CN112348666A - Method and device for determining system capacity - Google Patents
Method and device for determining system capacity Download PDFInfo
- Publication number
- CN112348666A CN112348666A CN202011168930.XA CN202011168930A CN112348666A CN 112348666 A CN112348666 A CN 112348666A CN 202011168930 A CN202011168930 A CN 202011168930A CN 112348666 A CN112348666 A CN 112348666A
- Authority
- CN
- China
- Prior art keywords
- capacity
- service subsystem
- determining
- factor
- load
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Debugging And Monitoring (AREA)
Abstract
The embodiment of the invention provides a method and a device for determining system capacity, wherein the method comprises the steps of collecting current operation information of each service subsystem in an application system, determining real-time information of each capacity influence factor from the current operation information of the service subsystems aiming at least one service subsystem, and determining the system capacity of the service subsystems according to the real-time information of each capacity influence factor and a capacity calculation model of the service subsystems. The system capacity is automatically calculated by combining the real-time information of each capacity influence factor, the weight of each capacity influence factor and the capacity calculation model, so that excessive manual intervention can be avoided, time and labor consumed by manually determining the system capacity can be reduced, the real-time performance and accuracy of system capacity determination can be ensured, and the problems of low evaluation efficiency and low accuracy caused by manual evaluation of the system capacity in the prior art can be solved.
Description
Technical Field
The embodiment of the invention relates to the field of financial technology (Fintech), in particular to a method and a device for determining system capacity.
Background
With the development of computer technology, more and more technologies are applied in the financial field, and the traditional financial industry is gradually changing to financial technology, but due to the requirements of the financial industry on safety and real-time performance, higher requirements are also put forward on the technologies. In the financial field, due to the concurrency of financial transaction execution, a banking system may execute a large amount of financial transactions for a period of time, which requires the banking system to provide a relatively sufficient capacity to process the financial transactions. Therefore, timely evaluating the capacity of the bank system is extremely important for ensuring the normal operation of the bank system.
The capacity evaluation method of the existing bank system is mainly based on manual evaluation of operation and maintenance personnel or cost operation planning personnel. Specifically, when the capacity of the bank system needs to be evaluated, the operation and maintenance staff or the cost operation planner obtains the operation state (such as the request processing time of each transaction of the system, the number of machines in the system, the machine load, and the like) of the bank system in the operation process, and calculates the operation data in the operation state to obtain the evaluation result of the capacity of the bank system. However, because the processing method mainly depends on the operation and maintenance personnel or the cost operation planning personnel to perform manual calculation, a long time and effort are needed, and because most of the bank systems to be evaluated are online running systems, the application after evaluation has hysteresis. In addition, because the estimated capacity is smaller than the actual capacity due to inaccurate estimation results in the previous period or careless mistakes in manual calculation, the system may cause transaction abnormality due to insufficient system load in the online process, which causes abnormal customer funds, thereby bringing serious loss to customers. It may also happen that the estimated capacity is much larger than the actual demand, which may result in wasted server resources and thus increased bank operating costs.
In summary, a method for determining system capacity is needed to solve the problems of low evaluation efficiency and low accuracy caused by manual evaluation of system capacity in the prior art.
Disclosure of Invention
The embodiment of the invention provides a method and a device for determining system capacity, which are used for solving the problems of low evaluation efficiency and low accuracy caused by manual evaluation of the system capacity in the prior art.
In a first aspect, an embodiment of the present invention provides a method for determining system capacity, including:
collecting current operation information of each service subsystem in an application system;
for at least one service subsystem, determining real-time information of each capacity influence factor from current operation information of the service subsystem;
determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem; the capacity calculation model is provided with weights of various capacity influence factors determined by historical operation information of the service subsystem.
In the above technical solution, the current operation information of each service subsystem in the application system is obtained in real time, and the real-time information of each capacity influencing factor is determined from the current operation information of the service subsystem for at least one service subsystem. And determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem. The system capacity is automatically calculated by combining the real-time information of each capacity influence factor, the weight of each capacity influence factor and the capacity calculation model, so that excessive manual intervention can be avoided, time and labor consumed by manually determining the system capacity can be reduced, the real-time performance and accuracy of system capacity determination can be ensured, and the problems of low evaluation efficiency and low accuracy caused by manual evaluation of the system capacity in the prior art can be solved.
Optionally, the weight of each capacity influencing factor is determined according to the following mode:
analyzing and processing the historical operation information of the application system based on an analytic hierarchy process, and determining each first influence factor as a scheme layer, each second influence factor as a criterion layer and the system capacity as a target layer; each capacity influencing factor comprises each first influencing factor and each second influencing factor;
and determining the first weight of each first influence factor and the second weight of each second influence factor based on an analytic hierarchy process.
In the above technical scheme, the analytic processing is performed on the historical operation information of the application system through the analytic hierarchy process, so that the first weight of each first influence factor and the second weight of each second influence factor can be quickly and accurately determined, and the subsequent capacity calculation model determines the system capacity of the service subsystem according to the first weight of each first influence factor and the second weight of each second influence factor.
Optionally, each first influencing factor includes a stand-alone processing capability and a machine number; the second influence factors include disaster tolerance capability, machine load, storage load and traffic load.
Optionally, determining the system capacity of the service subsystem according to the real-time information of each capacity influencing factor and the capacity calculation model of the service subsystem, includes:
determining a maximum system capacity of the service subsystem according to the following formula (1);
the formula (1) is:
a... (1) maximum system capacity ÷ (load condition × load weight) · maximum traffic processing capacity ÷ (load condition × load weight)
Determining the capacity of the disaster recovery system of the service subsystem according to the following formula (2);
the formula (2) is:
the disaster tolerance system capacity is the weight of the disaster tolerance capability x the maximum system capacity x the disaster tolerance capability
The load condition comprises machine load, storage load and traffic load, and the disaster tolerance capability is the ratio of the number of available machines to the total number of machines.
According to the technical scheme, the maximum system capacity and the disaster recovery system capacity of the service subsystem can be efficiently and accurately determined through the capacity calculation model, the system capacity of the service subsystem can be accurately determined in real time based on the maximum system capacity and the disaster recovery system capacity of the service subsystem, excessive manual intervention is not needed, time and labor consumed by manually determining the system capacity are reduced, and real-time performance and accuracy of system capacity determination are guaranteed.
Optionally, the service subsystem is an online service subsystem;
determining the maximum service processing capacity of the online service subsystem according to the following formula (3);
the formula (3) is:
the maximum TPS request amount (Transactions Per Second) represents the maximum number of requested Transactions Per unit time.
Optionally, the service subsystem is a batch service subsystem;
determining the maximum service processing capacity of the batch service subsystem according to the following formula (4);
the formula (4) is:
optionally, the method further comprises:
and determining the dynamic system capacity of the service subsystem according to the current operation information of the service subsystem.
In the technical scheme, the current operation information of the service subsystem is obtained in real time, and the dynamic system capacity of the service subsystem can be accurately determined in real time based on the current operation information of the service subsystem, so that support is provided for accurately determining the system capacity of the service subsystem in real time according to a capacity calculation model in the follow-up process.
In a second aspect, an embodiment of the present invention further provides an apparatus for determining system capacity, where the apparatus includes:
the acquisition unit is used for acquiring the current operation information of each service subsystem in the application system;
the processing unit is used for determining real-time information of each capacity influence factor from the current operation information of at least one service subsystem; determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem; the capacity calculation model is provided with weights of various capacity influence factors determined by historical operation information of the service subsystem.
Optionally, the processing unit is specifically configured to:
determining the weight of each capacity influencing factor according to the following mode:
analyzing and processing the historical operation information of the application system based on an analytic hierarchy process, and determining each first influence factor as a scheme layer, each second influence factor as a criterion layer and the system capacity as a target layer; each capacity influencing factor comprises each first influencing factor and each second influencing factor;
and determining the first weight of each first influence factor and the second weight of each second influence factor based on an analytic hierarchy process.
Optionally, the processing unit is specifically configured to:
each first influence factor comprises single machine processing capacity and machine number; the second influence factors include disaster tolerance capability, machine load, storage load and traffic load.
Optionally, the processing unit is specifically configured to:
determining a maximum system capacity of the service subsystem according to the following formula (1);
the formula (1) is:
a... (1) maximum system capacity ÷ (load condition × load weight) · maximum traffic processing capacity ÷ (load condition × load weight)
Determining the capacity of the disaster recovery system of the service subsystem according to the following formula (2);
the formula (2) is:
the disaster tolerance system capacity is the weight of the disaster tolerance capability x the maximum system capacity x the disaster tolerance capability
The load condition comprises machine load, storage load and traffic load, and the disaster tolerance capability is the ratio of the number of available machines to the total number of machines.
Optionally, the service subsystem is an online service subsystem;
the processing unit is specifically configured to:
determining the maximum service processing capacity of the online service subsystem according to the following formula (3);
the formula (3) is:
the maximum TPS request amount (Transactions Per Second) represents the maximum number of requested Transactions Per unit time.
Optionally, the service subsystem is a batch service subsystem;
the processing unit is specifically configured to:
determining the maximum service processing capacity of the batch service subsystem according to the following formula (4);
the formula (4) is:
the storage performance coefficient and the network delay coefficient are constants.
Optionally, the processing unit is further configured to:
and determining the dynamic system capacity of the service subsystem according to the current operation information of the service subsystem.
In a third aspect, an embodiment of the present invention provides a computing device, including:
a memory for storing a computer program;
and the processor is used for calling the computer program stored in the memory and executing the method for determining the system capacity according to the obtained program.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing a computer-executable program for causing a computer to execute a method for determining system capacity.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic diagram of a system architecture according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a method for determining system capacity according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a hierarchy model provided in an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating a relationship between system capacity and request number in an online subsystem capacity calculation model according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a relationship between system capacity and a number of faulty machines in an on-line subsystem capacity calculation model according to an embodiment of the present invention;
FIG. 6 is a diagram illustrating a capacity calculation result of an on-line subsystem according to an embodiment of the present invention;
FIG. 7 is a diagram illustrating a relationship between a batch task subsystem capacity and a batch task running time according to an embodiment of the present invention;
FIG. 8 is a diagram illustrating a relationship between the number of involved accounts and a processing duration of batch tasks according to an embodiment of the present invention;
FIG. 9 is a diagram illustrating a result of calculating capacity of a batch task subsystem according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of an apparatus for determining system capacity according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the following, some terms related to the embodiments of the present invention are first explained to facilitate understanding by those skilled in the art.
(1) An IT system: the application system mentioned in the embodiment of the present invention may refer to an Internet application system of a bank deposit business, and may also refer to other transaction application systems in the financial field, which is not limited in the embodiment of the present invention.
(2) Capacity: application systems (such as bank IT systems, etc.) can support the amount of server resources required for terminal service requests.
(3) Capacity platform: the method is used for carrying out capacity calculation on the capacity of an application system (such as a bank IT system) and feeding back a reasonable capacity evaluation result for reference of operation and maintenance personnel.
(4) TPS: transactions Per Second, i.e., Transactions/Second. Wherein, a transaction refers to a process that a client sends a request to a server and then the server reacts. The client starts timing when sending the request, and finishes timing after receiving the response of the server, so as to calculate the used time and the number of completed transactions. One transaction is a transaction.
As described above, some terms related to the embodiments of the present invention are described, and the technical features related to the embodiments of the present invention are described below.
In the prior art, the capacity of the bank system is evaluated manually, so that a long time and effort are needed, and the evaluated bank system is mostly an online running system, so that the application after evaluation has hysteresis. In addition, due to inaccurate manual evaluation results, the evaluated capacity is smaller than the actual capacity or the evaluated capacity is larger than the actual capacity, so that transaction abnormity occurs due to insufficient load of the system or server resources are wasted. Therefore, the embodiment of the invention replaces manpower by constructing the server capacity platform to quickly and accurately evaluate the system capacity, so as to solve the problems of low evaluation efficiency and low accuracy caused by manually evaluating the system capacity in the prior art.
To facilitate understanding of the embodiment of the present invention, a system architecture suitable for capacity calculation according to the embodiment of the present invention is first described by taking the system architecture shown in fig. 1 as an example. As shown in fig. 1, the system architecture may include a capacity platform and an external system, as shown in fig. 1. Since the application system (such as a bank deposit IT system) generally includes two parts, namely an online system and a batch task system, the objects collected by the capacity platform are the online system and the batch task system.
The capacity platform may include an acquisition module 101 and a capacity calculation module 102. The capacity calculation module may include an online system calculation module 1021 and a batch task system calculation module 1022.
The acquisition module 101 is configured to acquire operation information of an application system (e.g., a bank deposit IT system, etc.) (e.g., periodically acquire the operation information at an acquisition frequency of once every 1 second). And meanwhile, analyzing, classifying and integrating the collected running information of the application system. For example, the request amount of the online system collected every second is normalized (weighted average), and is sorted into 1 minute data, and the data is input to the capacity calculation module 102 for capacity calculation. The acquisition module may be disposed in the server to acquire the operation information of the application system, or may manually collect the operation information of the application system, classify and sort the operation information, and input the operation information into the capacity calculation module 102 to perform capacity calculation.
The capacity calculation module 102 is configured to input the data sorted by the collection module into a preset capacity calculation model to perform capacity calculation (for example, the operation data of the online system is input into the online system calculation module 1021 to perform capacity calculation, and the operation data of the batch task system is input into the batch task system calculation module 1022 to perform capacity calculation), so as to obtain a capacity calculation result, that is, the real-time capacity of the application system (for example, a bank IT system, and the like), and feed back the real-time capacity to a user (for example, an operation and maintenance person or a cost operation planning person, and the like), and the user may determine whether to perform capacity expansion or capacity reduction on the application system according to the capacity calculation result.
The external systems may include application systems (such as bank deposit IT systems, etc.) and users (such as operation and maintenance personnel or cost operation planner, etc.). The application system is used for providing the running information of the application system for the capacity platform (such as the actual request quantity of the online system, the execution time consumption of each online transaction, the batch task quantity of the batch task system, the execution time consumption of the batch task, the CPU load of a host computer in the system, the memory load, the I/O load of a hard disk, the use condition of storage service, the machine quantity and the like); the user is used for receiving a capacity calculation result obtained by the capacity platform performing capacity calculation on the operation information of the application system, and determining whether the application system needs to be subjected to capacity expansion or capacity reduction according to the capacity calculation result.
It should be noted that the structure shown in fig. 1 is only an example, and the embodiment of the present invention is not limited thereto.
Based on the above description, fig. 2 exemplarily shows a flow of a method for determining system capacity according to an embodiment of the present invention, and the flow may be performed by an apparatus for determining system capacity.
As shown in fig. 2, the process specifically includes:
And 203, determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem.
In step 201, in the embodiment of the present invention, the acquisition module may acquire the current operation information of each service subsystem in the application system in real time, for example, the acquisition module may be a single server or a server cluster, and the single server or the server cluster is used to acquire the current operation information of each service subsystem. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as cloud service, cloud computing, cloud storage, network service, an artificial intelligence platform, and the like. Of course, the acquisition module may also manually collect the current operation information of each service subsystem and classify and sort the information. In practical application scenarios, the embodiment of the present invention does not limit this. When the current operation information of each service subsystem is collected, for example, the current operation information of each service subsystem is collected once every 1 second or once every 2 seconds, the current operation information can be set according to the needs of the actual service scene, which is not limited in the embodiment of the present invention. The current operation information may include real-time information of each capacity influence factor of each service subsystem, for example, the collected current operation information is described by taking the service subsystem as an online subsystem and a batch task subsystem as examples, and the current operation information may include an actual request amount (TPS) of the online system, execution time of each online transaction, a batch task amount of the batch task system, execution time of the batch task, a host CPU load, a memory load, a hard disk I/O load in the system, a storage service usage (such as database capacity, database access times, and the like), a machine number, and the like. After the current operation information of each service subsystem is collected, the collected current operation information of each service subsystem is stored in a database.
In step 202 and step 203, for at least one service subsystem, real-time information of each capacity influencing factor may be determined from the current operation information of the service subsystem based on each capacity influencing factor of the capacity calculation model determined by the historical operation information of the service subsystem and the weight of each capacity influencing factor, and the real-time information of each capacity influencing factor is input into the capacity calculation model of the service subsystem, so as to determine the system capacity of the service subsystem. Because the system capacity is automatically calculated by combining the real-time information of each capacity influence factor and the capacity calculation model, excessive manual intervention can be avoided, time and labor consumed by manually determining the system capacity can be reduced, the real-time performance and accuracy of system capacity determination can be ensured, and the problems of low evaluation efficiency and low accuracy caused by manually evaluating the system capacity in the prior art can be solved. The capacity influencing factors can include first influencing factors and second influencing factors. The weight of each capacity influence factor can be determined based on the analytic hierarchy process and the historical operation information of the application system, and the specific process is as follows: analyzing and processing historical operation information of the application system based on an analytic hierarchy process, and determining each first influence factor as a scheme layer, each second influence factor as a criterion layer and system capacity as a target layer; and determining a first weight of each first influence factor and a second weight of each second influence factor based on an analytic hierarchy process. The historical operation information of the application system is analyzed and processed through an analytic hierarchy process, and the first weight of each first influence factor and the second weight of each second influence factor can be determined quickly and accurately, so that a subsequent capacity calculation model can determine the system capacity of the service subsystem according to the first weight of each first influence factor and the second weight of each second influence factor. Wherein, each first influencing factor can comprise a single machine processing capacity, a machine number and a thread number; each second influencing factor may include disaster recovery capability, machine load, storage load, and traffic load.
Illustratively, the embodiment of the present invention describes a process for determining the system capacity of the business subsystem by taking the application system as a bank deposit IT system as an example. The bank deposit IT system can comprise an online business subsystem and a batch business subsystem. It should be noted that, the scope of the application system is not limited in the embodiment of the present invention, and in other service application scenarios, the application systems corresponding to other service application scenarios may also use the technical solution provided by the present invention to implement the calculation of the system capacity.
Determining the maximum system capacity of the service subsystem according to the following formula (1), wherein the formula (1) is specifically as follows:
a... (1) maximum system capacity ÷ (load condition × load weight) · maximum traffic processing capacity ÷ (load condition × load weight)
Determining the capacity of the disaster recovery system of the service subsystem according to a formula (2), wherein the formula (2) is specifically as follows:
the disaster tolerance system capacity is the weight of the disaster tolerance capability x the maximum system capacity x the disaster tolerance capability
The load condition comprises machine load, storage load and traffic load, and the disaster tolerance capability is the ratio of the number of available machines to the total number of machines.
It should be noted that, if the service subsystem is an online service subsystem, the maximum service processing capability of the online service subsystem is determined according to the following formula (3). The formula (3) is specifically:
the maximum TPS request amount (Transactions Per Second) represents the maximum number of requested Transactions Per unit time.
If the service subsystem is a batch service subsystem, determining the maximum service processing capacity of the batch service subsystem according to the following formula (4). The formula (4) is specifically:
the storage performance coefficient and the network delay coefficient are constants.
In addition, in the specific implementation process of the embodiment of the invention, the dynamic system capacity of the service subsystem can be quickly and accurately determined according to the current operation information of the service subsystem. The dynamic system capacity of the service subsystem may be determined, for example, by monitoring the real-time request volume or batch task throughput of the service subsystem.
Illustratively, the following description continues with the application system as a bank deposit IT system, and a specific implementation process of the capacity calculation module involved in the embodiment of the present invention is described.
Because the bank deposit IT system can comprise an online subsystem and a batch task subsystem, the capacity calculation model related to the technical scheme of the invention can also be composed of two parts, namely an online subsystem capacity calculation model and a batch task subsystem capacity calculation model. It should be noted that the calculation factors depended on by the capacity calculation model according to the technical solution of the present invention mainly come from the collected data, such as the number of machines, the processable capacity of a single-instance application program, and the processing thread of a single instance. The calculation factors occupy different influence weights respectively, and different influences are generated on the output result of the capacity calculation model. Therefore, before executing the capacity calculation model, the respective influence weights of different calculation factors need to be obtained by combining with an actual business scene, and the capacity of the bank deposit IT system can be obtained by combining with the capacity calculation model for calculation.
The construction process of the capacity calculation model specifically comprises the following steps:
step 1: and analyzing the influence weight of the influence factor of the bank deposit IT system by using an Analytic Hierarchy Process (AHP) to obtain the influence factor and the influence weight of the capacity calculation model.
First, a hierarchical structure model in the analytic hierarchy process is introduced, referring to fig. 3, and fig. 3 is a schematic diagram of a hierarchical structure model according to an embodiment of the present invention. The hierarchical model comprises a destination layer (system capacity), a criterion layer (A1 disaster tolerance capability, A2 machine load, A3 database load, A4 system load) and a scheme layer (B1 machine number, B2 single machine processing capability, B3 configuration thread number).
The process of determining the impact factors and their impact weights using analytic hierarchy process is described below.
a. Determining an impact factor of the capacity calculation model.
When determining the influence weight among the factors of each layer, if only qualitative results (such as a1 disaster tolerance ratio of 0.3, a2 machine load ratio of 0.2, etc.) are obtained, the results calculated by the capacity calculation model are not easily approved by users or are easily inaccurate. Therefore, the embodiment of the invention adopts a consistent matrix method in the analytic hierarchy process, namely: all factors are not put together for comparison, but are compared with each other two by two; relative dimensions are adopted at this time to reduce the difficulty of comparing different factors of the properties with each other as much as possible so as to improve the accuracy. The consistent matrix method is a comparison showing the relative importance of all factors of the current layer to a certain factor of the previous layer (such as a criterion layer or a target layer). As shown in Table 1, element a of the uniform matrix methodijThe ith factor is shown relative toAs a result of the j factor comparison, this value is given using the 1-9 Scale method of Santy.
TABLE 1
b. By comparing every two of the consistent matrix methods, the key several influence factors in the system capacity calculation model are the number of machines, the processing capacity of a single machine, and the number of threads configured by the application example deployed on the single machine. After determining the influence factors affecting the capacity calculation formula, the respective weight ratios of the influence factors need to be analyzed. In the embodiment of the present invention, the number of machines is defined as B1, the stand-alone processing capability is defined as B2, and the number of threads configured by the application instance is defined as B3.
c. And determining the influence weight of the influence factor of the capacity calculation model.
Normalizing the factors (machine number B1, stand-alone processing capability B2 and configuration thread number B3) of each layer of the B layer to obtain a normalized vector W, namely W ═ W1,W2,…,WnAre multiplied by
WiAnd (3) representing the relative importance of the ith factor of the same layer to the factor of the previous layer, calculating the synthetic weight of each element (namely the influence factor) in each layer to the total target, and finally obtaining the total hierarchical ordering of the B layer.
In view of this, the three factors in layer B rank the total target (system capacity) as a1, a2, a3, and then each factor in layer B ranks factor A in layer A asjIs b1j,b2j,…bnj. The overall hierarchical ordering of the B layers is thus:
that is, the ith factor of the B layer has the weight of the total target
Because each factor in the layer B is obtained by comparing each factor in the layer A pairwise, the influence weight of each factor in the layer A can be reversely deduced through the influence weight of each factor in the layer B, and the influence weight of the ith factor in the layer A on the total target (system capacity) can be obtained.
d. Since there are other qualitative rules controlling the factors in the layer a (e.g. the rule of high availability of the service system, the rule of load requirement, etc.) in the actual operation of the system, the influence weight of the factors in the layer a needs to be obtained. The embodiment of the invention adopts a control variable method, for example, when the influence weight of A1 is calculated, the values of B2 of the single machine processing capacity and B3 of the configuration thread number are not changed. Because each factor in the layer B is obtained by comparing each factor in the layer A in pairs, the influence weight of each factor in the layer A can be reversely deduced through the influence weight of each factor in the layer B, namely the influence weight of A1 on B1 can be calculated according to the ith factor in the layer B (since the values of B2 and B3 are not changed, the influence weight can be considered as the influence weight of A1 on the total target). Wherein the influence weight of the ith factor of the B layer on the total target is
In addition, because each factor in the B layer is obtained by comparing each factor in the a layer two by two, a plurality of influence weights of a1 on B1 are calculated, and thus a plurality of influence weights of a1 need to be weighted and averaged to obtain a final influence weight of a1, that is, the influence weight proportion coefficient of the influence factor a1 is K1. And the calculation method of the other factors in the layer A is analogized. Meanwhile, a calculation formula of the weight of each factor in the layer A can be obtained, and the calculation formula is as follows:
f1＝A×Ki+(P+Q)................................................(6)
wherein f is1The weight of the 1 st factor of the layer B on the total target is shown, A represents each factor in the layer A, and KiThe weight of a certain factor in the layer A, i.e. the weight-to-weight ratio coefficient of a certain factor in the layer A, and P is a fixed value corresponding to the single-machine processing capability B2, which is a common ruleThe number, Q, is a fixed value corresponding to the configuration thread number B3 and is a constant.
When calculating the influence weight of a1, the values of the number of devices B1 and the number of configuration threads B3 may be controlled to be unchanged, that is, the influence weight of a1 on B2 may be calculated from the ith factor in the B layer (since the values of B1 and B3 are unchanged, the influence weight may be considered as the influence weight of a1 on the overall target), so that the final influence weight of a1, that is, the influence weight proportion coefficient of the influence factor a1 may be calculated by using the above calculation idea, and the method of calculating the remaining factors in the a layer may be analogized. Or when the influence weight of a1 is calculated, the values of the device number B1 and the stand-alone processing capacity B2 may be controlled to be unchanged, that is, the influence weight of a1 on B3 may be calculated according to the ith factor in the B layer (since the values of B1 and B2 are unchanged, the influence weight may be considered as the influence weight of a1 on the overall target), so that the final influence weight of a1, that is, the influence weight proportion coefficient of the influence factor a1 may be calculated by using the above calculation idea, and so on. Of course, when determining the influence weight of a2, A3, or a4, the above-mentioned technical solution of the embodiments of the present invention may also be used for calculation, and will not be described herein again.
Step 2: and constructing a capacity calculation model.
After determining the influence factors of the system capacity and the influence weights of the influence factors, a capacity calculation model can be constructed.
(1) And constructing an online system capacity calculation model.
a. And the value source of each layer of influence factors.
Since the stand-alone processing performance of a bank deposit IT system is mainly determined by the system program (i.e. the transaction request M that N servers can process within 1 second should approach a constant). The capacity of the online subsystem can thus be considered equal to the amount of requests (TPS) that the system can handle. Through analysis, the capacity of the online subsystem is mainly determined by the request Traffic (TPS) of the online subsystem, the request processing delay, the host CPU load, and the system disaster tolerance policy (disaster tolerance capability). Fig. 4 is a schematic diagram illustrating a relationship between a system capacity and a request number in an online subsystem capacity calculation model according to an embodiment of the present invention. As shown in fig. 4, when the request amount is larger, the more requests the online subsystem needs to process, the more CPU computing resources are consumed correspondingly, and the host CPU load is increased accordingly.
b. And substituting the influence factors of each layer and the influence weights of the influence factors of each layer into a formula for calculation.
Fig. 5 is a schematic diagram illustrating a relationship between system capacity and a number of faulty machines in an on-line subsystem capacity calculation model according to an embodiment of the present invention. As shown in fig. 5, when the online subsystem fails in the cluster, the corresponding cluster processing performance also decreases (when there are N machines in the cluster, and when there are N machines that fail, the processing performance in the cluster is lost by N/N), so that when the machine fails abnormally, the evaluation and calculation of the online subsystem capacity calculation model on the system capacity also may be affected, and therefore, the system disaster tolerance policy (disaster tolerance capability) is also a factor that needs to be considered by the online subsystem capacity calculation model.
According to the exploration relation obtained by the analytic hierarchy process, the influence weight of each layer of influence factors is sleeved, and the capacity formula model of the online subsystem can be deduced to be approximate to:
the maximum TPS request volume (Transactions Per Second) represents the maximum number of requested Transactions in a unit time, the load condition may include machine load, storage load, and traffic load, the load weight is an influence weight corresponding to each load, and α is an influence weight of the single machine processing capability.
As a disaster tolerance scenario needs to be considered, that is, a scenario in which a machine failure in a cluster is unavailable occurs, a real-time capacity model of the online subsystem is as follows:
wherein, K1The influence weight of the disaster tolerance capability in the disaster tolerance scene.
In view of this, for the capacity calculation module of the online subsystem, the current operation information of the online subsystem is input to the capacity calculation module of the online subsystem for capacity calculation, so as to obtain the real-time dynamic capacity and the real-time maximum capacity of the online subsystem, and the real-time dynamic capacity, the real-time maximum capacity, and the ratio of the real-time dynamic capacity to the real-time maximum capacity of the online subsystem are fed back to the user, so that the user can determine whether to perform capacity expansion or capacity reduction according to the result fed back by the capacity platform. Referring to fig. 6, a schematic diagram of a capacity calculation result of an online subsystem according to an embodiment of the present invention is provided, according to which a user can intuitively know real-time dynamic changes of a system capacity, so that the user can make a decision.
(2) And constructing a batch task system capacity calculation model.
a. And the value source of each layer of influence factors.
Because the batch task subsystem of the bank deposit system mainly processes batch entry, account checking, interest improvement and the like, the processing time of batch tasks has a large relationship with the capacity of the batch task subsystem, the number of accounts processed in the tasks, the computing capacity of batch threads, the storage service performance/capacity, the network transmission time consumption and the like. The batch task subsystem capacity is related to each capacity influence factor (which may include machine load, database load, disaster tolerance capability, number of machines, stand-alone processing capability, number of configuration threads, etc.) and the influence weight of each capacity influence factor. Referring to fig. 7, a schematic diagram of a relationship between a capacity of a batch task subsystem and a running time of a batch task according to an embodiment of the present invention is shown, and it can be known from the schematic diagram that the running time of the batch task subsystem is longer when the capacity of the batch task subsystem is larger. Referring to fig. 8, a schematic diagram of a relationship between the number of accounts involved in the batch task and the processing time of the batch task according to an embodiment of the present invention is provided, and according to the schematic diagram, when the number of accounts involved in the batch task is gradually increased, the processing time of the batch task is gradually increased, and then gradually becomes stable.
b. And substituting the influence factors of each layer and the influence weights of the influence factors of each layer into a formula for calculation.
According to the exploration relation obtained by the analytic hierarchy process, the influence weight of each layer of influence factors is sleeved, and then a capacity formula model of the batch task subsystem can be deduced as follows:
maximum capacity ═ stand-alone batch task processing capacity × machine number ÷ (load condition × load weight.) (10)
The load conditions may include machine loads, storage loads, and traffic loads, the load weight is an influence weight corresponding to each load condition, and the storage performance coefficient and the network delay coefficient are constants.
If there is change optimization, the capacity calculation model can also perform calculation correction according to the collected operation information, and revise the constant value again, so as to obtain the capacity calculation model of the batch task subsystem as follows:
a
Wherein, the batch processing capacity can be considered as a constant in a normal state (i.e. no change optimization is performed during the operation of the system).
Because a disaster tolerance scenario needs to be considered, that is, a scenario in which a machine failure in a cluster is unavailable occurs, a real-time capacity model of the batch task subsystem is as follows:
wherein, K1The influence weight of the disaster tolerance capability in the disaster tolerance scene.
In view of this, the volume calculation module of the batch task subsystem may calculate the processing capacity of the batch subsystem in the bank deposit IT system according to the collected current operation information of the bank deposit IT system, and may also perform volume calculation on the current operation information of the batch task subsystem according to the volume calculation module of the batch task subsystem, so as to obtain the real-time dynamic volume and the real-time maximum volume of the batch task subsystem, and feed back the real-time dynamic volume, the real-time maximum volume, and the ratio of the real-time dynamic volume to the real-time maximum volume of the batch task subsystem to the user, so that the user may determine whether to perform volume expansion or volume reduction according to the result fed back by the volume platform. The real-time dynamic capacity of the batch task subsystem can be determined by monitoring the current operation information (such as batch task processing amount, batch task execution time consumption, batch task processing capacity and the like) of the batch task subsystem. Referring to fig. 9, a schematic diagram of a capacity calculation result of a batch task subsystem according to an embodiment of the present invention is provided, according to which a user can intuitively know real-time dynamic changes of a system capacity, so that the user can make a decision.
The above embodiments show that the real-time information of each capacity influencing factor is determined from the current operation information of the service subsystem for at least one service subsystem by acquiring the current operation information of each service subsystem in the application system in real time. And determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem. The system capacity is automatically calculated by combining the real-time information of each capacity influence factor, the weight of each capacity influence factor and the capacity calculation model, so that excessive manual intervention can be avoided, time and labor consumed by manually determining the system capacity can be reduced, the real-time performance and accuracy of system capacity determination can be ensured, and the problems of low evaluation efficiency and low accuracy caused by manual evaluation of the system capacity in the prior art can be solved.
Based on the same technical concept, fig. 10 exemplarily shows an apparatus for determining system capacity, which may perform the flow of the method for determining system capacity according to an embodiment of the present invention.
As shown in fig. 10, the apparatus includes:
the acquisition unit 1001 is used for acquiring the current operation information of each service subsystem in the application system;
a processing unit 1002, configured to determine, for at least one service subsystem, real-time information of each capacity impact factor from current operation information of the service subsystem; determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem; the capacity calculation model is provided with weights of various capacity influence factors determined by historical operation information of the service subsystem.
Optionally, the processing unit 1002 is specifically configured to:
determining the weight of each capacity influencing factor according to the following mode:
analyzing and processing the historical operation information of the application system based on an analytic hierarchy process, and determining each first influence factor as a scheme layer, each second influence factor as a criterion layer and the system capacity as a target layer; each capacity influencing factor comprises each first influencing factor and each second influencing factor;
and determining the first weight of each first influence factor and the second weight of each second influence factor based on an analytic hierarchy process.
Optionally, the processing unit 1002 is specifically configured to:
each first influence factor comprises single machine processing capacity and machine number; the second influence factors include disaster tolerance capability, machine load, storage load and traffic load.
Optionally, the processing unit 1002 is specifically configured to:
determining a maximum system capacity of the service subsystem according to the following formula (1);
the formula (1) is:
a... (1) maximum system capacity ÷ (load condition × load weight) · maximum traffic processing capacity ÷ (load condition × load weight)
Determining the capacity of the disaster recovery system of the service subsystem according to the following formula (2);
the formula (2) is:
the disaster tolerance system capacity is the weight of the disaster tolerance capability x the maximum system capacity x the disaster tolerance capability
The load condition comprises machine load, storage load and traffic load, and the disaster tolerance capability is the ratio of the number of available machines to the total number of machines.
Optionally, the service subsystem is an online service subsystem;
the processing unit 1002 is specifically configured to:
determining the maximum service processing capacity of the online service subsystem according to the following formula (3);
the formula (3) is:
the maximum TPS request amount (Transactions Per Second) represents the maximum number of requested Transactions Per unit time.
Optionally, the service subsystem is a batch service subsystem;
the processing unit 1002 is specifically configured to:
determining the maximum service processing capacity of the batch service subsystem according to the following formula (4);
the formula (4) is:
the storage performance coefficient and the network delay coefficient are constants.
Optionally, the processing unit 1002 is further configured to:
and determining the dynamic system capacity of the service subsystem according to the current operation information of the service subsystem.
Based on the same technical concept, an embodiment of the present invention provides a computing device, including:
a memory for storing a computer program;
and the processor is used for calling the computer program stored in the memory and executing the method for determining the system capacity according to the obtained program.
Based on the same technical concept, embodiments of the present invention provide a computer-readable storage medium storing a computer-executable program for causing a computer to execute a method of determining system capacity.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present application and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A method of determining system capacity, comprising:
collecting current operation information of each service subsystem in an application system;
for at least one service subsystem, determining real-time information of each capacity influence factor from current operation information of the service subsystem;
determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem; the capacity calculation model is provided with weights of various capacity influence factors determined by historical operation information of the service subsystem.
2. The method of claim 1, wherein the weight of each capacity impact factor is determined according to:
analyzing and processing the historical operation information of the application system based on an analytic hierarchy process, and determining each first influence factor as a scheme layer, each second influence factor as a criterion layer and the system capacity as a target layer; each capacity influencing factor comprises each first influencing factor and each second influencing factor;
and determining the first weight of each first influence factor and the second weight of each second influence factor based on an analytic hierarchy process.
3. The method of claim 2, wherein each of said first influencing factors comprises stand-alone processing power and machine number; the second influence factors include disaster tolerance capability, machine load, storage load and traffic load.
4. The method of claim 3, wherein determining the system capacity of the service subsystem based on the real-time information of the capacity impact factors and a capacity calculation model of the service subsystem comprises:
determining a maximum system capacity of the service subsystem according to the following formula (1);
the formula (1) is:
a... (1) maximum system capacity ÷ (load condition × load weight) · maximum traffic processing capacity ÷ (load condition × load weight)
Determining the capacity of the disaster recovery system of the service subsystem according to the following formula (2);
the formula (2) is:
the disaster tolerance system capacity is the weight of the disaster tolerance capability x the maximum system capacity x the disaster tolerance capability
The load condition comprises machine load, storage load and traffic load, and the disaster tolerance capability is the ratio of the number of available machines to the total number of machines.
5. The method of claim 4, wherein the service subsystem is an online service subsystem;
determining the maximum service processing capacity of the online service subsystem according to the following formula (3);
the formula (3) is:
the maximum TPS request amount (Transactions Per Second) represents the maximum number of requested Transactions Per unit time.
6. The method of claim 4, wherein the service subsystem is a bulk service subsystem;
determining the maximum service processing capacity of the batch service subsystem according to the following formula (4);
the formula (4) is:
the storage performance coefficient and the network delay coefficient are constants.
7. The method of any of claims 1 to 6, further comprising:
and determining the dynamic system capacity of the service subsystem according to the current operation information of the service subsystem.
8. An apparatus for determining system capacity, comprising:
the acquisition unit is used for acquiring the current operation information of each service subsystem in the application system;
the processing unit is used for determining real-time information of each capacity influence factor from the current operation information of at least one service subsystem; determining the system capacity of the service subsystem according to the real-time information of each capacity influence factor and the capacity calculation model of the service subsystem; the capacity calculation model is provided with weights of various capacity influence factors determined by historical operation information of the service subsystem.
9. A computing device, comprising:
a memory for storing a computer program;
a processor for calling a computer program stored in said memory, for executing the method of any one of claims 1 to 7 in accordance with the obtained program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer-executable program for causing a computer to execute the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011168930.XA CN112348666A (en) | 2020-10-28 | 2020-10-28 | Method and device for determining system capacity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011168930.XA CN112348666A (en) | 2020-10-28 | 2020-10-28 | Method and device for determining system capacity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112348666A true CN112348666A (en) | 2021-02-09 |
Family
ID=74358832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011168930.XA Pending CN112348666A (en) | 2020-10-28 | 2020-10-28 | Method and device for determining system capacity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112348666A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115080363A (en) * | 2022-08-23 | 2022-09-20 | 中国中金财富证券有限公司 | System capacity evaluation method and device based on service log |
| US12009660B1 (en) | 2023-07-11 | 2024-06-11 | T-Mobile Usa, Inc. | Predicting space, power, and cooling capacity of a facility to optimize energy usage |
-
2020
- 2020-10-28 CN CN202011168930.XA patent/CN112348666A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115080363A (en) * | 2022-08-23 | 2022-09-20 | 中国中金财富证券有限公司 | System capacity evaluation method and device based on service log |
| US12009660B1 (en) | 2023-07-11 | 2024-06-11 | T-Mobile Usa, Inc. | Predicting space, power, and cooling capacity of a facility to optimize energy usage |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10942781B2 (en) | Automated capacity provisioning method using historical performance data | |
| US5537542A (en) | Apparatus and method for managing a server workload according to client performance goals in a client/server data processing system | |
| US10402225B2 (en) | Tuning resources based on queuing network model | |
| EP3226134B1 (en) | A method and system for scaling resources, and a computer program product | |
| CN109783237B (en) | Resource allocation method and device | |
| US10748072B1 (en) | Intermittent demand forecasting for large inventories | |
| CN109120715A (en) | Dynamic load balancing method under a kind of cloud environment | |
| CN112579273B (en) | Task scheduling method and device and computer readable storage medium | |
| CN112348666A (en) | Method and device for determining system capacity | |
| Upadhyay et al. | Scaled conjugate gradient backpropagation based sla violation prediction in cloud computing | |
| US20160080267A1 (en) | Monitoring device, server, monitoring system, monitoring method and program recording medium | |
| CN110619580A (en) | Whole-market multi-variety fund financing management system based on intelligent data processing platform | |
| US10824956B1 (en) | System and method for price estimation of reports before execution in analytics | |
| CN109117269A (en) | A kind of distributed system dispatching method of virtual machine, device and readable storage medium storing program for executing | |
| CN115080373A (en) | Performance detection method, device, equipment and medium of power distribution terminal operating system | |
| US20210217083A1 (en) | Method and system for optimizing resource redistribution | |
| CN114911613A (en) | Cross-cluster resource high-availability scheduling method and system in inter-cloud computing environment | |
| WO2018182442A1 (en) | Machine learning system and method for generating a decision stream and automonously operating device using the decision stream | |
| CA3182205A1 (en) | Financial risk assessment | |
| CN107844496B (en) | Statistical information output method and device | |
| CN116185797A (en) | Method, device and storage medium for predicting server resource saturation | |
| CN110570307A (en) | Whole-market multi-variety fund financing management system based on intelligent strategy platform | |
| Banicescu et al. | Towards the robustness of dynamic loop scheduling on large-scale heterogeneous distributed systems | |
| CN112685390B (en) | Database instance management method and device and computing equipment | |
| CN112181443B (en) | Automatic service deployment method and device and electronic equipment |
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 |