泰康保险集团股份有限公司 | Data processing method and device, electronic equipment and computer readable medium |
| CN111475108A (en) * | 2020-03-20 | 2020-07-31 | 平安国际智慧城市科技股份有限公司 | Distributed storage method, computer equipment and computer readable storage medium |
| CN111475108B (en) * | 2020-03-20 | 2023-11-28 | 深圳赛安特技术服务有限公司 | Distributed storage method, computer equipment and computer readable storage medium |
| CN112734103A (en) * | 2021-01-05 | 2021-04-30 | 烽火通信科技股份有限公司 | Video cold picture prediction method and device based on space-time sequence |
| CN114205416A (en) * | 2021-10-27 | 2022-03-18 | 北京旷视科技有限公司 | Resource caching method and device, electronic equipment and computer readable medium |
| CN114205416B (en) * | 2021-10-27 | 2024-03-12 | 北京旷视科技有限公司 | Resource caching method, device, electronic equipment and computer readable medium |
| CN114691033A (en) * | 2022-02-17 | 2022-07-01 | 阿里巴巴（中国）有限公司 | Data copying method, data storage system control method, device, equipment and medium |
| CN115185996A (en) * | 2022-07-19 | 2022-10-14 | 广州凯园软件科技有限公司 | Education training comprehensive evaluation system based on point system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109697018A (en) | The method and apparatus for adjusting memory node copy amount | |
| TWI620075B (en) | Server and cloud computing resource optimization method thereof for cloud big data computing architecture | |
| CN101013427B (en) | Method and system for managing data | |
| US11057490B2 (en) | Systems and methods for utilizing unused network capacity for prefetch requests | |
| KR20230070272A (en) | Computer-based systems, computing components, and computing objects configured to implement dynamic outlier bias reduction in machine learning models | |
| JP7420972B2 (en) | Methods, systems and apparatus for storing blockchain-based data | |
| Qi et al. | An invocation cost optimization method for web services in cloud environment | |
| CN103368867A (en) | Method and system of cached object communicating with secondary site through network | |
| CN108876379A (en) | A kind of method and apparatus for generating message | |
| CN106464669A (en) | Intelligent file pre-fetch based on access patterns | |
| Nguyen et al. | Two-stage robust edge service placement and sizing under demand uncertainty | |
| Yang et al. | Traffic Flow Prediction Model for Large‐Scale Road Network Based on Cloud Computing | |
| CN103685347A (en) | Method and device for allocating network resources | |
| Min et al. | An approach to resource and QoS-aware services optimal composition in the big service and Internet of Things | |
| CN109002925A (en) | Traffic prediction method and apparatus | |
| Choi et al. | Quality evaluation and best service choice for cloud computing based on user preference and weights of attributes using the analytic network process | |
| CN114423023B (en) | Mobile user-oriented 5G network edge server deployment method | |
| Tian et al. | A Pareto‐Based Adaptive Variable Neighborhood Search for Biobjective Hybrid Flow Shop Scheduling Problem with Sequence‐Dependent Setup Time | |
| Kim | A column generation heuristic for congested facility location problem with clearing functions | |
| Shao et al. | Cost‐Aware Placement Optimization of Edge Servers for IoT Services in Wireless Metropolitan Area Networks | |
| CN111044062A (en) | Path planning and recommending method and device | |
| CN109902847A (en) | Prediction divides the method and apparatus of library order volume | |
| US20220309338A1 (en) | Discrete optimization of configuration attributes | |
| Niu et al. | An asynchronous fixed-point algorithm for resource sharing with coupled objectives | |
| Cheng et al. | Globally optimal selection of web composite services based on univariate marginal distribution algorithm |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190430 |