TWI502371B - Cloud Evaluation System and Its Method - Google Patents

Description

Cloud computing evaluation system and method thereof

The invention relates to a cloud computing evaluation system and a method thereof, in particular to a cloud computing evaluation system capable of evaluating a system for planning to enter a cloud data center and suggesting a total value of cloud resources to be self-adapted. And its method.

In the era of global competition, both product manufacturers and service providers must effectively use their resources to maintain their competitive advantage. Resources generally refer to the manpower, equipment, capital, and capital that enterprises can use in their operations. How to effectively coordinate the operation of resources and make the company's operations more competitive is a major issue in the information society competition at this stage.

In general, the company's operating system includes direct equipment and indirect support equipment. Direct equipment systems generally refer to equipment resources that have direct contact with production or service targets, while indirect support systems are those that do not directly contact production or service targets. During the operation of the equipment, whether the direct or indirect equipment may cause damage to the parts due to neglect of maintenance, human negligence or improper use of software resources, the production or service activities may be unexpectedly shut down, causing great losses to the enterprise.

According to the iTHome 2013 CIO survey, the proportion of enterprise-based server virtualization is ranked first in software investment projects, indicating that companies are turning to investment virtualization technology to replace some physical servers. However, enterprises are investing in virtualization. An important issue is that enterprises must effectively evaluate the actual resource requirements of the system in order to determine the capacity of the cloud platform to be built, and ensure the future operation and development of the enterprise. The present invention is to propose a cloud-based evaluation system and method. To provide more accurate cloud resource requirements for servers.

In view of the above-mentioned problems of the prior art, the object of the present invention is to provide a resource usage data that can be fed back according to the resource usage data of the system and the cloud data center. In order to effectively improve the error of cloud resource estimation, it provides a cloud-based evaluation system and method for cloud resource demand recommendations for future systems that are stationed in the cloud data center.

According to the purpose of the present invention, a cloud computing evaluation system includes: a data receiving and analyzing module, configured to receive at least one hardware usage data of a system including at least one server, and the cloud data center The hardware of the server uses the data, and analyzes each hardware usage data to generate corresponding resource usage data respectively; a data database for storing each resource usage data; and a behavior model building module for self-data data The resource usage data is obtained in the library to respectively establish a corresponding resource usage behavior model; an evaluation learning module is used for the resource usage behavior model of the server before entering the cloud data center and the resource usage after entering the cloud data center The behavior model is analyzed, and one of the various cloud resources of the server is used to evaluate the error parameter value; a server cloud resource demand prediction module is used to obtain the resource usage behavior model of the server from the data database, and evaluate The cloud resource requirements required by the server to enter the cloud data center; and a system cloud resource Assessment module for aggregate amount of cloud resource requirements of each server separately, and each resource based on the use of behavioral models of the server at a different time interval, the proposed resource requirements of the cloud systems.

Preferably, the evaluation learning module further includes a resource usage behavior reading unit and a resource evaluation error calculation unit; the resource usage behavior reading unit is configured to read the resource usage behavior of the server before and after entering the cloud data center. The resource evaluation error calculation unit is configured to analyze the resource usage behavior model of the server to perform resource estimation error calculation, and calculate the evaluation error parameter value after comparing the resource usage data before and after the cloud data center.

Preferably, the server cloud resource demand prediction module further comprises a server resource usage behavior comparison unit and a server cloud resource requirement calculation unit; the server resource usage behavior comparison unit is configured to search according to the role of the server. The server of the same role performs the comparison of the resource usage behavior models; the server cloud resource requirement calculation unit 152 is configured to use the resource usage data of the server itself, and combine the evaluation error of the server with the similar resource usage behavior module. The parameter value is used to calculate the cloud resource requirements of the server.

According to the purpose of the present invention, a cloud computing evaluation method is further provided, which comprises the steps of: receiving at least one hardware usage data of a system including at least one server, and extracting and normalizing hardware usage data of all servers; Record data to a resource after using it for a resource Repository; respectively establish a resource usage behavior model corresponding to each resource usage data of all servers; perform a comparison of resource usage behavior models according to roles and resource types to determine a similar resource usage behavior model; The maximum value of the resource usage data and one of the similar resource behavior models are used to evaluate the error parameter value, calculate the cloud resource requirement of the server; and analyze the resource usage behavior model of all the servers in the system, and find all the servers according to the time interval. The eigenvalues with the largest time interval are summed up, and the maximum value is found from the eigenvalues of all time intervals as the recommended value of the overall cloud resource requirement of the system.

Preferably, the cloud computing evaluation method of the present invention further comprises the following steps: obtaining resource usage data of a server stationed in the cloud data center in a specific period according to the set learning frequency, and establishing a server in the cloud data. The resource usage behavior model after the center; according to the cloud evaluation number of the server, to the data database to obtain the resource usage behavior model of the server before entering the cloud data center; and according to the server before entering the cloud data center and after entering the cloud data center The resource usage behavior model is analyzed to calculate the evaluation error parameter value of the server resource usage behavior model.

Preferably, the resource usage behavior model includes a CPU resource usage model, a memory resource usage model, or other defined resource usage behavior models.

Preferably, the role includes a web role, a DB role, or other defined roles; the resource type includes a CPU resource type, a memory resource type, or other defined resource type.

Preferably, the cloud resource requirement includes at least a CPU configuration requirement value, a memory configuration requirement value, or other defined resource requirement values.

Preferably, the overall cloud resource requirement recommendation value includes the overall CPU resource requirement value of the system, the overall memory resource requirement value, the individual CPU resource requirement value of the server, the memory resource requirement value, or the overall requirement of other defined resources. value.

According to the above, the cloud computing evaluation system and method thereof have the following one or more advantages:

1. The present invention provides an overall cloud resource demand assessment mechanism for a system, which can establish a resource usage behavior model according to each resource usage data of each server in the system, and analyze usage behavior of all servers in the system, and provide overall cloud resource requirements of the system. Suggest.

2. The present invention provides a self-adapting learning mechanism that utilizes the cloud data center to return The federated virtual machine resource usage data is combined with the role-based resource usage behavior model to self-adjust the evaluation error parameter values of each resource.

3. The present invention provides a resource usage behavior model establishment mechanism, and establishes a role-based resource usage behavior model for resource usage data and server information of a server (physical machine or virtual machine number).

1‧‧‧Cloud Assessment System

11‧‧‧Data Receiving and Analysis Module

12‧‧‧Data Database

13‧‧‧ Behavioral Model Building Module

14‧‧‧Evaluation Learning Module

141‧‧‧Resource usage behavior reading unit

142‧‧‧Resource assessment error calculation unit

15‧‧‧Server Cloud Resource Demand Forecasting Module

151‧‧‧Server resource usage behavior comparison unit

152‧‧‧Server Cloud Resource Requirements Calculation Unit

16‧‧‧System Cloud Resource Evaluation Module

21~28‧‧‧Steps

3‧‧‧System hardware usage data

4‧‧‧Cloud data center server hardware usage data

6‧‧‧Server collection resources

61‧‧‧CPU resources

612‧‧‧Web role

614‧‧‧DB role

62‧‧‧ Memory resources

7‧‧‧Server resource usage data

71‧‧‧Server CPU resource usage data

8‧‧‧Resource usage behavior model

81‧‧‧Resource usage behavior model before server for cloud assessment

82‧‧‧Resource usage behavior model after the server is stationed in the cloud data center

83‧‧‧CPU resource usage behavior model

831‧‧‧CPU-Web Resource Usage Model-1

832‧‧‧CPU-Web Resource Usage Behavior Model-N

833‧‧‧CPU-DB Resource Usage Model-1

834‧‧‧CPU-DB Resource Usage Behavior Model-M

9‧‧‧System Cloud Assessment Report

Figure 1 is a block diagram of the cloud computing evaluation system of the present invention.

Figure 2 is a block diagram of the evaluation learning module of the present invention.

Figure 3 is a block diagram of the server cloud resource demand prediction module of the present invention.

Figure 4 is a flow chart of the cloud computing evaluation method of the present invention.

Figure 5 is a schematic diagram of the cloud computing evaluation application and resource usage data input of the present invention.

Figure 6 is a schematic diagram of the establishment of the resource usage behavior model of the present invention.

Figure 7 is a simplified schematic diagram of the evaluation learning mechanism of the present invention.

Figure 8 is a simplified schematic diagram of the server cloud resource demand forecasting operation of the present invention.

Figure 9 is a simplified schematic diagram of the cloud computing resource evaluation operation of the system of the present invention.

The technical features, contents, and advantages of the present invention, as well as the advantages thereof, can be understood by the present inventors, and the present invention will be described in detail with reference to the accompanying drawings. The subject matter is only for the purpose of illustration and description. It is not intended to be a true proportion and precise configuration after the implementation of the present invention. Therefore, the scope and configuration relationship of the attached drawings should not be interpreted or limited. First described.

The invention provides a cloud-based evaluation system and a method for self-adaptation of a role, which can evaluate the system that is planned to be stationed in the cloud data center, and propose the total value of the cloud resource to be configured.

The present invention provides a cloud-based evaluation system 1 that adapts itself to a role, as shown in FIG. 1 , which includes a data receiving and analyzing module 11 , a data database 12 , a behavior model building module 13 , an evaluation learning module 14 , and a servo . The cloud resource demand prediction module 15 and the system cloud resource evaluation module 16. The data receiving and analyzing module 11 can receive the existing system (including one or more The hardware usage data of the server and the hardware usage data of the server that has been stationed in the cloud data center are analyzed to generate corresponding resource usage data, and are stored in the data database 12. The behavior model building module 13 can obtain the resource usage data of the server in the data repository 12 and establish a corresponding resource usage behavior model. The evaluation learning module 14 can analyze the resource usage behavior model of the server before entering the cloud data center and the resource usage behavior model after entering the cloud data center, and recalculate the evaluation error parameter values of the cloud resources of the server. . The server cloud resource requirement prediction module 15 can obtain a resource usage behavior model of the server from the data repository 12 and evaluate various cloud resource requirements required for the server to enter the cloud data center; wherein the cloud resource may include A resource defined by a computing resource (CPU), memory, or other cloud data center. The system cloud resource evaluation module 16 can aggregate the amount of cloud resource requirements of all servers in a system, and propose various cloud resource requirements of the system according to various resource usage behaviors of different servers in different time intervals.

The evaluation learning module 14 provided by the present invention further includes a resource usage behavior reading unit 141 and a resource evaluation error calculation unit 142, as shown in FIG. The resource usage behavior reading unit 141 can read a resource usage behavior model of the server before and after entering the cloud data center. The resource evaluation error calculation unit 142 can analyze the resource usage behavior model of the server to perform resource estimation error calculation (comparing the resource usage data before and after entering the cloud data center), and calculate the evaluation error parameter value.

The server cloud resource requirement prediction module 15 of the present invention further includes a server resource usage behavior comparison unit 151 and a server cloud resource requirement calculation unit 152, as shown in FIG. The server resource usage behavior comparison unit 151 can search for servers of the same role according to the role of a server and perform resource usage behavior model comparison. The server cloud resource requirement calculation unit 152 can calculate the cloud resource requirement of the server according to the resource usage data of the server itself and the evaluation error parameter value of the server using the similar resource usage behavior.

The present invention proposes a cloud-based assessment method for self-adaptation of roles, as shown in FIG. 4, which includes a series of steps from inputting system resource usage data files to completing overall cloud resource resource requirement evaluation. First, in step 21, receiving hardware usage data of a system (including one or more servers), resource usage data of each server in the system will be captured and recorded to the data database 12; wherein the resource usage data At least CPU and memory usage data, or other defined resources Source usage data. Then, in step 22, for each resource usage data of each server in the system, each resource usage behavior model is established (for example, if the resource data includes CPU and memory usage data, each server will be separately Establish a CPU resource usage behavior model and a memory resource usage behavior model). In step 23, the role of the server (such as a web role or a DB role) is compared with the resource type to be compared (for example, a CPU resource type or a memory resource type), and the server resource is determined. Use a collection of behavioral models that use behavioral models that are similar. Then, in step 24, the cloud resource requirements of the server are calculated according to the maximum value of the server resource usage data and the evaluation error parameter value provided by the similar resource usage behavior model set. Step 25: Perform an overall cloud resource demand assessment operation, which can analyze resource usage behavior models of all servers in the system, find out the maximum eigenvalue of each server in the time interval according to the time interval, and perform eigenvalue summation. Then, find the maximum value from the eigenvalue plus total value of all time intervals as the recommended value of the overall cloud resource requirement of the system. In step 26, according to the set learning frequency, the cloud data center obtains resource usage data of the server stationed in the cloud data center in a specific period, and establishes a resource usage behavior model of the server after entering the cloud data center. Then, in step 27, for the server obtained in step 26, the resource usage behavior model of the server before entering the cloud data center is obtained from the data database. In step 28, the server uses the resource usage behavior model before the cloud data center to perform the analysis, and calculates the evaluation error parameter value of the resource group, and the evaluation error parameter value can be fed back to step 24.

Please refer to FIG. 5, which is a schematic diagram of cloud computing evaluation application and resource usage data input. The data receiving and analyzing module 11 can receive the system hardware usage data 3 uploaded by the user or the cloud data center server hardware usage data. 4. The system hardware usage data should contain one or more server information. Each server information should contain at least the operating system type and version used by the server and server resource usage data. The server resources should at least contain the CPU. Use data with both resources of memory. The cloud data center server hardware usage data 4 should contain one or more server information, and each server information should at least include the operating system type and version used by the server, server resource usage data, and the server cloud. The evaluation number; wherein the server resource should contain at least two data usage data of the CPU and the memory. The user-used system hardware usage data 3 will be parsed by the data receiving and analyzing module 11 and the resource usage data of each server will be generated, and each server will be assigned a server cloud computing evaluation. No. After that, the resource usage data of the server and the distributed server cloudization evaluation number are stored in the data repository 12. In addition, the data receiving and analyzing module 11 can obtain the cloud data center server hardware data file 4, and the data file should include the resource usage data of the server stationed in the cloud data center after completing the cloud computing evaluation in the past specific time interval. The resource usage data of each server will be stored in the data repository 12.

Referring to FIG. 6, a schematic diagram of the resource usage behavior model is shown. As described in (a) of FIG. 6, the behavior model building module 13 is executed according to the execution frequency (for example, every 30 minutes or 60 minutes). The data database 12 obtains the server resource usage data 7 of the server that needs to establish the resource usage behavior model (the server that needs to establish the resource usage behavior model includes each server in the system hardware usage data uploaded by the user, and the cloud The data center virtual machine hardware uses each server included in the data), and then the behavior model building module 13 generates a corresponding resource usage behavior model and writes it to the data repository 12. Figure 6 (b) illustrates the concept of establishing resource usage behavior. The resource usage data of each server will be cut according to the set time interval. For example, the server CPU resource usage data 71 is cut into 10 time intervals (t1, T2, t3, t4, t5, t6, t7, t8, t9, t10), and take the maximum resource usage data value from each time interval as the eigenvalue of the time interval. The resource usage behavior model establishment process is described as follows :

Step 1: Obtain resource usage data: obtain resource usage data D (resource) of a server, where resource can be CPU, memory or other defined resources, and all resource usage data collection time needs to be greater than a specific time (for example, 1 week) ).

Step 2: Cutting resource usage data: The resource usage data D (resource) will be cut according to the time interval setting value. For example, if the time interval setting value is 30 minutes, the resource usage data D (resource) will perform data division according to the 30-minute interval. If the resource usage data is collected once every 5 minutes, then each time interval will contain 6 (30 minutes / 5 minutes = 6) data values.

Step3: Establish a resource usage behavior model: the data values in each time interval can be represented as a series of data value sets {dx, dx+1, ..., dx+k}, and the maximum value in the data value set will be selected. For the eigenvalue d(ti) of the time interval, the eigenvalue combination of all time intervals is regarded as the server resource usage behavior model B (resource). For example, the resource usage behavior model can be expressed as: B ( resource )= { d ( t ₀ ), d ( t ₁ ), ..., d ( t _m )}, which indicates that the resource usage behavior model contains m+1 feature values.

See Figure 7 for a simplified diagram of the evaluation learning mechanism. The resource usage behavior reading unit 141 obtains the resource usage behavior model 82 after each server enters the cloud data center according to the set frequency (for example, every 8 hours or every 24 hours), and according to the servo of the server. The cloud computing evaluation number obtains the resource usage behavior model 81 before the cloud server performs the cloudization evaluation, and then delivers the above data to the resource evaluation error calculation unit. The resource evaluation error calculation unit 142 calculates the evaluation error parameter value according to the resource usage behavior model data of the pen server and the formula 1. The data behavior model data and the evaluation error parameter value are calculated as follows: the server has not been stationed in the cloud data center. The resource usage behavior model is defined as: B ( resource ) ^original ={ d ( t ₀ ) ^original , d ( t ₁ ) ^original ,......, d ( t _m ) ^original }; server stationed in the cloud data The resource usage behavior model after the center is defined as: B ( resource ) ^cloud ={ d ( t ₀ ) ^cloud , d ( t ₁ ) ^cloud ,......, d ( t _m ) ^cloud }; resource estimation error calculation The unit can derive the evaluation error parameter value of each server according to the following formula 1.

See Figure 8 for a simplified diagram of the server cloud resource demand forecasting operation. The server resource usage behavior comparison unit 151 receives a server resource usage behavior model data to be compared, and finds a similar resource usage behavior model according to the comparison parameter, and the comparison parameter includes similar value parameters and similar numbers. Parameters; wherein the similarity parameter is used to determine the similarity limit between the two resource usage behavior models (for example, if the similarity parameter is set to 90%, indicating that the two resources are compared using the behavioral model, the similarity value is greater than or equal to 90. % is considered to be similar, otherwise it is not similar, and the similar number parameter is used to determine to find several similar resource usage behavior models (for example, if the similar number parameter is set to 5, it means from all similar resources. In the behavioral model, select the top 5 most similar resource usage behavior models. If the number of similar resource usage behavior models is less than the similar number, only the existing similar numbers are taken. Finding a similar resource usage behavior model strategy is based on the resource type of the server (server collection resource 6, including CPU resource 61, memory resource 62 or other defined resource type, and role (eg, web role 612). , DB role 614 or other defined role)) to search. For example, if a server belongs to the Web role and the CPU resource usage behavior model 83 is to be compared, only the CPU-Web resource usage model-1 831 to the CPU-Web resource usage behavior model-N 832 will be searched (if it belongs to The DB role, at the same time, is to compare the CPU resource usage behavior model 83, and then search for the CPU-DB resource usage model-1 833 to the CPU-DB resource usage behavior model -M 834). The similarity value si,j is calculated as defined in Equation 2, i represents the server to be compared, j is represented as the server used for comparison, and m represents the number of time segments. After the comparison operation is completed, the cloud resource requirement calculation unit 152 calculates the average evaluation error value Δ B ^{c of the} server c based on all the error parameter values in the similar resource usage behavior model set obtained after the comparison and according to the formula 3. , where x represents a server with similar resource usage behavior. Then, according to formula 4, the cloud resource quantity value recommended by the server c is calculated. .

(Formula 2).

See Figure 9 for a simplified diagram of the System Cloud Resource Assessment job. The system cloud resource evaluation module 16 obtains the system number of the cloud computing evaluation that has not been completed according to the set frequency (for example, every 60 minutes), and obtains resources of all the servers of the system according to the system number to the data database. Use behavioral model data. After that, the system cloud resource evaluation module 16 aggregates the feature values of each time interval in the resource usage model of the server, as defined in Equation 5, assuming that the system has a total of x+1 servers, each server The resource usage behavior model has a total of m+1 time intervals, and the first time interval (t0), the second time interval (t1), and the m+1th time interval (tm) of the x+1 server are used. The feature values are summed separately to calculate the sum of the feature values d ( t _i ) ^total for each time interval (ti). After the summation of each time interval is completed, the sum of the largest eigenvalues is found from the m+1 time intervals according to formula 6 as the recommended value of the cloud resource requirement of the system sname. . The system cloud resource assessment module 16 will then generate a system cloud assessment report 9 .

In summary, the cloud computing evaluation system and the method thereof according to the present invention can effectively improve the error of the cloud resource estimation according to the resource usage data of the system and the resource usage data fed back to the cloud data center, and provide the future cloud data center. The system is more accurate in terms of cloud resource requirements. In addition, by analyzing the resource usage behavior of each server in different time intervals in the system, the cloud resource requirements of the system as a whole can be more accurately evaluated, which can effectively reduce the excessive allocation of cloud resources. issue.

The above is intended to be illustrative only and not limiting. Any equivalent modifications or alterations to the spirit and scope of the invention are intended to be included in the scope of the appended claims.

1‧‧‧Cloud Assessment System

11‧‧‧Data Receiving and Analysis Module

12‧‧‧Data Database

13‧‧‧ Behavioral Model Building Module

14‧‧‧Evaluation Learning Module

15‧‧‧Server Cloud Resource Demand Forecasting Module

16‧‧‧System Cloud Resource Evaluation Module

Claims (11)

A cloud computing evaluation system, comprising: a data receiving and analyzing module, configured to receive at least one hardware usage data of a system including at least one server, and the hardware of the server that has been stationed in the cloud data center Using data, and analyzing each of the hardware usage data to generate corresponding resource usage data respectively; a data database for storing each resource usage data; and a behavior model building module for using the data database Obtaining each resource usage data to respectively establish a corresponding resource usage behavior model; an evaluation learning module, configured to target the resource usage behavior model of the server before entering the cloud data center and the cloud data center The resource usage behavior model is analyzed, and one of the cloud resources of the server is recalculated to evaluate the error parameter value; a server cloud resource demand prediction module is configured to obtain each resource of the server from the data database. Use a behavioral model and evaluate the cloud resource requirements required for the server to enter the cloud data center; and a system cloud Resource assessment module for aggregate amount of cloud resource requirements for each of the servers, respectively, and the use of resources according to each of the behavioral model of each segment of the server at different times, the proposed resource requirements of the cloud systems. The cloud computing evaluation system of claim 1, wherein the evaluation learning module further comprises a resource usage behavior reading unit and a resource evaluation error calculation unit; the resource usage behavior reading unit is configured to read the The resource usage behavior model of the server before and after entering the cloud data center; the resource evaluation error calculation unit is configured to analyze the resource usage behavior model of the server to perform resource estimation error calculation, and compare and enter the cloud data center. The evaluation error parameter value is calculated after the resource usage data of the former and the resident. The cloud computing resource evaluation system of claim 1, wherein the server cloud resource requirement prediction module further comprises a server resource usage behavior comparison unit and a server cloud resource requirement calculation unit; the server resource Using the behavior comparison unit to search for the server of the same role according to the role of the server and performing an alignment of the resource usage behavior model; the server cloud resource requirement calculation unit 152 is configured to use the server according to the server itself The resource usage data is combined with the value of the evaluation error parameter of the server having the similar resource usage behavior module to calculate the cloud resource requirement of the server. The cloud computing evaluation system described in claim 1, wherein the cloud resource requirements include a computing resource (CPU) and a memory. For example, the cloud computing evaluation system described in claim 4, wherein the resource usage behavior model includes a CPU resource usage model and a memory resource usage model. A cloud computing evaluation method, comprising the steps of: receiving at least one hardware usage data of a system including at least one server, and extracting and normalizing all the hardware usage data of the server as a resource usage data Recording to a data database; respectively establishing a resource usage behavior model corresponding to each resource usage data of all the servers; performing a comparison of the resource usage behavior models according to the role and the resource type to determine the similarity a resource usage behavior model; calculating a cloud resource requirement of the server according to a maximum value of the resource usage data of the server and a similar one of the resource behavior models; and analyzing all the servers in the system The resource uses a behavior model to find all the eigenvalues of the server in the time interval according to the time interval and sum the eigenvalues, and find the maximum value from the eigenvalues of all time intervals as the system. The recommended value for the overall cloud resource requirements. For example, the cloud computing evaluation method described in claim 6 further includes the following steps: obtaining the resource usage data of the server stationed in the cloud data center in a specific period according to the set learning frequency to the cloud data center, and establishing The resource usage behavior model of the server after entering the cloud data center; according to the cloudization evaluation number of the server, obtaining the resource usage behavior model of the server before entering the cloud data center; and The server analyzes the resource usage behavior model before entering the cloud data center, and calculates the evaluation error parameter value of the resource usage behavior model of the server. The cloud-based evaluation method described in claim 7, wherein the resource usage behavior model includes a CPU resource usage model and a memory resource usage model. A cloud computing evaluation method as described in claim 7 wherein the role includes a web role And the DB role; the resource type includes the CPU resource type and the memory resource type. The cloud computing evaluation method described in claim 7 is characterized in that the cloud resource requirement includes at least a CPU configuration requirement value and a memory configuration requirement value. For example, in the cloud computing evaluation method described in claim 7, wherein the overall cloud resource requirement recommendation value includes the overall CPU resource requirement value of the system, the overall memory resource requirement value, and the value of the individual CPU resource requirements of the server in the system. Memory resource demand value.