CN117112236A - Jurisdictional server configuration method and system based on data inrush current and volatility prediction - Google Patents

Jurisdictional server configuration method and system based on data inrush current and volatility prediction Download PDF

Info

Publication number
CN117112236A
CN117112236A CN202311368351.3A CN202311368351A CN117112236A CN 117112236 A CN117112236 A CN 117112236A CN 202311368351 A CN202311368351 A CN 202311368351A CN 117112236 A CN117112236 A CN 117112236A
Authority
CN
China
Prior art keywords
access
processing resource
partition
occupation
processing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202311368351.3A
Other languages
Chinese (zh)
Other versions
CN117112236B (en
Inventor
赵延军
袁一鹏
赵建伟
王春明
李业庭
廖磊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Shuguangzhao Information Technology Co ltd
Original Assignee
Shandong Shuguangzhao Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong Shuguangzhao Information Technology Co ltd filed Critical Shandong Shuguangzhao Information Technology Co ltd
Priority to CN202311368351.3A priority Critical patent/CN117112236B/en
Publication of CN117112236A publication Critical patent/CN117112236A/en
Application granted granted Critical
Publication of CN117112236B publication Critical patent/CN117112236B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08Configuration management of networks or network elements
    • H04L41/0896Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention provides a jurisdictional server configuration method and system based on data inrush current and volatility prediction, and relates to the technical field of computers, wherein the method comprises the following steps: acquiring processing resources and bandwidth resources of each partition in an ith detection period; acquiring access quantity and processing resource occupation proportion at a plurality of moments; determining the response time length of the processing resources according to the access quantity and the occupation proportion of the processing resources; determining an access mode type according to the access quantity, the occupation proportion of the processing resources and the response time length of the processing resources; determining a configuration adjustment strategy according to the type of the access mode, the processing resources and the bandwidth resources, and the access quantity and the occupation proportion of the processing resources; and according to the configuration adjustment strategy, the processing resource and the bandwidth resource of the (i+1) th detection period are configured. According to the invention, the access requirement can be responded in time when the access data fluctuates, so that the continuously-changing access requirement is met.

Description

Jurisdictional server configuration method and system based on data inrush current and volatility prediction
Technical Field
The invention relates to the technical field of computers, in particular to a jurisdictional server configuration method and system based on data inrush current and volatility prediction.
Background
In the related art, the server may divide the partitions for different jurisdictions, so that the respective partitions are used to correspond to access requests of the respective jurisdictions, and the processing resources and bandwidth resources of each partition are generally fixed, and the processing resources and bandwidth resources of the respective partition may be configured according to historical access data of the jurisdiction. However, access data of each jurisdiction may fluctuate, so that it is difficult for a fixed resource allocation mode to meet access requirements when the access data fluctuates.
The information disclosed in the background section of the application is only for enhancement of understanding of the general background of the application and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
The embodiment of the application provides a jurisdictional server configuration method and a jurisdictional server configuration system based on data inrush current and volatility prediction, which can timely respond to access requirements when access data fluctuates, so as to meet the continuously-changing access requirements.
According to a first aspect of an embodiment of the present application, there is provided a jurisdictional server configuration method based on data inrush current and volatility prediction, including:
In the ith detection period, processing resources and bandwidth resources of each partition of the district server are obtained, wherein i is a positive integer;
the access quantity and the processing resource occupation proportion of each partition in the ith detection period at a plurality of moments are obtained;
determining the response time length of the processing resources according to the access amount and the occupation proportion of the processing resources of each partition at a plurality of moments;
determining the access mode type of each partition according to the access amount and the occupation proportion of the processing resources of each partition at a plurality of moments and the response time length of the processing resources;
determining a configuration adjustment strategy of each partition in the (i+1) th detection period according to the access mode type of each partition, the processing resources and bandwidth resources of each partition in the (i) th detection period, and the access amount and the processing resource occupation proportion of each partition at a plurality of moments, wherein the configuration adjustment strategy comprises a bandwidth resource adjustment amount and a processing resource adjustment amount;
and according to the configuration adjustment strategy, processing resources and bandwidth resources of the partitions in the (i+1) th detection period are configured.
According to one embodiment of the present invention, determining a processing resource response duration according to the access amounts and the processing resource occupation ratios of the respective partitions at a plurality of moments includes:
Obtaining access quantity vectors of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation vector of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a first response time length according to the access quantity vector, the processing resource occupation vector and the time interval among all the moments;
obtaining access quantity fluctuation functions of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation fluctuation function of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a second response time length according to the access quantity fluctuation function and the processing resource occupation fluctuation function;
and determining the response time length of the processing resource according to the first response time length and the second response time length.
According to one embodiment of the present invention, determining the first response time period according to the access amount vector, the processing resource occupation vector, and the time interval between the respective times includes:
according to the formula
Obtaining a first similarityWherein, for processing the resource occupancy vector, +.>For the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth detection period,/for the kth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the >For the access amount of the kth time in the access amount vector of the jth partition in the ith detection period, k is a positive integer less than or equal to n, n is the number of a plurality of times, +.>For the maximum value of the elements in the access quantity vector of the jth partition in the ith detection period, +.>For the minimum value of the elements in the access quantity vector of the jth partition in the ith detection period,/->Is a displacement matrix of the step s, s is an integer which is more than or equal to 0 and less than or equal to n-1,
indicating ++in case the step number s is the kth step>Otherwise->,/>In the case of a number of steps s, < > in>And->Covariance matrix between->Is->An inverse matrix of (a);
determining a number of steps at which the first similarity reaches a maximum value;
and determining the first response time length according to the step number when the first similarity reaches the maximum value and the time interval between the moments.
According to one embodiment of the present invention, determining a second response time period according to the access amount fluctuation function and the processing resource occupation fluctuation function includes:
according to the formula
Obtaining a second degree of similarityWherein t is the time in the ith detection period,/and>for the start time of the ith detection period, < +.>For the end time of the ith detection period, < +. >For the access volume fluctuation function +.>For processing resource occupancy fluctuation functions, +.>Is the displacement duration;
determining a displacement time length when the second similarity reaches a maximum value;
and taking the displacement time length when the second similarity reaches the maximum value as the second response time length.
According to one embodiment of the present invention, determining the processing resource response time length according to the first response time length and the second response time length includes:
according to the formula
Obtaining the processing resource response time lengthWherein->For the first response time length, +.>And the second response time length is the second response time length.
According to one embodiment of the invention, the access mode type of each partition is determined according to the access amount and the occupation proportion of the processing resources of each partition at a plurality of moments and the response time of the processing resources, and the access mode type of each partition comprises one of the following steps:
when the maximum value of the access quantity at a plurality of moments is larger than or equal to the access quantity threshold value, the maximum value of the occupation proportion of the processing resources is smaller than the occupation proportion threshold value, and the response time of the processing resources is smaller than the time threshold value, determining the access mode as a low-occupation high-frequency access mode;
determining an access mode as a high-occupancy low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than an access amount threshold, the maximum processing resource occupancy proportion is smaller than an occupancy proportion threshold, and the processing resource response length is larger than or equal to a duration threshold;
Determining an access mode as a high-occupancy high-frequency access mode under the condition that the maximum access amount at a plurality of moments is larger than or equal to an access amount threshold value, the maximum processing resource occupancy proportion is larger than or equal to an occupancy proportion threshold value, and the processing resource response length is larger than or equal to a duration threshold value;
and determining the access mode as a low-occupancy and low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than the access amount threshold, the maximum processing resource occupancy proportion is smaller than the occupancy proportion threshold and the processing resource response time length is smaller than the time length threshold.
According to one embodiment of the present invention, determining a configuration adjustment policy of each partition in the (i+1) th detection period according to the access mode type of each partition, the processing resource and the bandwidth resource of each partition in the (i) th detection period, and the access amount and the processing resource occupation ratio of the plurality of moments, includes:
determining a processing resource adjustment amount according to the difference between the maximum value of the processing resource occupation proportion and the threshold value of the occupation proportion and the processing resource of the ith detection period, and determining a bandwidth resource adjustment amount according to the difference between the maximum value of the access amount and the threshold value of the access amount at a plurality of moments and the bandwidth resource of the ith detection period, wherein the bandwidth resource adjustment amount is increased in a low-occupation high-frequency access mode, the processing resource adjustment amount is decreased in a high-occupation low-frequency access mode, the processing resource adjustment amount is increased in a high-occupation low-frequency access mode, the bandwidth resource adjustment amount is decreased in a high-occupation high-frequency access mode, the processing resource adjustment amount and the bandwidth resource adjustment amount are increased in a high-occupation high-frequency access mode, and the processing resource adjustment amount and the bandwidth resource adjustment amount are decreased in a low-occupation low-frequency access mode.
According to a second aspect of embodiments of the present invention, there is provided a jurisdictional server configuration system based on data inrush current and volatility prediction, comprising:
the configuration acquisition module is used for acquiring processing resources and bandwidth resources of each partition of the district server in the ith detection period, wherein i is a positive integer;
the state acquisition module is used for acquiring the access quantity and the processing resource occupation proportion of each partition in the ith detection period at a plurality of moments;
the resource response time length module is used for determining the response time length of the processing resources according to the access quantity and the occupation proportion of the processing resources of each partition at a plurality of moments;
the access mode type module is used for determining the access mode type of each partition according to the access quantity and the processing resource occupation proportion of each partition at a plurality of moments and the processing resource response time length;
the configuration adjustment strategy module is used for determining a configuration adjustment strategy of each partition in the (i+1) th detection period according to the access mode type of each partition, the processing resources and bandwidth resources of each partition in the (i) th detection period, the access quantity and the processing resource occupation proportion of the plurality of moments, wherein the configuration adjustment strategy comprises a bandwidth resource adjustment quantity and a processing resource adjustment quantity;
And the configuration module is used for configuring the processing resources and the bandwidth resources of each partition in the (i+1) th detection period according to the configuration adjustment strategy.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
obtaining access quantity vectors of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation vector of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a first response time length according to the access quantity vector, the processing resource occupation vector and the time interval among all the moments;
obtaining access quantity fluctuation functions of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation fluctuation function of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a second response time length according to the access quantity fluctuation function and the processing resource occupation fluctuation function;
and determining the response time length of the processing resource according to the first response time length and the second response time length.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
According to the formula
Obtaining a first similarityWherein, for processing the resource occupancy vector, +.>For the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth detection period,/for the kth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the>For the access amount of the kth time in the access amount vector of the jth partition in the ith detection period, k is a positive integer less than or equal to n, n is the number of a plurality of times, +.>For the maximum value of the elements in the access quantity vector of the jth partition in the ith detection period, +.>For the minimum value of the elements in the access quantity vector of the jth partition in the ith detection period,/->The displacement matrix is the step s, s is greater than or equal toAn integer of 0 to less than or equal to n-1,
indicating ++in case the step number s is the kth step>Otherwise,/>In the case of a number of steps s, < > in>And->Covariance matrix between->Is->An inverse matrix of (a);
determining a number of steps at which the first similarity reaches a maximum value;
and determining the first response time length according to the step number when the first similarity reaches the maximum value and the time interval between the moments.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
according to the formula
Obtaining a second degree of similarityWherein t is the time in the ith detection period,/and>for the start time of the ith detection period, < +.>For the end time of the ith detection period, < +.>For the access volume fluctuation function +.>For processing resource occupancy fluctuation functions, +.>Is the displacement duration;
determining a displacement time length when the second similarity reaches a maximum value;
and taking the displacement time length when the second similarity reaches the maximum value as the second response time length.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
according to the formula
Obtaining the processing resource response time lengthWherein->For the first response time length, +.>And the second response time length is the second response time length.
According to one embodiment of the invention, the access pattern type module is further adapted to one of:
when the maximum value of the access quantity at a plurality of moments is larger than or equal to the access quantity threshold value, the maximum value of the occupation proportion of the processing resources is smaller than the occupation proportion threshold value, and the response time of the processing resources is smaller than the time threshold value, determining the access mode as a low-occupation high-frequency access mode;
determining an access mode as a high-occupancy low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than an access amount threshold, the maximum processing resource occupancy proportion is smaller than an occupancy proportion threshold, and the processing resource response length is larger than or equal to a duration threshold;
Determining an access mode as a high-occupancy high-frequency access mode under the condition that the maximum access amount at a plurality of moments is larger than or equal to an access amount threshold value, the maximum processing resource occupancy proportion is larger than or equal to an occupancy proportion threshold value, and the processing resource response length is larger than or equal to a duration threshold value;
and determining the access mode as a low-occupancy and low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than the access amount threshold, the maximum processing resource occupancy proportion is smaller than the occupancy proportion threshold and the processing resource response time length is smaller than the time length threshold.
According to one embodiment of the invention, the configuration adjustment policy module is further configured to:
determining a processing resource adjustment amount according to the difference between the maximum value of the processing resource occupation proportion and the threshold value of the occupation proportion and the processing resource of the ith detection period, and determining a bandwidth resource adjustment amount according to the difference between the maximum value of the access amount and the threshold value of the access amount at a plurality of moments and the bandwidth resource of the ith detection period, wherein the bandwidth resource adjustment amount is increased in a low-occupation high-frequency access mode, the processing resource adjustment amount is decreased in a high-occupation low-frequency access mode, the processing resource adjustment amount is increased in a high-occupation low-frequency access mode, the bandwidth resource adjustment amount is decreased in a high-occupation high-frequency access mode, the processing resource adjustment amount and the bandwidth resource adjustment amount are increased in a high-occupation high-frequency access mode, and the processing resource adjustment amount and the bandwidth resource adjustment amount are decreased in a low-occupation low-frequency access mode.
According to a third aspect of embodiments of the present invention, there is provided a jurisdictional server configuration apparatus based on data inrush current and volatility prediction, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored by the memory to perform the jurisdictional server configuration method based on data inrush and volatility prediction.
According to a fourth aspect of embodiments of the present invention, there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the jurisdictional server configuration method based on data inrush and volatility prediction.
According to the district server configuration method based on data inrush current and volatility prediction, the resource response time of each partition can be obtained in the ith detection period, so that the access mode type of each partition is determined, whether the bandwidth resources and the processing resources configured by each partition can meet the access requirements or not is determined, if the access requirements are not met, configuration adjustment can be performed in time, and therefore the access requirements can be responded in time when the access data fluctuates, and the continuously-changing access requirements are met. In the process of determining the resource response time length, the method of solving the similarity after displacement and determining the maximum value of a plurality of similarities obtained after multiple displacements can solve the problem that the access quantity change is earlier than the time difference caused by the change of the processing resource occupation proportion, normalize the access quantity vector, solve the problem that the access quantity vector is different from the value range and unit of the processing resource occupation vector, and further eliminate the scale difference of different physical meanings by using the inverse matrix of the covariance matrix between the processing resource occupation vector and the displaced vector, thereby obtaining the first similarity with higher accuracy and improving the accuracy of the first response time length. And the maximum waveform similarity of the access quantity fluctuation function and the processing resource occupation fluctuation function can be solved by utilizing normalization to eliminate the difference of the value range and dimension of the processing resource occupation proportion and the access quantity and using a mode of integrating the product of the normalized access quantity fluctuation function and the processing resource occupation fluctuation function after waveform translation, so that the second response time length is obtained, and the accuracy of the second response time length is improved. And the first response time length and the second response time length can be weighted and summed, so that the first response time length and the second response time length are mutually checked, and the accuracy of processing the resource response time length is improved. Further, the access mode type can be determined, and configuration adjustment strategies can be set for various access mode types, so that access requirements can be met more accurately, and the utilization rate of bandwidth resources and processing resources can be improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed. Other features and aspects of the present invention will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the invention or the solutions of the prior art, the drawings which are necessary for the description of the embodiments or the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other embodiments may be obtained from these drawings without inventive effort to a person skilled in the art,
FIG. 1 schematically illustrates a flow diagram of a jurisdictional server configuration method based on data inrush and volatility prediction in accordance with an embodiment of the present invention;
FIG. 2 schematically illustrates a schematic diagram of a jurisdictional server configuration system based on data inrush and volatility prediction in accordance with an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The technical scheme of the invention is described in detail below by specific examples. The following embodiments may be combined with each other, and some embodiments may not be repeated for the same or similar concepts or processes.
FIG. 1 schematically illustrates a flow diagram of a jurisdictional server configuration method based on data inrush and volatility prediction, as shown in FIG. 1, according to an embodiment of the present invention, the method comprising:
step S101, in the ith detection period, processing resources and bandwidth resources of each partition of a district server are obtained, wherein i is a positive integer;
step S102, access quantity and processing resource occupation proportion of each partition in the ith detection period at a plurality of moments are obtained;
step S103, determining the response time length of the processing resources according to the access amount and the occupation proportion of the processing resources of each partition at a plurality of moments;
step S104, determining the access mode type of each partition according to the access amount and the processing resource occupation proportion of each partition at a plurality of moments and the processing resource response time length;
step S105, determining a configuration adjustment strategy of each partition in the (i+1) th detection period according to the access mode type of each partition, the processing resources and bandwidth resources of each partition in the (i) th detection period, and the access amount and the processing resource occupation proportion of each partition at a plurality of moments, wherein the configuration adjustment strategy comprises a bandwidth resource adjustment amount and a processing resource adjustment amount;
And S106, configuring the processing resources and the bandwidth resources of each partition in the (i+1) th detection period according to the configuration adjustment strategy.
According to the district server configuration method based on data inrush current and volatility prediction, the resource response time of each partition can be obtained in the ith detection period, so that the access mode type of each partition is determined, whether the bandwidth resources and the processing resources configured by each partition can meet the access requirements or not is determined, if the access requirements are not met, configuration adjustment can be performed in time, and therefore the access requirements can be responded in time when the access data fluctuates, and the continuously-changing access requirements are met.
According to one embodiment of the invention, the district server comprises a plurality of subareas, each subarea can be configured with a responsive processing resource and a bandwidth resource, the processing resource is used for responding to the instruction in the access data, carrying out operation processing to obtain a processing result corresponding to the instruction, and the bandwidth resource is used for receiving the access data and feeding back the processing result to a sender of the access data.
According to an embodiment of the present invention, in step S101, in the current ith detection period, processing resources and bandwidth resources configured by each partition of the district server may be obtained, so that the current configuration status of each partition may be determined, and when the configured resources cannot meet the access requirement and adjustment is performed, the current configuration status is used as a starting point of adjustment.
According to one embodiment of the present invention, in step S102, the access amount and the processing resource occupation ratio of each time may be collected at a plurality of times in the current ith detection period, so as to determine the resource usage situation of each partition in the current ith detection period, determine whether the resource configuration of each partition needs to be modified in the next detection period, and serve as a data basis for determining the adjustment amount for adjusting the resource configuration.
According to one embodiment of the present invention, in step S103, a processing resource response time period may be determined based on the access amount and the processing resource occupation ratio of each partition at each time in the current i-th detection period. The processing resource response duration may represent a duration between when the partition receives an access to the partition and responds to the access through its own processing resource, if the processing resource is sufficient, or the processing resource occupation amount of the access is smaller, the processing resource response duration is shorter, otherwise, if the processing resource is insufficient, or the processing resource occupation amount of the access is larger, the processing resource response duration is longer.
According to one embodiment of the present invention, step S103 may include: obtaining access quantity vectors of all the partitions according to the access quantity of all the partitions at a plurality of moments; obtaining the processing resource occupation vector of each partition according to the processing resource occupation proportion of each partition at a plurality of moments; determining a first response time length according to the access quantity vector, the processing resource occupation vector and the time interval among all the moments; obtaining access quantity fluctuation functions of all the partitions according to the access quantity of all the partitions at a plurality of moments; obtaining the processing resource occupation fluctuation function of each partition according to the processing resource occupation proportion of each partition at a plurality of moments; determining a second response time length according to the access quantity fluctuation function and the processing resource occupation fluctuation function; and determining the response time length of the processing resource according to the first response time length and the second response time length.
According to one embodiment of the present invention, the access amount vector of each partition may be determined according to the access amounts of each partition at a plurality of times, for example, the access amount of each time may be used as an element in the access amount vector, that is, the dimension of the access amount vector is equal to the number of the plurality of times, and the element of the access vector is the access amount of each time.
According to one embodiment of the present invention, the processing resource occupation vector of each partition is obtained according to the processing resource occupation ratios of each partition at a plurality of times, for example, the processing resource occupation ratio of each time may be used as an element in the processing resource occupation vector, that is, the dimension of the processing resource occupation vector is equal to the number of the plurality of times, and the element of the processing resource occupation vector is the processing resource occupation ratio of each time.
According to one embodiment of the invention, the first response time period may be determined based on the above access volume vector, the processing resource occupancy vector, and the time interval between the moments, i.e. the first response time period is determined based on the correlation between the access volume vector and the processing resource occupancy vector. However, the access amount vector and the processing resource occupation vector cannot directly calculate the similarity to determine the correlation, firstly, the units of the elements of the access amount vector and the processing resource occupation vector are inconsistent, the units of the elements of the access amount vector and the processing resource occupation vector are directly compared with each other, and have no physical meaning, and further, the fluctuation of the access amount vector and the processing resource occupation vector has a time difference, for example, when a certain partition receives an access, the access amount is increased immediately, and the processing resource occupation ratio needs a certain response time to be increased, so that even though the two have the correlation, the similarity is not high, and the correlation cannot be directly determined by using the similarity.
According to one embodiment of the present invention, to overcome the above-described problems, the present invention uses the following method to determine the correlation of the access volume vector and the processing resource occupancy vector, and thus the first response time period. Determining a first response duration according to the access amount vector, the processing resource occupation vector and the time interval between each moment, including:
obtaining a first similarity according to formula (1)
(1)
Wherein, for processing the resource occupation vector,for the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth detection period,/for the kth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the>For the access amount of the kth time in the access amount vector of the jth partition in the ith detection period, k is a positive integer less than or equal to n, n is the number of a plurality of times, +.>For the maximum value of the elements in the access quantity vector of the jth partition in the ith detection period, +.>For the ith detectionMinimum value of element in access quantity vector of jth partition in cycle,/>For the displacement matrix of the s-th step, s is an integer greater than or equal to 0 and less than or equal to n-1, the displacement matrix is represented by the following formula (2):
(2)
indicating ++in case the step number s is the kth step>Otherwise,/>In the case of a number of steps s, < > in >And->Covariance matrix between->Is->An inverse matrix of (a); determining a number of steps at which the first similarity reaches a maximum value; and determining the first response time length according to the step number when the first similarity reaches the maximum value and the time interval between the moments.
According to one embodiment of the present invention, in the formula (1), since the units of the access amount vector and the processing resource occupation vector are not identical, the elements in the access amount vector are positive integers, which represent accesses received at the same timeThe number of requests and the elements in the processing resource occupancy vector are proportional, e.g., in the form of a percentage, representing the ratio of the number of occupied resources to the total amount of resources in the processing resource, and therefore, the dimensions of the two may be unified first, e.g., the ratio between each element in the access volume vector, and the difference between the access volume maximum (i.e., the maximum of the elements in the access volume vector of the jth partition in the ith detection period) and the access volume minimum (i.e., the minimum of the elements in the access volume vector of the jth partition in the ith detection period), may be solved, and a vector consisting of the above ratios may be obtained Therefore, each element in the vector is also a ratio, and the value range and unit of the element in the access vector are consistent.
According to one embodiment of the present invention, although the above vectors are consistent with the range and unit of access vector, there are still problems that the physical meaning is inconsistent and the correlation cannot be directly determined. Therefore, directly solving the similarity between the vector and the processing resource occupation vector fails to take into account the similarity difference caused by the response time difference between the vector and the processing resource occupation vector, and the response time difference (i.e., the first response time duration) cannot be solved. Therefore, the above-mentioned vector can be shifted by using the shift matrix shown in the above formula (2), and after each shift, the similarity between the shifted vector and the processing resource occupation vector is calculated, and since the time when the access is received to cause the access amount to change is earlier than the time when the resource occupation ratio changes, the shift matrix shown in the above formula (2) can shift the element of the above-mentioned vector backward, that is, the first element in the vector before shift is shifted backward to become the second element in the vector after shift, the second element in the vector before shift is shifted backward to become the last element in the vector before shift of the third element … … in the vector after shift, the last element in the vector before shift is discarded, and the first element in the vector after shift is complemented with 0.
According to one embodiment of the present invention, after each step of displacement, the similarity between the displaced vector and the processing resource occupation vector may be calculated according to formula (1), and after n-1 displacements, n-1 similarities are obtained, where the maximum value of the n-1 similarities is the first similarity.
In an example, the 1 st row element of the displacement matrix is: line 1 and column 1 areThis means that when the number of steps of the displacement s is 0, i.e. no displacement is performed, the element is 1, otherwise 0, and the other elements on line 1 are all 0. The 2 nd row element of the displacement matrix is: row 2, column 1->Representing that when the number of steps of the displacement s is 1, the element is 1, otherwise 0, the element of row 2 and column 2 is +.>The remaining elements are 0. The 3 rd row element of the displacement matrix is: row 3, column 1->Representing that the element is 1 when the number of steps of the displacement s is 2, otherwise 0, row 3, column 2 is +.>Row 3 column 3 +.>The remaining elements are 0. Similarly, the nth row element of the forward displacement matrix is: column 1 of row n is->Representing that when the step number s of the displacement is n-1, the element is 1, otherwise 0, the nth row is listed as +.>Indicating whether or not the element is 1 when the number of steps s of the displacement is n-2Then the n-th column of the n-th row of 0 … … is +. >
According to the embodiment of the invention, the method for solving the similarity after the displacement and determining the maximum value of the similarity obtained after multiple displacements can solve the problem that the access quantity is changed earlier than the time difference caused by the change of the occupation proportion of processing resources, but the two problems still have inconsistent physical meanings. Therefore, in the process of solving the similarity by using the formula (1) after each step of displacement, the inverse matrix of the covariance matrix in the formula (1) can be used for eliminating the scale difference of vectors with different physical meanings, so as to obtain the statistical characteristic distance or difference of the vectors after the occupation vector of the processing resources and the displacement. And subtracting the characteristic distance or the difference from 1 to obtain the similarity of the processing resource occupation vector and the displaced vector, so that the maximum similarity after multiple displacements can be used as the first similarity.
According to one embodiment of the present invention, the above first similarity is a similarity calculated at each step of displacement after a plurality of steps of displacement, and the maximum value of the similarity is selected, so that the first similarity (i.e., the maximum value of the similarity) corresponds to the number of steps of a certain displacement, that is, the number of steps of the displacement is the number of steps when the first similarity reaches the maximum value, that is, the similarity between the processing resource occupation vector and the displaced vector reaches the maximum value at the time of the displacement.
According to an embodiment of the present invention, as described above, the dimensions of the processing resource occupation vector and the access volume vector are equal to the number of times in the i-th detection period, so that each data bit in the normalized access volume vector can correspond to one time, shifting by one step each time is equivalent to shifting all data in the normalized access volume vector by one time, so that the number of steps when the first similarity reaches the maximum value is equivalent to the number of times when the first similarity reaches the maximum value, and the number of times is multiplied by the time interval of each time, and the first response time length, that is, the time difference between the access volume variation caused by partition receiving access and the processing resource occupation ratio variation can be obtained.
In this way, the method of solving the similarity after displacement and determining the maximum value of the multiple similarities obtained after multiple displacements can solve the problem that the access quantity changes earlier than the time difference caused by the change of the occupation proportion of the processing resources, normalize the access quantity vector, solve the problem that the access quantity vector is different from the value range and unit of the processing resource occupation vector, and further eliminate the scale difference of different physical meanings by using the inverse matrix of the covariance matrix between the processing resource occupation vector and the displaced vector, thereby obtaining the first similarity with higher accuracy and improving the accuracy of the first response time length.
According to one embodiment of the invention, the above determines the first response time period based on the access amount vector and the processing resource occupancy vector. The second response time length can be determined in another angle, namely, the time difference between the access quantity change caused by the partition receiving access and the processing resource occupation proportion change is determined in another mode, and the time difference obtained in two modes is used for mutual verification, so that the accuracy of the time difference is improved.
According to one embodiment of the invention, the access amount fluctuation function of each partition can be obtained according to the access amount of each partition at a plurality of moments, for example, the access amount of each moment can be fitted to obtain the access amount fluctuation function, and the access amount fluctuation function can be used for describing the relationship between the access amount of the partition and time.
According to one embodiment of the invention, the processing resource occupation fluctuation function of each partition can be obtained according to the processing resource occupation proportion of each partition at a plurality of moments, for example, the processing resource occupation proportion of each moment can be fitted to obtain the processing resource occupation fluctuation function, and the processing resource occupation fluctuation function can be used for describing the relationship between the processing resource occupation proportion of the partition and time.
According to one embodiment of the invention, a second response time period may be determined based on the access amount fluctuation function and the processing resource occupancy fluctuation function, which stepMay include: obtaining a second similarity according to equation (3)
(3)
Wherein t is the time in the ith detection period,for the start time of the ith detection period, < +.>For the end time of the ith detection period, < +.>For the access volume fluctuation function +.>For processing resource occupancy fluctuation functions, +.>Is the displacement duration; determining a displacement time length when the second similarity reaches a maximum value; and taking the displacement time length when the second similarity reaches the maximum value as the second response time length.
According to one embodiment of the present invention, as described above, the value ranges and dimensions of the processing resource occupation ratios and the access amounts are different, and thus, the difference between the value ranges and dimensions can be eliminated using the ratio of the access amount fluctuation function to the difference between the access amount maximum value and the access amount minimum value.
According to one embodiment of the invention, since there is a time difference between the change of the access amount and the change of the proportion of occupied processing resources caused by the access of the partition, the waveform similarity of the normalized access amount fluctuation function and the waveform similarity of the processing resource occupation fluctuation function are not high, and the waveform similarity of the normalized access amount fluctuation function can be improved when the normalized access amount fluctuation function is translated, for example, the waveform elevation is translated to the same moment And/or translating the waveform dip to the same time, etc. In order to determine the maximum value of waveform similarity of the normalized access quantity fluctuation function after translation, the normalized access quantity fluctuation function can be translated by using the displacement time length, the translation quantity of the normalized access quantity fluctuation function after backward translation is equal to the displacement time length, and the integral maximum value of the product of the normalized access quantity fluctuation function and the processing resource occupation fluctuation function can be determined in the translation process, and the integral maximum value is utilizedThe multiplied arctangent function is normalized, so that the maximum waveform similarity, namely the second similarity, of the access quantity fluctuation function and the processing resource occupation fluctuation function can be obtained. Further, the translation amount (i.e., the displacement period) when the integrated maximum value is obtained is the displacement period when the second similarity reaches the maximum value. The displacement duration can be used as a time difference between the access amount change caused by the partition access acceptance and the change of the processing resource occupation proportion, namely, the second response duration.
In this way, the normalization can be used to eliminate the difference of the value range and dimension of the occupation proportion of the processing resources and the access quantity, and the maximum waveform similarity of the access quantity fluctuation function and the occupation fluctuation function of the processing resources is solved by integrating the product of the normalized access quantity fluctuation function and the normalized access quantity fluctuation function after the waveform is translated, so that the second response time is obtained, and the accuracy of the second response time is improved.
According to one embodiment of the present invention, the first response time period and the second response time period are obtained above, and may be checked against each other using both, and in an example, may be weighted and summed, and the result of the weighted and summed may be used as the processing resource response time period.
According to one embodiment of the present invention, determining the processing resource response time length according to the first response time length and the second response time length includes: obtaining the processing resource response time length according to the formula (4)
(4)
Wherein,for the first response time length, +.>And the second response time length is the second response time length.
According to one embodiment of the present invention, the first similarity represents a maximum value of the vector similarity obtained after normalization and displacement processing of the access amount vector, that is, a maximum value of the similarity representing the change of the access amount data and the change of the resource occupation ratio caused by the partition receiving access, and the second similarity represents a maximum value of the waveform similarity obtained after normalization and displacement processing of the access amount fluctuation function, that is, a maximum value of the similarity representing the fluctuation of the access amount and the fluctuation of the resource occupation ratio caused by the partition receiving access. Both can represent the relevance between the access quantity and the resource occupation proportion caused by partition receiving access, so that the proportion of the partition receiving access to the sum of the partition receiving access and the resource occupation proportion can be used as weight, namely, the ratio between the first similarity and the sum of the first similarity and the second similarity is used as the weight of the first response time length, the ratio between the second similarity and the sum of the first similarity and the second similarity is used as the weight of the second response time length, and therefore the first response time length and the second response time length are weighted and summed, the first response time length and the second response time length are mutually checked, and the accuracy of processing the resource response time length is improved.
According to an embodiment of the present invention, in step S104, the access mode type of each partition may be determined based on the access amounts and the processing resource occupation ratios of each partition at a plurality of times, and the processing resource response time obtained above. For example, it may be determined whether the number of accesses received by a partition is large and/or whether the processing resources that need to be occupied are large. It may then be determined whether the configuration of the partition is sufficient to meet the access requirements.
According to one embodiment of the present invention, step S104 may include one of the following: when the maximum value of the access quantity at a plurality of moments is larger than or equal to the access quantity threshold value, the maximum value of the occupation proportion of the processing resources is smaller than the occupation proportion threshold value, and the response time of the processing resources is smaller than the time threshold value, determining the access mode as a low-occupation high-frequency access mode; determining an access mode as a high-occupancy low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than an access amount threshold, the maximum processing resource occupancy proportion is smaller than an occupancy proportion threshold, and the processing resource response length is larger than or equal to a duration threshold; determining an access mode as a high-occupancy high-frequency access mode under the condition that the maximum access amount at a plurality of moments is larger than or equal to an access amount threshold value, the maximum processing resource occupancy proportion is larger than or equal to an occupancy proportion threshold value, and the processing resource response length is larger than or equal to a duration threshold value; and determining the access mode as a low-occupancy and low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than the access amount threshold, the maximum processing resource occupancy proportion is smaller than the occupancy proportion threshold and the processing resource response time length is smaller than the time length threshold.
According to one embodiment of the invention, in the low-occupation high-frequency access mode, the access amount accepted by the partition is larger, but the amount of processing resources occupied by each access is smaller, each access request can be rapidly processed by the processing resources of the partition, so that the occupation amount of the processing resources is smaller, and the response time of the processing resources is shorter. Of course, the proportion of the accumulated processing resources occupied by the huge access amount is larger and even exceeds the proportion threshold value due to the huge access amount, but the processing resources of single access are smaller, so that even if the access amount is huge, the access requests can be rapidly processed by the processing resources of the partition, and the response time of the processing resources is smaller.
According to one embodiment of the invention, in the high-occupation low-frequency access mode, the access amount accepted by the partition is smaller, but the amount of processing resources occupied by each access is larger, so that the occupation proportion of the processing resources is larger, and the processing resources are difficult to respond to all the accesses in time, so that the response time of the processing resources is longer.
According to one embodiment of the invention, in a high-occupation high-frequency access mode, the access amount accepted by the partition is larger, the amount of processing resources occupied by each access is larger, so that the occupation proportion of the processing resources is larger, the processing resources are difficult to respond to all accesses in time, and the response time of the processing resources is longer. Of course, although the access amount is large and the amount of processing resources occupied by each access is large, the processing resource response time is short, and the access amount is relatively uniform in time due to the large access amount and the single access of the processing resources, and the situation of access congestion does not occur, so that the situation that the processing of the processing resources is not performed for a long time by the instructions in some accesses does not occur, but in this case, there is still a risk of insufficient processing resources or bandwidth resources.
According to one embodiment of the invention, in the low-occupation low-frequency access mode, the access amount accepted by the partition is smaller, the processing resource amount occupied by each access is smaller, each access request can be rapidly processed by the processing resource of the partition, the occupation amount of the processing resource is smaller, and the response time of the processing resource is shorter.
According to one embodiment of the present invention, after determining the access pattern type in the above manner, in step S105, a configuration adjustment policy of each partition in the (i+1) th detection period may be determined according to the access pattern type of each partition, the processing resource and the bandwidth resource of each partition in the (i) th detection period, and the access amount and the processing resource occupation ratio. This step may include: determining a processing resource adjustment amount according to the difference between the maximum value of the processing resource occupation proportion and the threshold value of the occupation proportion and the processing resource of the ith detection period, and determining a bandwidth resource adjustment amount according to the difference between the maximum value of the access amount and the threshold value of the access amount at a plurality of moments and the bandwidth resource of the ith detection period, wherein the bandwidth resource adjustment amount is increased in a low-occupation high-frequency access mode, the processing resource adjustment amount is decreased in a high-occupation low-frequency access mode, the processing resource adjustment amount is increased in a high-occupation low-frequency access mode, the bandwidth resource adjustment amount is decreased in a high-occupation high-frequency access mode, the processing resource adjustment amount and the bandwidth resource adjustment amount are increased in a high-occupation high-frequency access mode, and the processing resource adjustment amount and the bandwidth resource adjustment amount are decreased in a low-occupation low-frequency access mode.
In an example, in the low-occupancy high-frequency access mode, the amount of access to be received is larger, the access frequency is higher, the bandwidth resources are more intense, but the occupancy proportion of the processing resources is lower, and the processing resources have a margin, so that the bandwidth resource adjustment amount is an increment, and the processing resource adjustment amount is a decrement. The bandwidth resource adjustment amount may be determined based on a difference between the access amount maximum value and the access amount threshold value, for example, a ratio of the difference to the access amount threshold value is 60%, and the bandwidth resource of the i+1th detection period may be increased by 60%, and the increased ratio may be multiplied by the bandwidth resource of the i detection period, thereby obtaining the bandwidth resource adjustment amount. The processing resource adjustment amount may be determined based on a difference between a maximum value of the processing resource occupation ratio and an occupation ratio threshold, for example, the maximum value of the processing resource occupation ratio is 30%, the occupation ratio threshold is 50%, the difference is 20%, and a ratio of the difference to the occupation ratio threshold is 40%, the processing resource of the i+1th detection period may be reduced by 40%, and the reduced ratio may be multiplied by the processing resource of the i detection period, thereby obtaining the processing resource adjustment amount.
In an example, in the high-occupancy low-frequency access mode, the amount of access to be received is smaller, the access frequency is lower, and a margin exists in the bandwidth resource, but the occupancy proportion of the processing resource is higher, and the processing resource is more intense, so that the adjustment amount of the processing resource is an increment, and the adjustment amount of the bandwidth resource is a decrement. The bandwidth resource adjustment amount may be determined based on a difference between the access amount maximum value and the access amount threshold value, for example, a ratio of the difference to the access amount threshold value is 40%, and the bandwidth resource of the i+1th detection period may be reduced by 40%, and the reduced ratio may be multiplied by the bandwidth resource of the i detection period, thereby obtaining the bandwidth resource adjustment amount. The processing resource adjustment amount may be determined based on a difference between a maximum value of the processing resource occupation ratio and an occupation ratio threshold, for example, the maximum value of the processing resource occupation ratio is 80%, the occupation ratio threshold is 50%, the difference is 30%, and the ratio of the difference to the occupation ratio threshold is 60%, and then the processing resource of the i+1th detection period may be increased by 60%, and the increased ratio may be multiplied by the processing resource of the i detection period, thereby obtaining the processing resource adjustment amount.
In an example, in a high-occupancy high-frequency access mode, the amount of access to be received is larger, the access frequency is higher, the bandwidth resources are more intense, and the occupancy proportion of the processing resources is higher, so that the processing resource adjustment amount and the bandwidth resource adjustment amount are both incremental. The bandwidth resource adjustment amount may be determined based on a difference between the access amount maximum value and the access amount threshold value, for example, a ratio of the difference to the access amount threshold value is 60%, and the bandwidth resource of the i+1th detection period may be increased by 60%, and the increased ratio may be multiplied by the bandwidth resource of the i detection period, thereby obtaining the bandwidth resource adjustment amount. The processing resource adjustment amount may be determined based on a difference between a maximum value of the processing resource occupation ratio and an occupation ratio threshold, for example, the maximum value of the processing resource occupation ratio is 80%, the occupation ratio threshold is 50%, the difference is 30%, and the ratio of the difference to the occupation ratio threshold is 60%, and then the processing resource of the i+1th detection period may be increased by 60%, and the increased ratio may be multiplied by the processing resource of the i detection period, thereby obtaining the processing resource adjustment amount.
In an example, in the low-occupancy low-frequency access mode, the amount of access to be received is small, the access frequency is low, the bandwidth resource has a margin, the occupancy proportion of the processing resource is low, and the processing resource has a margin, so that the processing resource adjustment amount and the bandwidth resource adjustment amount are both reduced. The bandwidth resource adjustment amount may be determined based on a difference between the access amount maximum value and the access amount threshold value, for example, a ratio of the difference to the access amount threshold value is 40%, and the bandwidth resource of the i+1th detection period may be reduced by 40%, and the reduced ratio may be multiplied by the bandwidth resource of the i detection period, thereby obtaining the bandwidth resource adjustment amount. The processing resource adjustment amount may be determined based on a difference between a maximum value of the processing resource occupation ratio and an occupation ratio threshold, for example, the maximum value of the processing resource occupation ratio is 30%, the occupation ratio threshold is 50%, the difference is 20%, and a ratio of the difference to the occupation ratio threshold is 40%, the processing resource of the i+1th detection period may be reduced by 40%, and the reduced ratio may be multiplied by the processing resource of the i detection period, thereby obtaining the processing resource adjustment amount.
According to one embodiment of the present invention, if the configuration adjustment amount of a certain partition is an increment, the increment may be from a partition whose configuration adjustment amount is a decrement, for example, if the processing resource adjustment amount of a certain partition is an increment, the processing resource adjustment amount increased by the partition is from the processing resource adjustment amount decreased by other partitions, and similarly, if the bandwidth resource adjustment amount of a certain partition is an increment, the bandwidth resource adjustment amount increased by the partition is from the bandwidth resource adjustment amount decreased by other partitions. However, if the total amount of configuration provided by the partition with the reduced configuration adjustment is insufficient to satisfy the configuration increment requirement with all the configuration adjustment being increased, the total amount of configuration may be allocated in proportion, for example, the processing resource provided by the partition with the reduced configuration adjustment is 100GB, the adjusted processing resource required by the 2 partitions with the increased configuration adjustment is 120GB and 80GB, respectively, the increment requirement of the processing resource is 200GB in total, the processing resource provided by the partition is insufficient to satisfy the increment requirement of the processing resource, and thus the processing resource provided by the partition may be allocated in proportion, that is, the processing resource provided by the first partition is 60GB, and the processing resource provided by the second partition is 40GB.
According to one embodiment of the present invention, in step S106, bandwidth resources and processing resources of each partition may be adjusted according to the above configuration adjustment policy, and access requests of each partition may be processed in the i+1th detection period using the adjusted processing resources and bandwidth resources.
According to the district server configuration method based on data inrush current and volatility prediction, the resource response time of each partition can be obtained in the ith detection period, so that the access mode type of each partition is determined, whether the bandwidth resources and the processing resources configured by each partition can meet the access requirements or not is determined, if the access requirements are not met, configuration adjustment can be performed in time, and therefore the access requirements can be responded in time when the access data fluctuates, and the continuously-changing access requirements are met. In the process of determining the resource response time length, the method of solving the similarity after displacement and determining the maximum value of a plurality of similarities obtained after multiple displacements can solve the problem that the access quantity change is earlier than the time difference caused by the change of the processing resource occupation proportion, normalize the access quantity vector, solve the problem that the access quantity vector is different from the value range and unit of the processing resource occupation vector, and further eliminate the scale difference of different physical meanings by using the inverse matrix of the covariance matrix between the processing resource occupation vector and the displaced vector, thereby obtaining the first similarity with higher accuracy and improving the accuracy of the first response time length. And the maximum waveform similarity of the access quantity fluctuation function and the processing resource occupation fluctuation function can be solved by utilizing normalization to eliminate the difference of the value range and dimension of the processing resource occupation proportion and the access quantity and using a mode of integrating the product of the normalized access quantity fluctuation function and the processing resource occupation fluctuation function after waveform translation, so that the second response time length is obtained, and the accuracy of the second response time length is improved. And the first response time length and the second response time length can be weighted and summed, so that the first response time length and the second response time length are mutually checked, and the accuracy of processing the resource response time length is improved. Further, the access mode type can be determined, and configuration adjustment strategies can be set for various access mode types, so that access requirements can be met more accurately, and the utilization rate of bandwidth resources and processing resources can be improved.
FIG. 2 schematically illustrates a schematic diagram of a jurisdictional server configuration system based on data inrush and volatility prediction, as shown in FIG. 2, including:
a configuration obtaining module 101, configured to obtain, in an ith detection period, processing resources and bandwidth resources of each partition of the district server, where i is a positive integer;
a state obtaining module 102, configured to obtain access amounts and processing resource occupation ratios of each partition in the ith detection period at multiple times;
a resource response duration module 103, configured to determine a processing resource response duration according to the access amounts and the processing resource occupation ratios of the respective partitions at multiple times;
an access mode type module 104, configured to determine an access mode type of each partition according to the access amounts and the occupation ratios of processing resources of each partition at multiple times, and the response time of the processing resources;
a configuration adjustment policy module 105, configured to determine a configuration adjustment policy of each partition in the (i+1) th detection period according to the access mode type of each partition, the processing resource and the bandwidth resource of each partition in the (i) th detection period, and the access amounts and the processing resource occupation ratios of the multiple moments, where the configuration adjustment policy includes a bandwidth resource adjustment amount and a processing resource adjustment amount;
And the configuration module 106 is configured to configure the processing resources and the bandwidth resources of the i+1st detection period of each partition according to the configuration adjustment policy.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
obtaining access quantity vectors of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation vector of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a first response time length according to the access quantity vector, the processing resource occupation vector and the time interval among all the moments;
obtaining access quantity fluctuation functions of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation fluctuation function of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a second response time length according to the access quantity fluctuation function and the processing resource occupation fluctuation function;
and determining the response time length of the processing resource according to the first response time length and the second response time length.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
According to the formula
Obtaining a first similarityWherein, for processing the resource occupancy vector, +.>For the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth detection period,/for the kth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the>For the access amount of the kth time in the access amount vector of the jth partition in the ith detection period, k is a positive integer less than or equal to n, n is the number of a plurality of times, +.>For the maximum value of the elements in the access quantity vector of the jth partition in the ith detection period, +.>For the minimum value of the elements in the access quantity vector of the jth partition in the ith detection period,/->Is a displacement matrix of the step s, s is an integer which is more than or equal to 0 and less than or equal to n-1,
indicating ++in case the step number s is the kth step>Otherwise,/>In the case of a number of steps s, < > in>And->Covariance matrix between->Is->An inverse matrix of (a);
determining a number of steps at which the first similarity reaches a maximum value;
and determining the first response time length according to the step number when the first similarity reaches the maximum value and the time interval between the moments.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
according to the formula
Obtaining a second degree of similarityWherein t is the time in the ith detection period,/and>for the start time of the ith detection period, < +.>For the end time of the ith detection period, < +.>For the access volume fluctuation function +.>For processing resource occupancy fluctuation functions, +.>Is the displacement duration;
determining a displacement time length when the second similarity reaches a maximum value;
and taking the displacement time length when the second similarity reaches the maximum value as the second response time length.
According to one embodiment of the present invention, the resource response time duration module is further configured to:
according to the formula
Obtaining the processing resource response time lengthWherein->For the first response time length, +.>And the second response time length is the second response time length.
According to one embodiment of the invention, the access pattern type module is further adapted to one of:
when the maximum value of the access quantity at a plurality of moments is larger than or equal to the access quantity threshold value, the maximum value of the occupation proportion of the processing resources is smaller than the occupation proportion threshold value, and the response time of the processing resources is smaller than the time threshold value, determining the access mode as a low-occupation high-frequency access mode;
determining an access mode as a high-occupancy low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than an access amount threshold, the maximum processing resource occupancy proportion is smaller than an occupancy proportion threshold, and the processing resource response length is larger than or equal to a duration threshold;
Determining an access mode as a high-occupancy high-frequency access mode under the condition that the maximum access amount at a plurality of moments is larger than or equal to an access amount threshold value, the maximum processing resource occupancy proportion is larger than or equal to an occupancy proportion threshold value, and the processing resource response length is larger than or equal to a duration threshold value;
and determining the access mode as a low-occupancy and low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than the access amount threshold, the maximum processing resource occupancy proportion is smaller than the occupancy proportion threshold and the processing resource response time length is smaller than the time length threshold.
According to one embodiment of the invention, the configuration adjustment policy module is further configured to:
determining a processing resource adjustment amount according to the difference between the maximum value of the processing resource occupation proportion and the threshold value of the occupation proportion and the processing resource of the ith detection period, and determining a bandwidth resource adjustment amount according to the difference between the maximum value of the access amount and the threshold value of the access amount at a plurality of moments and the bandwidth resource of the ith detection period, wherein the bandwidth resource adjustment amount is increased in a low-occupation high-frequency access mode, the processing resource adjustment amount is decreased in a high-occupation low-frequency access mode, the processing resource adjustment amount is increased in a high-occupation low-frequency access mode, the bandwidth resource adjustment amount is decreased in a high-occupation high-frequency access mode, the processing resource adjustment amount and the bandwidth resource adjustment amount are increased in a high-occupation high-frequency access mode, and the processing resource adjustment amount and the bandwidth resource adjustment amount are decreased in a low-occupation low-frequency access mode.
According to one embodiment of the present invention, there is provided a jurisdictional server configuration apparatus based on data inrush and volatility prediction, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored by the memory to perform the jurisdictional server configuration method based on data inrush and volatility prediction.
According to one embodiment of the present invention, a computer readable storage medium having stored thereon computer program instructions which when executed by a processor implement the jurisdictional server configuration method based on data inrush current and volatility prediction.
The present invention may be a method, apparatus, system, and/or computer program product. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for performing various aspects of the present invention.
It will be appreciated by persons skilled in the art that the embodiments of the invention described above and shown in the drawings are by way of example only and are not limiting. The objects of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been shown and described in the examples and embodiments of the invention may be modified or practiced without departing from the principles described.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.

Claims (10)

1. A jurisdictional server configuration method based on data inrush current and volatility prediction, comprising:
in the ith detection period, processing resources and bandwidth resources of each partition of the district server are obtained, wherein i is a positive integer;
the access quantity and the processing resource occupation proportion of each partition in the ith detection period at a plurality of moments are obtained;
determining the response time length of the processing resources according to the access amount and the occupation proportion of the processing resources of each partition at a plurality of moments;
determining the access mode type of each partition according to the access amount and the occupation proportion of the processing resources of each partition at a plurality of moments and the response time length of the processing resources;
Determining a configuration adjustment strategy of each partition in the (i+1) th detection period according to the access mode type of each partition, the processing resources and bandwidth resources of each partition in the (i) th detection period, and the access amount and the processing resource occupation proportion of each partition at a plurality of moments, wherein the configuration adjustment strategy comprises a bandwidth resource adjustment amount and a processing resource adjustment amount;
and according to the configuration adjustment strategy, processing resources and bandwidth resources of the partitions in the (i+1) th detection period are configured.
2. The jurisdictional server configuration method based on data inrush current and volatility prediction of claim 1, wherein determining a processing resource response duration based on the access amounts and the processing resource occupancy ratios of the respective partitions at a plurality of times comprises:
obtaining access quantity vectors of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation vector of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a first response time length according to the access quantity vector, the processing resource occupation vector and the time interval among all the moments;
Obtaining access quantity fluctuation functions of all the partitions according to the access quantity of all the partitions at a plurality of moments;
obtaining the processing resource occupation fluctuation function of each partition according to the processing resource occupation proportion of each partition at a plurality of moments;
determining a second response time length according to the access quantity fluctuation function and the processing resource occupation fluctuation function;
and determining the response time length of the processing resource according to the first response time length and the second response time length.
3. The jurisdictional server configuration method based on data inrush and volatility prediction of claim 2, wherein determining a first response time period based on the access volume vector, the processing resource occupancy vector, and a time interval between respective moments comprises:
according to the formula
Obtaining a first similarityWherein, for processing the resource occupancy vector, +.>For the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the processing resource occupation vector of the jth detection period,/for the kth partition in the ith detection period,/for the processing resource occupation ratio of the kth moment in the>For the access amount of the kth time in the access amount vector of the jth partition in the ith detection period, k is a positive integer less than or equal to n, n is the number of a plurality of times, +. >For the maximum value of the elements in the access quantity vector of the jth partition in the ith detection period, +.>For the minimum value of the elements in the access quantity vector of the jth partition in the ith detection period,/->Is a displacement matrix of the step s, s is an integer which is more than or equal to 0 and less than or equal to n-1,
indicating ++in case the step number s is the kth step>Otherwise->,/>In the case of a number of steps s, < > in>And->Covariance matrix between->Is->An inverse matrix of (a);
determining a number of steps at which the first similarity reaches a maximum value;
and determining the first response time length according to the step number when the first similarity reaches the maximum value and the time interval between the moments.
4. The jurisdictional server configuration method based on data inrush current and volatility prediction of claim 3, wherein determining a second response time period based on the access volume fluctuation function and the processing resource occupancy fluctuation function comprises:
according to the formula
Obtaining a second degree of similarityWherein t is the time in the ith detection period,/and>for the start time of the ith detection period, < +.>For the end time of the ith detection period, < +.>For the access volume fluctuation function +.>For processing resource occupancy fluctuation functions, +. >Is the displacement duration;
determining a displacement time length when the second similarity reaches a maximum value;
and taking the displacement time length when the second similarity reaches the maximum value as the second response time length.
5. The jurisdictional server configuration method based on data inrush current and volatility prediction of claim 4, wherein determining the processing resource response time period from the first response time period and the second response time period comprises:
according to the formula
Obtaining the processing resource response time lengthWherein->For the first response time length, +.>And the second response time length is the second response time length.
6. The jurisdictional server configuration method based on data inrush current and volatility prediction of claim 1, wherein determining the access pattern type of each partition according to the access amount and the processing resource occupation ratio of each partition at a plurality of times and the processing resource response time length comprises one of:
when the maximum value of the access quantity at a plurality of moments is larger than or equal to the access quantity threshold value, the maximum value of the occupation proportion of the processing resources is smaller than the occupation proportion threshold value, and the response time of the processing resources is smaller than the time threshold value, determining the access mode as a low-occupation high-frequency access mode;
Determining an access mode as a high-occupancy low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than an access amount threshold, the maximum processing resource occupancy proportion is smaller than an occupancy proportion threshold, and the processing resource response length is larger than or equal to a duration threshold;
determining an access mode as a high-occupancy high-frequency access mode under the condition that the maximum access amount at a plurality of moments is larger than or equal to an access amount threshold value, the maximum processing resource occupancy proportion is larger than or equal to an occupancy proportion threshold value, and the processing resource response length is larger than or equal to a duration threshold value;
and determining the access mode as a low-occupancy and low-frequency access mode under the condition that the maximum access amount at a plurality of moments is smaller than the access amount threshold, the maximum processing resource occupancy proportion is smaller than the occupancy proportion threshold and the processing resource response time length is smaller than the time length threshold.
7. The jurisdictional server configuration method based on data inrush current and volatility prediction of claim 6, wherein determining a configuration adjustment policy for each partition at the i+1 th detection period according to the access pattern type of each partition, the processing resources and bandwidth resources of each partition at the i detection period, and the access amount and the processing resource occupation ratio of the plurality of times, comprises:
Determining a processing resource adjustment amount according to the difference between the maximum value of the processing resource occupation proportion and the threshold value of the occupation proportion and the processing resource of the ith detection period, and determining a bandwidth resource adjustment amount according to the difference between the maximum value of the access amount and the threshold value of the access amount at a plurality of moments and the bandwidth resource of the ith detection period, wherein the bandwidth resource adjustment amount is increased in a low-occupation high-frequency access mode, the processing resource adjustment amount is decreased in a high-occupation low-frequency access mode, the processing resource adjustment amount is increased in a high-occupation low-frequency access mode, the bandwidth resource adjustment amount is decreased in a high-occupation high-frequency access mode, the processing resource adjustment amount and the bandwidth resource adjustment amount are increased in a high-occupation high-frequency access mode, and the processing resource adjustment amount and the bandwidth resource adjustment amount are decreased in a low-occupation low-frequency access mode.
8. A jurisdictional server configuration system based on data inrush and volatility prediction, comprising:
the configuration acquisition module is used for acquiring processing resources and bandwidth resources of each partition of the district server in the ith detection period, wherein i is a positive integer;
The state acquisition module is used for acquiring the access quantity and the processing resource occupation proportion of each partition in the ith detection period at a plurality of moments;
the resource response time length module is used for determining the response time length of the processing resources according to the access quantity and the occupation proportion of the processing resources of each partition at a plurality of moments;
the access mode type module is used for determining the access mode type of each partition according to the access quantity and the processing resource occupation proportion of each partition at a plurality of moments and the processing resource response time length;
the configuration adjustment strategy module is used for determining a configuration adjustment strategy of each partition in the (i+1) th detection period according to the access mode type of each partition, the processing resources and bandwidth resources of each partition in the (i) th detection period, the access quantity and the processing resource occupation proportion of the plurality of moments, wherein the configuration adjustment strategy comprises a bandwidth resource adjustment quantity and a processing resource adjustment quantity;
and the configuration module is used for configuring the processing resources and the bandwidth resources of each partition in the (i+1) th detection period according to the configuration adjustment strategy.
9. A jurisdictional server configuration apparatus based on data inrush and volatility prediction, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored in the memory to perform the method of any of claims 1-7.
10. A computer readable storage medium, having stored thereon computer program instructions which, when executed by a processor, implement the method of any of claims 1-7.
CN202311368351.3A 2023-10-23 2023-10-23 Jurisdictional server configuration method and system based on data inrush current and volatility prediction Active CN117112236B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311368351.3A CN117112236B (en) 2023-10-23 2023-10-23 Jurisdictional server configuration method and system based on data inrush current and volatility prediction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311368351.3A CN117112236B (en) 2023-10-23 2023-10-23 Jurisdictional server configuration method and system based on data inrush current and volatility prediction

Publications (2)

Publication Number Publication Date
CN117112236A true CN117112236A (en) 2023-11-24
CN117112236B CN117112236B (en) 2024-02-20

Family

ID=88809461

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311368351.3A Active CN117112236B (en) 2023-10-23 2023-10-23 Jurisdictional server configuration method and system based on data inrush current and volatility prediction

Country Status (1)

Country Link
CN (1) CN117112236B (en)

Citations (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6263359B1 (en) * 1997-05-22 2001-07-17 International Business Machines Corporation Computer resource proportional utilization and response time scheduling
US20080167977A1 (en) * 2007-01-04 2008-07-10 Auvenshine John J Determining proportionate use of system resources by applications executing in a shared hosting environment
CN101695034A (en) * 2009-10-09 2010-04-14 清华大学 Device and method for adjusting monitoring parameter of WEB service
CN104283946A (en) * 2014-09-26 2015-01-14 东北大学 Resource self-adaptive adjusting system and method of multiple virtual machines under single physical machine
US20160292608A1 (en) * 2015-04-03 2016-10-06 Alibaba Group Holding Limited Multi-cluster management method and device
CN107026877A (en) * 2016-01-29 2017-08-08 华为技术有限公司 The method and apparatus that resource is managed in cloud platform
US20180109469A1 (en) * 2016-10-17 2018-04-19 International Business Machines Corporation Systems and methods for controlling process priority for efficient resource allocation
CN109783219A (en) * 2017-11-10 2019-05-21 北京信息科技大学 A kind of cloud resource Optimization Scheduling and device
CN111343244A (en) * 2018-08-23 2020-06-26 贵州白山云科技股份有限公司 Cloud distribution network scheduling method and device and computer readable storage medium
CN111625859A (en) * 2020-05-20 2020-09-04 北京百度网讯科技有限公司 Resource access control method and device, electronic equipment and storage medium
US20200382582A1 (en) * 2019-05-05 2020-12-03 Wangsu Science & Technology Co., Ltd. Resource scheduling method and system
CN112131005A (en) * 2020-09-25 2020-12-25 新华三大数据技术有限公司 Resource adjustment strategy determination method and device
CN112416590A (en) * 2020-11-24 2021-02-26 平安普惠企业管理有限公司 Server system resource adjusting method and device, computer equipment and storage medium
CN114205232A (en) * 2021-11-01 2022-03-18 湖北省楚天云有限公司 Interface service platform capable of automatically expanding capacity
JP7103705B1 (en) * 2021-12-21 2022-07-20 北京穿楊科技有限公司 Cluster-based capacity reduction processing method and equipment
CN115002044A (en) * 2022-05-26 2022-09-02 平安银行股份有限公司 Method and device for controlling data transmission, computer equipment and storage medium
CN115134368A (en) * 2022-08-31 2022-09-30 中信建投证券股份有限公司 Load balancing method, device, equipment and storage medium
US20220329541A1 (en) * 2019-12-26 2022-10-13 Huawei Cloud Computing Technologies Co., Ltd. Bandwidth Scheduling Method, Traffic Transmission Method, and Related Product
CN115202889A (en) * 2022-09-13 2022-10-18 阿里巴巴(中国)有限公司 Computing resource adjusting method and computing system

Patent Citations (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6263359B1 (en) * 1997-05-22 2001-07-17 International Business Machines Corporation Computer resource proportional utilization and response time scheduling
US20080167977A1 (en) * 2007-01-04 2008-07-10 Auvenshine John J Determining proportionate use of system resources by applications executing in a shared hosting environment
CN101695034A (en) * 2009-10-09 2010-04-14 清华大学 Device and method for adjusting monitoring parameter of WEB service
CN104283946A (en) * 2014-09-26 2015-01-14 东北大学 Resource self-adaptive adjusting system and method of multiple virtual machines under single physical machine
US20160292608A1 (en) * 2015-04-03 2016-10-06 Alibaba Group Holding Limited Multi-cluster management method and device
CN107026877A (en) * 2016-01-29 2017-08-08 华为技术有限公司 The method and apparatus that resource is managed in cloud platform
US20180109469A1 (en) * 2016-10-17 2018-04-19 International Business Machines Corporation Systems and methods for controlling process priority for efficient resource allocation
CN109783219A (en) * 2017-11-10 2019-05-21 北京信息科技大学 A kind of cloud resource Optimization Scheduling and device
CN111343244A (en) * 2018-08-23 2020-06-26 贵州白山云科技股份有限公司 Cloud distribution network scheduling method and device and computer readable storage medium
US20200382582A1 (en) * 2019-05-05 2020-12-03 Wangsu Science & Technology Co., Ltd. Resource scheduling method and system
US20220329541A1 (en) * 2019-12-26 2022-10-13 Huawei Cloud Computing Technologies Co., Ltd. Bandwidth Scheduling Method, Traffic Transmission Method, and Related Product
CN111625859A (en) * 2020-05-20 2020-09-04 北京百度网讯科技有限公司 Resource access control method and device, electronic equipment and storage medium
CN112131005A (en) * 2020-09-25 2020-12-25 新华三大数据技术有限公司 Resource adjustment strategy determination method and device
CN112416590A (en) * 2020-11-24 2021-02-26 平安普惠企业管理有限公司 Server system resource adjusting method and device, computer equipment and storage medium
CN114205232A (en) * 2021-11-01 2022-03-18 湖北省楚天云有限公司 Interface service platform capable of automatically expanding capacity
JP7103705B1 (en) * 2021-12-21 2022-07-20 北京穿楊科技有限公司 Cluster-based capacity reduction processing method and equipment
CN115002044A (en) * 2022-05-26 2022-09-02 平安银行股份有限公司 Method and device for controlling data transmission, computer equipment and storage medium
CN115134368A (en) * 2022-08-31 2022-09-30 中信建投证券股份有限公司 Load balancing method, device, equipment and storage medium
CN115202889A (en) * 2022-09-13 2022-10-18 阿里巴巴(中国)有限公司 Computing resource adjusting method and computing system

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
YANG, ZZ等: "Dynamic Configuration of Reverse Proxy Cache based on Multi-Dimensional Time Series Prediction of Visit Traffic", PROCEEDINGS OF 2016 IEEE 7TH INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING AND SERVICE SCIENCE (ICSESS 2016) *
姚兆凡: "轻量级Web服务器Nginx的研究与优化", 中国优秀硕士学位论文全文数据库 信息科技辑 *
王健宗;谌炎俊;谢长生;: "面向云存储的I/O资源效用优化调度算法研究", 计算机研究与发展, no. 08 *

Also Published As

Publication number Publication date
CN117112236B (en) 2024-02-20

Similar Documents

Publication Publication Date Title
US20220253214A1 (en) Data processing method, apparatus, device, and readable storage medium
CN109408590B (en) Method, device and equipment for expanding distributed database and storage medium
CN110636388A (en) Service request distribution method, system, electronic equipment and storage medium
CN109407997B (en) Data processing method, device and equipment and readable storage medium
CN107480771B (en) Deep learning-based activation function realization method and device
WO2022257389A1 (en) Task delay risk detection method, electronic device and storage medium
CN104932898A (en) Method for selecting to-be-increased components based on improved multi-target particle swam optimization algorithm
CN117112236B (en) Jurisdictional server configuration method and system based on data inrush current and volatility prediction
CN112506399B (en) Information publishing method of book reading page, computing device and storage medium
CN114090204A (en) Resource prediction method and device
CN113342474A (en) Method, device and storage medium for forecasting customer flow and training model
CN110287014B (en) Method, system and storage medium for scheduling memory bandwidth of computer system
CN113452546A (en) Dynamic quality of service management for deep learning training communications
CN116150067A (en) Bandwidth adjustment method, device, electronic equipment and storage medium
CN110069340B (en) Thread number evaluation method and device
CN115499513A (en) Data request processing method and device, computer equipment and storage medium
CN110377533B (en) Cache replacement method and device for cloud computing and edge computing
CN110134575B (en) Method and device for calculating service capacity of server cluster
CN111767137B (en) System deployment method and device, electronic equipment and storage medium
CN115114028B (en) Task allocation method and device for secondary control of electric power simulation
CN113282405B (en) Load adjustment optimization method and terminal
CN108270851B (en) Data storage method and device
CN113822445B (en) Model integrated prediction method, system, electronic equipment and storage medium
CN117744466A (en) Self-adaptive agent model synchronous construction method suitable for flood multi-element prediction
CN112684972B (en) Data storage method and device based on distributed block storage

Legal Events

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