CN111767137B - System deployment method and device, electronic equipment and storage medium - Google Patents
System deployment method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111767137B CN111767137B CN202010495898.XA CN202010495898A CN111767137B CN 111767137 B CN111767137 B CN 111767137B CN 202010495898 A CN202010495898 A CN 202010495898A CN 111767137 B CN111767137 B CN 111767137B
- Authority
- CN
- China
- Prior art keywords
- access
- unit time
- service
- average
- preset
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Mathematical Physics (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Debugging And Monitoring (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
According to the system deployment method, the device, the electronic equipment and the storage medium, whether the difference degree of the unit time access quantity at the two detection moments meets the preset difference condition is judged according to the unit time access quantity at the current detection moment of the system and the unit time access quantity at the previous detection moment of the system, if so, the situation that the unit time access quantity is greatly increased or greatly reduced is indicated to possibly occur to the system, at the moment, the target service mechanism number corresponding to the unit time access quantity at the current detection moment is determined according to the preset queuing theory model corresponding to the system, and the service mechanism number currently deployed by the system is adjusted to the target service mechanism number. When the access amount per unit time in the system is greatly increased or reduced, the service mechanism number of the system is adjusted to be suitable for the current access amount per unit time, and the problems of resource waste or too slow access processing caused by unreasonable configuration of the service mechanism number in the system are avoided.
Description
Technical Field
The present application relates to the field of computers, and in particular, to a system deployment method, apparatus, electronic device, and storage medium.
Background
In the actual use process of the system, the phenomenon that the number of threads or QPS (which is abbreviated as Queries Per Second and means the query rate per second, is a measure of how much access a specific query server processes in a specified time) is very unevenly distributed often occurs in one day, for example, the number of accesses is very large in only certain time periods in one day, the number of threads in the system is relatively large, the QPS is relatively high, and the access in other times is relatively low. This causes a problem: when the project is deployed, the high access quantity is met, a plurality of service mechanisms are deployed, or fewer service mechanisms are deployed according to the daily average access quantity. From the point of view of resource utilization, a small number of service institutions should be deployed, but from the point of view of access success rate, user access satisfying peak hours is required, so that it is difficult to balance the two in many service institutions.
Disclosure of Invention
In order to solve the technical problem of how to deploy a service mechanism for a system, the application provides a system deployment method, a device, electronic equipment and a storage medium.
In a first aspect, an embodiment of the present application provides a system deployment method, including:
acquiring the unit time access quantity of the current detection moment and the unit time access quantity of the previous detection moment of the system;
judging whether the difference degree between the unit time access amount at the current detection moment and the unit time access amount at the previous detection moment meets a preset difference condition or not;
if yes, determining the number of target service mechanisms corresponding to the unit time access amount at the current detection moment according to a pre-constructed queuing theory model corresponding to the system;
and adjusting the number of service institutions currently deployed by the system to the target number of service institutions.
In one possible implementation manner, determining whether the difference between the unit time access amount at the current detection time and the unit time access amount at the previous detection time meets a preset difference condition includes:
calculating the absolute value of the difference value between the unit time access quantity of the current detection moment and the unit time access quantity of the previous detection moment;
judging whether the absolute value of the difference value is larger than a preset first threshold value or not;
if the difference is larger than the preset difference condition, the difference degree between the unit time access quantity at the current detection moment and the unit time access quantity at the previous detection moment is determined to be satisfied.
In one possible implementation manner, determining, according to a pre-constructed queuing theory model corresponding to the system, a target service mechanism number corresponding to the unit time access amount at the current detection moment includes:
acquiring the number of preset alternative service institutions and a preset average service rate, wherein the average service rate is the access amount processed by a single service institution in unit time;
inputting the unit time access quantity, the number of alternative service mechanisms and the average service rate of the current detection moment into a pre-constructed queuing theory model corresponding to the system;
and determining the minimum number of the candidate service institutions which enable the parameters output by the queuing theory model to meet the preset requirement as the target service institution number corresponding to the unit time access quantity of the current detection moment.
In one possible implementation manner, before determining the target service organization number corresponding to the unit time access amount of the current detection moment according to a pre-constructed queuing theory model corresponding to the system, the method further includes:
acquiring operation parameters of the system in a preset time period, wherein the operation parameters comprise: the method comprises the following steps of accessing quantity in each unit time, average service rate, access failure number in each unit time and access processing time length corresponding to each unit time, wherein the average service rate is the access quantity processed in unit time of a single service mechanism;
Calculating the probability distribution of the access amount of the system in the preset time period according to the operation parameters;
calculating a service queue length of the system according to the operation parameters and the access volume probability distribution, wherein the service queue length is the highest access volume which can be accommodated in the system in unit time and is subjected to queuing waiting processing;
and determining a queuing theory model corresponding to the system according to preset queuing rules, service captain and average service rate.
In one possible implementation manner, calculating the probability distribution of the access amount of the system in the preset time period according to the operation parameters includes:
according to the acquired access amount of the system in each unit time in a preset time period, calculating the average access amount of the system in each unit time in the preset time period;
taking the average access quantity of the system in each unit time in the preset time period as a parameter of poisson distribution, and taking a probability function of the preset poisson distribution to obtain the probability distribution of the access quantity of the system in the preset time period.
In one possible implementation, calculating a service captain of the system according to the operation parameter and the access probability distribution includes:
Determining a busy period of the access volume probability distribution;
determining an average access amount and an average access failure number corresponding to the busy period according to the operation parameters;
subtracting the average service rate after subtracting the average access failure number corresponding to the busy period from the average access quantity corresponding to the busy period to obtain a difference value corresponding to the busy period;
and taking the difference value corresponding to the busy period as a service captain of the system.
In one possible implementation manner, the operation parameters further include the number of service institutions corresponding to each unit time;
determining a queuing theory model corresponding to the system according to preset queuing rules, service rules, the service queue length and the average service rate, wherein the queuing theory model comprises the following steps:
calculating the actual average waiting time length of the system according to the operation parameters and the access quantity probability distribution, wherein the actual average waiting time length is the average waiting time length of the access in-line waiting in the system;
determining an alternative queuing theory model matched with a preset queuing rule, a preset service queue length and a preset average service rate;
inputting the visit amount, the average service rate and the corresponding service mechanism number of the unit time contained in the operation parameters into the alternative queuing theory model to obtain the simulated average waiting time corresponding to each alternative queuing theory model;
Calculating absolute values of differences between the simulated average waiting time lengths corresponding to the alternative queuing theory models and the actual average waiting time lengths;
and determining an alternative queuing theory model with the minimum absolute value of the difference value between the corresponding simulated average waiting time length and the actual average waiting time length as the queuing theory model.
In one possible implementation, calculating the actual average waiting time of the system according to the operation parameter and the access probability distribution includes:
determining busy and idle periods of the access probability distribution;
determining an average access processing duration corresponding to the busy period according to the operation parameter, and taking the average access processing duration as a first average access processing duration;
determining an average access processing time length corresponding to the idle period according to the operation parameters, and taking the average access processing time length as a second average access processing time length;
calculating a difference value between the first average access processing time length and the second average access processing time length;
and taking the difference value between the first average access processing time length and the second average access processing time length as the actual average waiting time length of the system.
In a second aspect, an embodiment of the present application further provides a system deployment apparatus, including:
The access amount acquisition module is used for acquiring the access amount of the system in unit time at the current detection moment and the access amount of the system in unit time at the previous detection moment;
the judging module is used for judging whether the difference degree between the unit time access quantity at the current detection moment and the unit time access quantity at the previous detection moment meets a preset difference condition or not;
the target service mechanism number determining module is used for determining the target service mechanism number corresponding to the unit time access amount at the current detection moment according to a pre-constructed queuing theory model corresponding to the system if the difference degree between the unit time access amount at the current detection moment and the unit time access amount at the previous detection moment meets a preset difference condition;
and the adjusting module is used for adjusting the number of the currently deployed service institutions of the system to the target number of the service institutions.
In a third aspect, an embodiment of the present application further provides an electronic device, including: the system deployment method according to the first aspect comprises a processor and a memory, wherein the processor is used for executing a data processing program stored in the memory so as to realize the system deployment method according to the first aspect.
In a fourth aspect, an embodiment of the present application further provides a storage medium, where one or more programs are stored, where the one or more programs are executable by one or more processors to implement the system deployment method in the first aspect.
Compared with the prior art, the technical scheme provided by the embodiment of the application has the following advantages:
according to the system deployment method provided by the embodiment of the application, whether the difference degree of the unit time access quantity at the two detection moments meets the preset difference condition is judged according to the unit time access quantity at the current detection moment of the system and the unit time access quantity at the previous detection moment of the system, if so, the condition that the unit time access quantity is greatly increased or greatly reduced is indicated to possibly occur to the system, in order to prevent the number of service mechanisms currently deployed by the system from being too small or too large, the number of target service mechanisms corresponding to the unit time access quantity at the current detection moment is determined according to the preset queuing theory model corresponding to the system, and the number of service mechanisms currently deployed by the system is adjusted to the number of target service mechanisms. When the access amount of the system in unit time is greatly increased or reduced, the service mechanism number of the system is timely adjusted to be suitable for the current access amount of the unit time, so that resource waste caused by low access amount and more service mechanisms is avoided, the resource utilization rate is improved, and the problem that access processing is too slow caused by high access amount and less service mechanisms is also avoided.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
In order to more clearly illustrate the embodiments of the application or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a flowchart of a system deployment method according to an embodiment of the present application;
FIG. 2 is a flowchart of a method for determining the number of target service institutions according to an embodiment of the present application;
FIG. 3 is a block diagram of a system deployment device according to an embodiment of the present application;
fig. 4 is a schematic diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Fig. 1 is a flowchart of a system deployment method according to an embodiment of the present application, as shown in fig. 1, where the method includes the following steps:
s11, acquiring the unit time access quantity of the current detection moment of the system and the unit time access quantity of the previous detection moment.
The current detection time and the previous detection time are two adjacent detection times, the previous detection time is the previous detection time of the current detection time, the specific detection time and the time interval between the two adjacent detection times are set according to specific conditions, for example, the time interval between the two adjacent detection times can be 1 hour or 1 minute, that is, the unit time access amount of the system is detected every 1 hour or 1 minute.
The access amount per unit time is an access amount per unit time, wherein the length of the unit time is also set according to the specific situation, and may be 1 second, for example.
If the system is one where read operations are frequent, then the access may be the QPS of the system, which is an abbreviation for Queries Per Second, meaning the query rate per second.
If the system is a system with frequent write operations or a proxy class system, the access amount may be the thread number of the system.
S12, judging whether the difference degree between the unit time access amount at the current detection moment and the unit application access amount at the previous detection moment meets a preset difference condition or not.
Whether the difference degree of the access amount per unit time of two adjacent detection moments meets the difference condition is judged, so that whether the access amount per unit time is greatly increased or greatly reduced is judged.
S13, if the number of the target service mechanisms corresponding to the unit time access amount at the current detection moment is determined according to a pre-constructed queuing theory model corresponding to the system.
When the access amount per unit time is greatly increased or greatly reduced, the number of service institutions currently deployed by the system may be too small or too large, so in order to provide the utilization rate of resources, the target service institution number adapted to the access amount per unit time at the current detection moment is determined through a preset queuing theory model corresponding to the system.
The queuing theory model can be one of M/M/1, M/D/1, M/Ek/1, M/M/c/≡/M, M/M/c/N/≡, M/Ek/c/N and the like according to the characteristics of the system.
The service mechanism can be a virtual machine or a server and other devices capable of providing services, and the number of the service mechanism is the number of the service mechanism.
S14, adjusting the number of service institutions currently deployed by the system to the target number of service institutions.
If the difference between the unit time access amount at the current detection time and the unit time access amount at the previous detection time meets the difference condition, the current service mechanism number may not well meet the requirement of the access amount at the current detection time, so in order to enable the system to meet the requirement of the unit time access amount at the current detection time, the number of currently deployed service mechanisms is adjusted to the target service mechanism number, the deployed service mechanisms are service mechanisms in a use state, a plurality of service mechanisms may exist in the system, but not necessarily each service mechanism is a deployed service mechanism, and only the service mechanism in the use state is the deployed service mechanism.
According to the system deployment method provided by the embodiment of the application, whether the difference degree of the unit time access quantity at the two detection moments meets the preset difference condition is judged according to the unit time access quantity at the current detection moment of the system and the unit time access quantity at the previous detection moment of the system, if so, the condition that the unit time access quantity is greatly increased or greatly reduced is indicated to possibly occur to the system, in order to prevent the number of service mechanisms currently deployed by the system from being too small or too large, the number of target service mechanisms corresponding to the unit time access quantity at the current detection moment is determined according to the preset queuing theory model corresponding to the system, and the number of service mechanisms currently deployed by the system is adjusted to the number of target service mechanisms. When the unit time access amount in the system is greatly increased or reduced, the service mechanism number of the system is timely adjusted to be suitable for the current unit time access amount, so that the problem that the access amount is low, the resource waste caused by the large number of service mechanisms is avoided, the resource utilization rate is improved, the problem that the access amount is high, and the access processing is too slow caused by the small number of service mechanisms is also avoided.
On the basis of the above embodiment, S12 determines whether the difference between the unit time access amount at the current detection time and the unit time access amount at the previous detection time satisfies the preset difference condition, which may be the following manner:
calculating the absolute value of the difference value between the unit time access amount of the current detection moment and the unit time access amount of the previous detection moment, judging whether the absolute value of the difference value is larger than a preset first threshold value, if so, determining that the difference degree between the unit time access amount of the current detection moment and the unit time access amount of the previous detection moment meets a preset difference condition, and if not, determining that the difference degree between the unit time access amount of the current detection moment and the unit time access amount of the previous detection moment does not meet the preset difference condition.
The first threshold is a value set according to specific conditions.
If the absolute value of the difference between the unit time access amount at the current detection time and the unit time access amount at the previous detection time is larger than the preset first threshold value, the unit time access amount at the current detection time is greatly increased or greatly reduced compared with the unit time access amount at the previous detection time.
In this embodiment, the difference degree of the access amount per unit time of the two adjacent detection moments is determined by calculating the absolute value of the difference value, so as to determine whether the access amount is increased or reduced greatly, and the calculation is simple.
On the basis of the above embodiment, S12 determines whether the difference between the unit time access amount at the current detection time and the unit time access amount at the previous detection time meets the preset difference condition, or may further adopt the following manner:
calculating the ratio of the unit time access amount of the current detection moment to the unit time access amount of the previous detection moment, if the ratio is larger than a preset second threshold value or smaller than a preset third threshold value, determining that the difference degree between the unit time access amount of the current detection moment and the unit time access amount of the previous detection moment meets a preset difference condition, otherwise, determining that the difference degree between the unit time access amount of the current detection moment and the unit time access amount of the previous detection moment does not meet the preset difference condition.
The second threshold and the third threshold are values set according to practical situations, for example, the second threshold is 1.2, and the third threshold is 0.8.
Wherein the ratio of the access amount per unit time at the current detection time to the access amount per unit time at the previous detection time is obtained by dividing the access amount per unit time at the current detection time by the access amount per unit time at the previous detection time
In this embodiment, the ratio may indicate whether the access amount per unit time at the current detection time is increased or decreased compared with the access amount per unit time at the previous detection time, and the magnitude of the increase or decrease, and when the ratio is greater than 1, the increase is indicated, and when the ratio is smaller than 1, the decrease is indicated, and the difference between the access amount per unit time at the current detection time and the access amount per unit time at the previous detection time is indicated by the ratio, which is more visual.
On the basis of the above embodiment, as shown in fig. 2, S13, determining, according to a queuing theory model corresponding to the system, the number of target service institutions corresponding to the unit time access amount at the current detection time, may include:
s131, acquiring the number of preset alternative service institutions and a preset average service rate, wherein the average service rate is the access amount processed by a single service institution in unit time.
The number of the alternative service mechanisms is the number of service mechanisms which can be provided by the system, one system can have a plurality of alternative service mechanism numbers, for example, 6 available service mechanisms in one system can be provided by the system, and the number of the service mechanisms which can be provided by the system is 1, 2, 3, 4, 5 and 6, and the number of the 6 service mechanisms can be used as the number of the alternative service mechanisms of the system.
The average service rate is determined according to the processing capacity of the service mechanism in the system, for example, the service mechanism is a server, and the corresponding average service rate can be determined according to the number of CPU cores of the server.
S132, inputting the unit time access quantity, the number of alternative service mechanisms and the average service rate of the current detection moment into a pre-constructed queuing theory model corresponding to the system.
If there are a plurality of candidate service mechanisms, the candidate service mechanism numbers are respectively input into the queuing theory model.
The queuing theory model takes the number of service institutions, the average service rate and the access amount in unit time as input, and takes parameters such as average queue length, average stay time, average waiting time, average busy period, system state and the like as output.
Wherein, average captain: refers to the mathematical expectation of the amount of access within the system (including the amount of access being processed and the amount of access waiting to be processed in line), denoted Ls.
Average row captain: the mathematical expectation of the amount of access in the system waiting to be processed is referred to as Lq.
Average residence time: refers to the mathematical expectation of accessing the time of stay (including the time of queuing and the time processed) within the system, denoted Ws.
Average waiting time period: refers to a mathematical expectation of access to queuing latency in a queuing system, denoted Wq.
Average busy period: the mathematical expectation of the length of continuous busy time (the time from the access to the idle service to the service being idle again) of the service is denoted as Tb.
The state of the system: refers to the amount of access in the system.
S133, determining the minimum number of the candidate service mechanisms which enable the parameters output by the queuing theory model to meet the preset requirements, wherein the minimum number of the candidate service mechanisms is used as the target service mechanism number corresponding to the unit time access quantity of the current detection moment.
When the number of the input alternative service institutions is different, the values of the parameters output by the queuing theory model are also different.
The method comprises the steps of presetting a threshold value of a parameter output by any one or more queuing theory models, such as a threshold value of an average queue length, a threshold value of an average stay length, a threshold value of an average waiting time length, a threshold value of an average busy period and the like, comparing the parameter value output by the queuing theory models with a preset corresponding parameter threshold value according to the unit time access amount of the current detection moment, the number of alternative service mechanisms and the average service rate, and determining that the corresponding comparison result meets preset conditions (the preset conditions are also preset according to requirements, the preset conditions are different according to different preset conditions of parameters, for example, the preset conditions can be the number of alternative service mechanisms with average stay length smaller than the threshold value of the average stay length), and when the corresponding comparison result of the number of the alternative service mechanisms meets the preset conditions, the number of the alternative service mechanisms can meet the requirement of the unit time access amount of the current detection moment, and the number of the alternative service mechanisms with minimum number of the unit time access amount required by the current detection moment can be regarded as the number of the target service mechanisms.
For example, the number of the alternative service mechanisms is 1, 2, 3, 4, 5 and 6, the 6 numbers are respectively input into queuing theory models corresponding to the system, average stay time lengths corresponding to the number of the alternative service mechanisms are obtained, the average stay time lengths corresponding to the number of the alternative service mechanisms are respectively compared with a preset threshold value of the average stay time length, the number of the alternative service mechanisms, of which the corresponding average stay time length is smaller than the threshold value of the average stay time length, is determined to be 4, 5 and 6, and finally the number of the alternative service mechanisms is determined to be 4.
In this embodiment, the number of alternative service mechanisms that can meet the access amount per unit time at the current detection moment is determined by comparing the output parameter of the queuing theory with a preset parameter threshold, which is simple and quick, and the minimum value of the number of alternative service mechanisms that meet the access amount per unit time at the current detection moment is selected as the target service mechanism number, thereby improving the utilization rate of a single service mechanism.
On the basis of the above embodiment, before determining the number of target service mechanisms corresponding to the unit time access amount at the current detection time according to the pre-constructed queuing theory model corresponding to the system in S13, it is further required to construct a queuing theory model corresponding to the system, where the constructing a queuing theory model corresponding to the system may include the following steps:
Step 1: acquiring operation parameters of the system in a preset time period, wherein the operation parameters comprise: the method comprises the following steps of accessing quantity in each unit time, average service rate, access failure number in each unit time and access processing time length corresponding to each unit time, wherein the average service rate is the access quantity processed in each unit time of a single service mechanism.
The duration and the specific time of the preset time period are set according to specific situations, for example, the duration of the preset time period can be one day.
If the system is one where read operations are frequent, then the access may be the QPS of the system, which is an abbreviation for Queries Per Second, meaning the query rate per second.
If the system is a system with frequent write operations or a proxy class system, the access amount may be the thread number of the system.
Wherein the access processing time length is an average time length consumed for processing each access.
The access failure number of each unit time and the corresponding access processing time length of each unit time can be obtained from the log of the system.
The access failure number per unit time can also be calculated according to the following formula:
number of access failures per unit time = number of access per unit time-number of access successes per unit time, wherein the number of access successes per unit time can be directly obtained.
And 2, calculating the probability distribution of the access amount of the system in the preset time period according to the operation parameters.
Wherein the access probability distribution can be calculated as follows:
according to the acquired access quantity of the system in each unit time in a preset time period, calculating the average access quantity of the system in each unit time in the preset time period, taking the average access quantity of the system in each unit time in the preset time period as a parameter of poisson distribution, and taking the average access quantity into a probability function of the preset poisson distribution to obtain the probability distribution of the access quantity of the system in the preset time period.
Wherein the probability function of the preset poisson distribution is as follows:
where k represents the actual access amount per unit time in the preset time period, and λ represents the average access amount per unit time in the preset time period.
And 3, calculating a service captain of the system according to the operation parameters and the access probability distribution, wherein the service captain is the highest access quantity which can be accommodated in the system and is subjected to queuing waiting processing in unit time.
The service captain may calculate in the following manner:
and determining the busy period of the access quantity probability distribution, determining the average access quantity and the average access failure number corresponding to the busy period according to the operation parameters, subtracting the average access failure number corresponding to the busy period from the average access quantity corresponding to the busy period, subtracting the average service rate to obtain a difference value corresponding to the busy period, and taking the difference value corresponding to the busy period as a service captain of the system.
Since a busy period may include a plurality of unit time, the average access amount corresponding to the busy period is the average access amount corresponding to each unit time in the busy period, and the average access failure number corresponding to the busy period is the average access failure number corresponding to each unit time in the busy period.
If only one busy period exists in the access quantity probability distribution, taking the difference value corresponding to the busy period as a service team leader of the system; if a plurality of busy periods exist in the access quantity probability distribution, taking the average value of the difference values corresponding to the busy periods as the service team leader of the system.
And 4, determining a queuing theory model corresponding to the system according to preset queuing rules, service rules, the service queue length and the average service rate.
The queuing rules mainly include:
queuing mode: wait for the system// instant system, wherein wait for the service organization to be made while being occupied, visit and wait for the processing, visit and process failure while being occupied for the service organization while being instant, wait for the order rule of the queuing mode of the system to be made: first come first served, anytime served, priority first served, etc.
Queuing system capacity: limited// unlimited;
the number of queuing queues is single column// multiple columns;
whether to exit halfway/disable;
Whether inter-column transfer is enabled/disabled.
The service rules mainly include:
number of service institutions: single// multiple;
service organization arrangement form: parallel// serial// mix;
service mode of service organization: one by one// batch by batch;
service time distribution: random// deterministic;
whether the service time distribution is smooth: smooth// non-smooth.
Corresponding queuing rules and service rules are preset according to the characteristics of the system.
According to the existing queuing theory model selection mode, a queuing theory model matched with a preset queuing rule, a service rule, the service captain and the average service rate is selected as a queuing theory model corresponding to the system, and the specific selection process is not repeated here, wherein the service captain is used for determining the capacity of the queuing system in the queuing theory model.
In this embodiment, the queuing theory model corresponding to the system can be constructed in the above manner, so that it is ensured that the constructed queuing theory model can be used for determining the number of target service institutions.
Since a plurality of queuing theory models may be determined according to queuing rules, service rules, and the service queue length and average service rate, in order to make the resulting queuing theory model more suitable for the system, on the basis of the above embodiment, the operation parameters further include the number of service institutions corresponding to each unit time, and in step 4, determining the queuing theory model corresponding to the system according to the preset queuing rules, service rules, and the service queue length and average service rate may be as follows:
And calculating the actual average waiting time length of the system according to the operation parameters and the access volume probability distribution, determining an alternative queuing theory model matched with the queuing rule, the service queue length and the average service rate, inputting the access volume of unit time, the average service rate and the corresponding service mechanism number contained in the operation parameters into the queuing theory model, inputting the alternative queuing theory model to obtain the simulated average waiting time length, calculating the absolute value of the difference value between the simulated average waiting time length and the actual average waiting time length, and determining the alternative queuing theory model with the minimum absolute value of the difference value between the corresponding simulated average waiting time length and the actual average waiting time length as the queuing theory model.
Wherein the average waiting time is the average time for which the access is queued in the system.
Wherein the corresponding number of service institutions is the number of service institutions corresponding to the access amount per unit time input in the alternative queuing theory model, for example, the time corresponding to the input access amount per unit time is 2019, 2, 12, 14, 3 minutes, 16 seconds, and the time corresponding to the input number of service institutions is 2019, 2, 12, 14, 3 minutes, 16 seconds.
In this embodiment, the queuing theory model corresponding to the system determined by adopting the above manner has high accuracy.
On the basis of the above embodiment, the actual average waiting time of the system may be calculated from the operation parameter and the access amount probability distribution in the following manner:
determining busy period and idle period of the access probability distribution, determining average access processing time length corresponding to the busy period according to the operation parameters, determining average access processing time length corresponding to the idle period according to the operation parameters as first average access processing time length, calculating difference value of the first average access processing time length and the second average access processing time length as second average access processing time length, and taking the difference value of the first average access processing time length and the second average access processing time length as actual average waiting time length of the system.
I.e. the actual average waiting time of the system = first average access processing time-second average access processing time.
The average access processing time length corresponding to the busy period is the average value of the access processing time length corresponding to each unit time in the busy period, and the average access processing time length corresponding to the idle period is the average value of the access processing time length corresponding to each unit time in the idle period.
If the access amount probability distribution comprises a plurality of busy periods and/or a plurality of idle periods, taking the average value of the average access processing duration corresponding to all the busy periods as a first average access processing duration, and taking the average value of the average access processing duration corresponding to all the idle periods as a second average access processing duration.
In one possible implementation, step 4 may further determine a queuing theory model corresponding to the system in the following manner:
and determining an alternative queuing theory model matched with the queuing rule, the service queue length and the average service rate, adopting simulation software to simulate the alternative queuing theory model according to the operation parameters to obtain simulated access response success numbers corresponding to each unit time in a preset time period corresponding to each alternative queuing theory model, subtracting the access failure number of each unit time from the access quantity of each unit time contained in the operation parameters to obtain actual access response success numbers corresponding to each unit time in the preset time period, respectively calculating error values corresponding to each alternative queuing theory model according to the simulated access response success numbers and the actual access response success numbers corresponding to each alternative queuing theory model, and selecting the alternative queuing theory model with the minimum error value as the queuing theory model corresponding to the system.
For each alternative queuing theory model, its corresponding error value may be calculated in the following manner:
wherein S is j Representing the error value corresponding to the alternative queuing theory model j, C ji Representing the success number of the simulation access response of the ith unit time of the alternative queuing theory model j in the preset time period, R i Representing the actual access response success number of the ith unit time in the preset time period, and m represents the total number of unit time contained in the preset time period.
The simulation software can be MATLAB simulation software.
In this embodiment, the queuing theory model is screened in a simulation and calculation manner, so that the queuing theory model obtained finally is more accurate.
According to the system deployment method provided by the embodiment of the application, when the system access quantity is increased or decreased sharply, the queuing theory model corresponding to the system is started, whether the number of the current service mechanisms is sufficient or excessive is analyzed in real time, then the number of the target service mechanisms obtained according to the queuing theory model is called, the related interfaces are used for realizing the dynamic expansion or reduction of the number of the service mechanisms in the system, the utilization rate of resources is improved through the dynamic expansion or reduction of the number of the service mechanisms, the resources of the service mechanisms (virtual machines or servers) can be acquired and released as required, the service satisfaction degree of users is further improved, the number of the service mechanisms deployed in the system can be increased dynamically when the access quantity is increased, efficient service is provided for the users, and the probability of access failure is reduced.
On the basis of the embodiment, when the service success rate of the system is reduced, the number of service mechanisms deployed in the system can be adjusted according to the queuing theory model corresponding to the system, so that the service success rate is improved, wherein the service success rate can be determined by analyzing the logs returned by the interface.
Fig. 3 is a block diagram of a system deployment apparatus according to an embodiment of the present application, where, as shown in fig. 3, the apparatus may include:
an access amount obtaining module 301, configured to obtain an access amount per unit time at a current detection time and an access amount per unit time at a previous detection time of the system;
a judging module 302, configured to judge whether a difference between the unit time access amount at the current detection time and the unit time access amount at the previous detection time meets a preset difference condition;
a target service mechanism number determining module 303, configured to determine, if a difference between a unit time access amount at a current detection time and a unit time access amount at a previous detection time meets a preset difference condition, a target service mechanism number corresponding to the unit time access amount at the current detection time according to a queuing theory model that is pre-constructed and corresponds to the system;
and the adjusting module 304 is configured to adjust the number of service institutions currently deployed by the system to the target number of service institutions.
On the basis of the above embodiment, the judging module 302 is specifically configured to:
calculating the absolute value of the difference value between the unit time access quantity of the current detection moment and the unit time access quantity of the previous detection moment;
Judging whether the absolute value of the difference value is larger than a preset first threshold value or not;
if the difference is larger than the preset difference condition, the difference degree between the unit time access quantity at the current detection moment and the unit time access quantity at the previous detection moment is determined to be satisfied.
On the basis of the above embodiment, the target service institution number determining module 303 is specifically configured to:
acquiring the number of preset alternative service institutions and a preset average service rate, wherein the average service rate is the access amount processed by a single service institution in unit time;
inputting the unit time access quantity, the number of alternative service mechanisms and the average service rate of the current detection moment into a pre-constructed queuing theory model corresponding to the system;
and determining the minimum number of the candidate service institutions which enable the parameters output by the queuing theory model to meet the preset requirement as the target service institution number corresponding to the unit time access quantity of the current detection moment.
On the basis of the above embodiment, the apparatus further includes: and the model construction module is used for constructing a queuing theory model corresponding to the system before determining the target service mechanism number corresponding to the unit time access amount at the current detection moment according to the pre-constructed queuing theory model corresponding to the system.
The model building module may include:
the acquisition sub-module is used for acquiring the operation parameters of the system in a preset time period, wherein the operation parameters comprise: the method comprises the following steps of accessing quantity in each unit time, average service rate, access failure number in each unit time and access processing time length corresponding to each unit time, wherein the average service rate is the access quantity processed in unit time of a single service mechanism;
the probability distribution determining submodule is used for calculating the probability distribution of the access quantity of the system in the preset time period according to the operation parameters;
a service queue length determining sub-module, configured to calculate a service queue length of the system according to the operation parameter and the access volume probability distribution, where the service queue length is a highest access volume of queuing waiting processing that can be accommodated in the system in unit time;
and the model determination submodule is used for determining a queuing theory model corresponding to the system according to preset queuing rules, service rules, the service queue length and the average service rate.
On the basis of the above embodiment, the probability distribution determining submodule is specifically configured to:
according to the acquired access quantity of the system in each unit time, calculating the average access quantity of the system in each unit time;
Taking the average access quantity of the system in each unit time as a parameter of poisson distribution, and taking a probability function of preset poisson distribution to obtain the access quantity probability distribution of the system in the preset time period.
On the basis of the above embodiment, the service captain determination submodule is specifically configured to:
determining a busy period of the access volume probability distribution;
determining an average access amount and an average access failure number corresponding to the busy period according to the operation parameters;
subtracting the average access failure number corresponding to the busy period from the average access quantity corresponding to the busy period, subtracting the average service rate to obtain a difference value corresponding to the busy period, and taking the difference value corresponding to the busy period as a service captain of the system.
On the basis of the embodiment, the operation parameters further comprise the number of service institutions corresponding to each unit time;
the model determination submodule is specifically used for:
calculating the actual average waiting time length of the system according to the operation parameters and the access quantity probability distribution, wherein the actual average waiting time length is the average waiting time length of the access in-line waiting in the system;
determining an alternative queuing theory model matched with a preset queuing rule, a preset service queue length and a preset average service rate;
Inputting the visit amount, the average service rate and the corresponding service mechanism number of the unit time contained in the operation parameters into the alternative queuing theory model to obtain the simulated average waiting time corresponding to each alternative queuing theory model;
calculating absolute values of differences between the simulated average waiting time lengths corresponding to the alternative queuing theory models and the actual average waiting time lengths;
and determining an alternative queuing theory model with the minimum absolute value of the difference value between the corresponding simulated average waiting time length and the actual average waiting time length as the queuing theory model.
On the basis of the above embodiment, the model determining submodule calculates an actual average waiting time of the system according to the operation parameter and the visit probability distribution specifically includes:
determining busy and idle periods of the access probability distribution;
determining an average access processing duration corresponding to the busy period according to the operation parameter, and taking the average access processing duration as a first average access processing duration;
determining an average access processing time length corresponding to the idle period according to the operation parameters, and taking the average access processing time length as a second average access processing time length;
and calculating the difference value between the first average access processing time length and the second average access processing time length as the actual average waiting time length of the system.
According to the system deployment device provided by the embodiment of the application, when the access quantity of the system is increased or decreased sharply, the queuing theory model corresponding to the system is started, whether the number of the current service mechanisms is sufficient or excessive is analyzed in real time, then the appropriate target service mechanism number is obtained according to the queuing theory model corresponding to the system, the related interfaces are called to realize the dynamic expansion or reduction of the service mechanism number in the system, the utilization rate of resources is improved through the dynamic expansion or reduction of the service mechanism number, the service mechanism resources can be acquired and released as required, the service satisfaction of a user is further improved, the number of the service mechanisms deployed in the system can be increased dynamically when the QPS is high, efficient service is provided for the user, and the probability of access failure is reduced.
In another embodiment of the present application, there is also provided an electronic device, as shown in fig. 4, including a processor 401, a communication interface 402, a memory 403, and a communication bus 404, where the processor 401, the communication interface 402, and the memory 403 complete communication with each other through the communication bus 404;
a memory 403 for storing a computer program;
the processor 401, when executing the program stored in the memory 403, implements the following steps:
Acquiring the unit time access quantity of the current detection moment and the unit time access quantity of the previous detection moment of the system;
judging whether the difference degree between the unit time access amount at the current detection moment and the unit time access amount at the previous detection moment meets a preset difference condition or not;
if yes, determining the number of target service mechanisms corresponding to the unit time access amount at the current detection moment according to a pre-constructed queuing theory model corresponding to the system;
and adjusting the number of service institutions currently deployed by the system to the target number of service institutions.
The communication bus 404 mentioned above for the electronic device may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, etc. The communication bus 404 may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, only one thick line is shown in fig. 4, but not only one bus or one type of bus.
The communication interface 402 is used for communication between the electronic device and other devices described above.
The memory 403 may include a random access memory (Random Access Memory, abbreviated as RAM) or may include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.
The processor 401 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (Digital Signal Processing, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field-programmable gate arrays (Field-Programmable Gate Array, FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
In another embodiment of the present application, there is further provided a computer readable storage medium, where a system deployment method program is stored, where the system deployment method program when executed by a processor implements the steps of any of the system deployment methods described above.
When the embodiment of the application is specifically implemented, the above embodiments can be referred to, and the application has corresponding technical effects.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is only a specific embodiment of the application to enable those skilled in the art to understand or practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. A system deployment method, comprising:
acquiring the unit time access quantity of the current detection moment and the unit time access quantity of the previous detection moment of the system;
judging whether the difference degree between the unit time access amount at the current detection moment and the unit time access amount at the previous detection moment meets a preset difference condition or not;
acquiring operation parameters of the system in a preset time period, wherein the operation parameters comprise: the method comprises the following steps of accessing quantity in each unit time, average service rate, access failure number in each unit time and access processing duration corresponding to each unit time, wherein the average service rate is the access quantity processed in unit time of a single service mechanism; calculating the probability distribution of the access quantity of the system in a preset time period according to the operation parameters; calculating a service captain of the system according to the operation parameters and the access probability distribution, wherein the service captain is the highest access quantity which can be accommodated in the system and is subjected to queuing waiting processing in unit time; determining a queuing theory model corresponding to the system according to preset queuing rules, service captain and average service rate;
If yes, determining the number of target service mechanisms corresponding to the unit time access amount at the current detection moment according to a pre-constructed queuing theory model corresponding to the system, wherein the method comprises the following steps: acquiring the number of preset alternative service institutions and a preset average service rate, wherein the average service rate is the access amount processed by a single service institution in unit time; inputting the unit time access quantity, the number of alternative service mechanisms and the average service rate of the current detection moment into a pre-constructed queuing theory model corresponding to the system; determining the minimum number of the candidate service mechanisms which enable the parameters output by the queuing theory model to meet the preset requirements as the target service mechanism number corresponding to the unit time access quantity of the current detection moment;
and adjusting the number of service institutions currently deployed by the system to the target number of service institutions.
2. The method according to claim 1, wherein determining whether a degree of difference between the access amount per unit time at the current detection time and the access amount per unit time at the previous detection time satisfies a preset difference condition includes:
calculating the absolute value of the difference value between the unit time access quantity of the current detection moment and the unit time access quantity of the previous detection moment;
Judging whether the absolute value of the difference value is larger than a preset first threshold value or not;
if the difference is larger than the preset difference condition, the difference degree between the unit time access quantity at the current detection moment and the unit time access quantity at the previous detection moment is determined to be satisfied.
3. The method of claim 1, wherein calculating an access volume probability distribution of the system over the preset time period based on the operating parameters comprises:
according to the acquired access amount of the system in each unit time in a preset time period, calculating the average access amount of the system in each unit time in the preset time period;
taking the average access quantity of the system in each unit time in the preset time period as a parameter of poisson distribution, and taking a probability function of the preset poisson distribution to obtain the probability distribution of the access quantity of the system in the preset time period.
4. The method of claim 1, wherein calculating a service captain of the system based on the operating parameters and the access probability distribution comprises:
determining a busy period of the access volume probability distribution;
determining an average access amount and an average access failure number corresponding to the busy period according to the operation parameters;
Subtracting the average service rate after subtracting the average access failure number corresponding to the busy period from the average access quantity corresponding to the busy period to obtain a difference value corresponding to the busy period;
and taking the difference value corresponding to the busy period as a service captain of the system.
5. The method of claim 1, wherein the operating parameters further comprise a number of service organizations corresponding to each unit time;
determining a queuing theory model corresponding to the system according to preset queuing rules, service rules, the service queue length and the average service rate, wherein the queuing theory model comprises the following steps:
calculating the actual average waiting time length of the system according to the operation parameters and the visit amount probability distribution, wherein the average waiting time length is the average waiting time length of visit queuing in the system;
determining an alternative queuing theory model matched with a preset queuing rule, a preset service queue length and a preset average service rate;
inputting the visit amount, the average service rate and the corresponding service mechanism number of the unit time contained in the operation parameters into the alternative queuing theory model to obtain the simulated average waiting time corresponding to each alternative queuing theory model;
calculating absolute values of differences between the simulated average waiting time lengths corresponding to the alternative queuing theory models and the actual average waiting time lengths;
And determining an alternative queuing theory model with the minimum absolute value of the difference value between the corresponding simulated average waiting time length and the actual average waiting time length as the queuing theory model.
6. The method of claim 5, wherein calculating an actual average wait time for the system based on the operating parameters and the access probability distribution comprises:
determining busy and idle periods of the access probability distribution;
determining an average access processing duration corresponding to the busy period according to the operation parameter, and taking the average access processing duration as a first average access processing duration;
determining an average access processing time length corresponding to the idle period according to the operation parameters, and taking the average access processing time length as a second average access processing time length;
calculating a difference value between the first average access processing time length and the second average access processing time length;
and taking the difference value between the first average access processing time length and the second average access processing time length as the actual average waiting time length of the system.
7. A system deployment apparatus, comprising:
the access amount acquisition module is used for acquiring the access amount of the system in unit time at the current detection moment and the access amount of the system in unit time at the previous detection moment;
The judging module is used for judging whether the difference degree between the unit time access quantity at the current detection moment and the unit time access quantity at the previous detection moment meets a preset difference condition or not;
a model building module, the model building module comprising: the acquisition sub-module is used for acquiring the operation parameters of the system in a preset time period, wherein the operation parameters comprise: the method comprises the following steps of accessing quantity in each unit time, average service rate, access failure number in each unit time and access processing duration corresponding to each unit time, wherein the average service rate is the access quantity processed in unit time of a single service mechanism; the probability distribution determining submodule is used for calculating the probability distribution of the access quantity of the system in a preset time period according to the operation parameters; the service queue length determining submodule is used for calculating the service queue length of the system according to the operation parameters and the access quantity probability distribution, wherein the service queue length is the highest access quantity which can be accommodated in the system in unit time and is subjected to queuing waiting processing; the model determining submodule is used for determining a queuing theory model corresponding to the system according to preset queuing rules, service captain and average service rate;
the target service mechanism number determining module is configured to determine, according to a queuing theory model corresponding to the system, a target service mechanism number corresponding to a unit time access amount at a current detection time, if a difference between the unit time access amount at the current detection time and a unit time access amount at a previous detection time meets a preset difference condition, including: acquiring the number of preset alternative service institutions and a preset average service rate, wherein the average service rate is the access amount processed by a single service institution in unit time; inputting the unit time access quantity, the number of alternative service mechanisms and the average service rate of the current detection moment into a pre-constructed queuing theory model corresponding to the system; determining the minimum number of the candidate service mechanisms which enable the parameters output by the queuing theory model to meet the preset requirements as the target service mechanism number corresponding to the unit time access quantity of the current detection moment;
And the adjusting module is used for adjusting the number of the currently deployed service institutions of the system to the target number of the service institutions.
8. An electronic device, comprising: a processor and a memory, the processor being configured to execute a data processing program stored in the memory, to implement the system deployment method of any of claims 1-6.
9. A storage medium storing one or more programs executable by one or more processors to implement the system deployment method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010495898.XA CN111767137B (en) | 2020-06-03 | 2020-06-03 | System deployment method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010495898.XA CN111767137B (en) | 2020-06-03 | 2020-06-03 | System deployment method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111767137A CN111767137A (en) | 2020-10-13 |
| CN111767137B true CN111767137B (en) | 2023-09-01 |
Family
ID=72720031
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010495898.XA Active CN111767137B (en) | 2020-06-03 | 2020-06-03 | System deployment method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111767137B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5923873A (en) * | 1995-12-28 | 1999-07-13 | Lucent Technologies Inc. | Method for determining server staffing in management of finite server queueing systems |
| JP2005148799A (en) * | 2003-11-11 | 2005-06-09 | Fujitsu Ltd | Method for processing information, method for deriving service time and method for adjusting number of processing unit |
| CN103632037A (en) * | 2013-11-07 | 2014-03-12 | 成都国科海博信息技术股份有限公司 | Method for computing server average waiting time based on working vacation |
| CN104092756A (en) * | 2014-07-09 | 2014-10-08 | 东南大学 | Cloud storage system resource dynamic allocation method based on DHT mechanism |
| CN107844376A (en) * | 2017-11-21 | 2018-03-27 | 北京星河星云信息技术有限公司 | Resource allocation method, computing system, medium and the server of computing system |
| CN110138732A (en) * | 2019-04-03 | 2019-08-16 | 平安科技（深圳）有限公司 | Response method, device, equipment and the storage medium of access request |
| CN110196767A (en) * | 2018-03-05 | 2019-09-03 | 腾讯科技（深圳）有限公司 | Service resource control method, device, equipment and storage medium |
| WO2019169724A1 (en) * | 2018-03-09 | 2019-09-12 | 平安科技（深圳）有限公司 | Server concurrency control method and device, computer device, and storage medium |
| CN111010303A (en) * | 2019-12-20 | 2020-04-14 | 北京爱奇艺科技有限公司 | Server control method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110231476A1 (en) * | 2010-03-22 | 2011-09-22 | Hung Tso-Sung | Information service platform equipped with dynamic distribution operators |
-
2020
- 2020-06-03 CN CN202010495898.XA patent/CN111767137B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5923873A (en) * | 1995-12-28 | 1999-07-13 | Lucent Technologies Inc. | Method for determining server staffing in management of finite server queueing systems |
| JP2005148799A (en) * | 2003-11-11 | 2005-06-09 | Fujitsu Ltd | Method for processing information, method for deriving service time and method for adjusting number of processing unit |
| CN103632037A (en) * | 2013-11-07 | 2014-03-12 | 成都国科海博信息技术股份有限公司 | Method for computing server average waiting time based on working vacation |
| CN104092756A (en) * | 2014-07-09 | 2014-10-08 | 东南大学 | Cloud storage system resource dynamic allocation method based on DHT mechanism |
| CN107844376A (en) * | 2017-11-21 | 2018-03-27 | 北京星河星云信息技术有限公司 | Resource allocation method, computing system, medium and the server of computing system |
| CN110196767A (en) * | 2018-03-05 | 2019-09-03 | 腾讯科技（深圳）有限公司 | Service resource control method, device, equipment and storage medium |
| WO2019169724A1 (en) * | 2018-03-09 | 2019-09-12 | 平安科技（深圳）有限公司 | Server concurrency control method and device, computer device, and storage medium |
| CN110138732A (en) * | 2019-04-03 | 2019-08-16 | 平安科技（深圳）有限公司 | Response method, device, equipment and the storage medium of access request |
| CN111010303A (en) * | 2019-12-20 | 2020-04-14 | 北京爱奇艺科技有限公司 | Server control method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111767137A (en) | 2020-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2019332682B2 (en) | Cost optimized dynamic resource allocation in a cloud infrastructure | |
| JP7020616B2 (en) | Methods and systems for resource scheduling | |
| CN108491122B (en) | Click event response method, computer readable storage medium and terminal equipment | |
| US9483393B1 (en) | Discovering optimized experience configurations for a software application | |
| CN112764936B (en) | Edge calculation server information processing method and device based on deep reinforcement learning | |
| JP2003504720A (en) | Improved real and perceived response times, user interfaces, and security using usage patterns | |
| CN112506643A (en) | Load balancing method and device of distributed system and electronic equipment | |
| CN108762885B (en) | Virtual machine creating method and device, management equipment and terminal equipment | |
| CN114490078A (en) | Dynamic capacity reduction and expansion method, device and equipment for micro-service | |
| CN111767137B (en) | System deployment method and device, electronic equipment and storage medium | |
| CN113452546A (en) | Dynamic quality of service management for deep learning training communications | |
| CN110347477B (en) | Service self-adaptive deployment method and device in cloud environment | |
| CN116360990B (en) | Distributed computing task rationality pre-judging method, system, equipment and storage medium | |
| CN114866563A (en) | Capacity expansion method, device, system and storage medium | |
| CN110069340B (en) | Thread number evaluation method and device | |
| Rao et al. | CoSL: A coordinated statistical learning approach to measuring the capacity of multi-tier websites | |
| CN110083517A (en) | A kind of optimization method and device of user's portrait confidence level | |
| CN111711650B (en) | Network request scheduling method, device, equipment and storage medium | |
| US11972302B1 (en) | User-aware request processing in application services | |
| CN114546279B (en) | IO request prediction method and device, storage node and readable storage medium | |
| US8626476B2 (en) | Determining the impact of elements in a technology system | |
| CN115442432B (en) | Control method, device, equipment and storage medium | |
| KR102255252B1 (en) | Method and server for deciding summary value from big raw data | |
| CN112579281B (en) | Resource allocation method, device, electronic equipment and storage medium | |
| CN114116213A (en) | Method, device, equipment and storage medium for estimating excess load in distributed system |
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 |