WO2006046297A1 - 分析方法及び装置 - Google Patents
分析方法及び装置 Download PDFInfo
- Publication number
- WO2006046297A1 WO2006046297A1 PCT/JP2004/016051 JP2004016051W WO2006046297A1 WO 2006046297 A1 WO2006046297 A1 WO 2006046297A1 JP 2004016051 W JP2004016051 W JP 2004016051W WO 2006046297 A1 WO2006046297 A1 WO 2006046297A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- delay time
- server
- storage unit
- average
- stored
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0817—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
- G06F11/3419—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment by assessing time
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3447—Performance evaluation by modeling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0852—Delays
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
- G06F11/3419—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment by assessing time
- G06F11/3423—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment by assessing time where the assessed time is active or idle time
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/86—Event-based monitoring
Definitions
- the present invention relates to an analysis technique relating to a response of a computer system.
- Japanese Patent Application Laid-Open Publication No. 2004-21756 describes the response performance of each application under various usage conditions for one or more applications operating on an information system, with a limited number of experiments.
- the technique to evaluate automatically is disclosed. More specifically, when multiple load-in experiments corresponding to various usage situations of the application are performed multiple times, the quantity relating to the application usage situation, the quantity relating to the response performance of the application, and the quantity relating to the hardware 'resource usage situation Hardware 'resources' quantity of response performance is obtained, and an estimation formula group describing the dependency between the quantities is created. Therefore, it is possible to evaluate the response performance of the application using the estimation formula group.
- this technique requires an “experiment” and cannot be analyzed with normal processing.
- Patent Document 1 Japanese Patent Laid-Open No. 2004-21756
- the object of the present invention is to provide a computer for analyzing the response of the computer system using information that can easily acquire the computer power of the analysis target (hereinafter also referred to as monitoring target). Is to provide.
- An analysis method is an analysis method for analyzing a response of a computer 'system including a plurality of servers, wherein the computer' system power also uses CPU usage rate data of each of the plurality of servers.
- the average delay time in each server is estimated using the CPU usage rate stored in the CPU usage rate storage unit and the CPU usage rate of each server stored in the CPU usage rate storage unit and the request frequency stored in the request frequency storage unit.
- an estimation step stored in the server delay time storage unit is an estimation step stored in the server delay time storage unit.
- the estimation step described above uses the CPU usage rate of each server stored in the CPU usage rate storage unit and the request frequency stored in the request frequency storage unit. Estimated average CPU consumption per unit and stores consumption CPU time in the consumption CPU time storage, and stores average consumption CPU time and CPU usage per request for each server stored in the consumption CPU time storage. The average delay time of each server is estimated using the CPU usage rate of each server stored in the server, and the server delay It may also include a server delay time estimation step stored in the time storage unit.
- a regression analysis is performed using the CPU usage rate and request frequency of each server in the time zone specified in advance. You may make it estimate the average consumption CPU time per request.
- the time zone specified in advance it is possible to exclude the time zone in which the user's request is not processed so much, and the calculation accuracy can be improved.
- a coefficient value representing the relationship between the average consumed CPU time per request of the server and the average delay time in the server is an element for determining the coefficient value.
- Read the corresponding coefficient value by referring to the matrix storage unit that stores each CPU usage rate for each predetermined unit and the number of CPUs, and calculate the above server's average consumption CPU time per request from the coefficient value
- the average delay time may be calculated.
- the above coefficient value is a function of the CPU usage rate and the number of CPUs, so it can be calculated each time. However, since the calculation amount actually increases, as described above to increase the processing speed. It may be stored in the matrix storage unit.
- a step of estimating an average delay time for each category may be further included.
- the average delay time for each layer may be calculated with the layer as a category. For example, it is for extracting a problem for every work.
- the method may further include a step of estimating an average delay time of the entire computer system using the data stored in the server delay time storage unit and storing it in the system delay time storage unit.
- the above computer stored in the system delay time storage section is generated at a location other than the server due to the difference from the average delay time of the entire system.
- the method may further include a step of estimating the delay time. If the delay time that occurs outside the server is shorter than the average delay time of the entire computer system, the estimation is inappropriate for some reason, and it is possible to detect such a case.
- the correlation coefficient between the total of the average consumed CPU time and the request frequency is calculated, the reliability of the average delay time for each category is determined based on the correlation coefficient, and the reliability data is stored. And a correction step of correcting the average delay time for each category based on the reliability of the average delay time for each category stored in the reliability data storage unit and storing in the storage device.
- the average delay time is used as it is with high reliability, and a large correction is made for the average delay time with low reliability.
- the correction step described above is highly reliable for the average delay time for each category!
- the step of sorting in order, the reliability is high, the average delay time for each category is accumulated in order, and the accumulated average delay time is less than the measured delay value and has the maximum value.
- Steps to specify the order of degrees and the delay time of the next order after the specified order of reliability accumulate the average delay time for each category until the specified order of reliability And a step of correcting the difference from the value obtained by doing so.
- the CPU usage rate of each server is changed according to the changed request frequency and stored in the storage device. Estimating the average delay time in each server using the CPU usage rate of each server after the change and storing it in the storage device, and the server delay time storage unit and each server before and after the change stored in the storage device And outputting the average delay time in a comparable manner. It is possible to know how the delay time changes with changes in the request frequency.
- the CPU usage rate of each server is changed according to the number of CPUs after the change, and stored in the storage device.
- Each server using the CPU usage rate and the number of CPUs after the change.
- a step of calculating an average consumed CPU time per one request for each server in accordance with the number of servers after the change, and storing the CPU time in the storage device Using the average consumption CPU time per request of each server after change stored in the device, the CPU usage rate of each server after change is calculated and stored in the storage device, and stored in the storage device. Estimating the average delay time of each server after the change using the average CPU consumption per request of each changed server and the CPU usage rate of each server after the change, and storing it in the storage device May be further included. For example, if the number of servers is increased, it can be tested how much the delay time is reduced, and the effectiveness of the server can be determined by whether or not the investment is appropriate.
- the method may further include a step of estimating an average delay time for each category defined by the step and storing the estimated delay time in a storage device.
- a program for causing a computer to execute the analysis method described above can be created.
- This program is stored in a storage medium such as a flexible disk, a CD-ROM, a magneto-optical disk, a semiconductor memory, a hard disk, or the like. Stored in the device. Also, it may be distributed as a digital signal via a network. Note that intermediate processing results are temporarily stored in a storage device such as a memory.
- FIG. 1 is a diagram for explaining the principle of the present invention.
- FIG. 2 is a diagram for explaining the principle of the present invention.
- FIG. 3 is a diagram for explaining the entire system in the embodiment of the present invention.
- FIG. 4A is a functional block of the delay time analyzer in the example of the present invention.
- FIG. 4B is a functional block of the delay time analyzer in the example of the present invention.
- FIG. 5 is a diagram showing a main processing flow in the embodiment of the present invention.
- FIG. 6 is a diagram showing an example of acquired data.
- FIG. 7 is a diagram for explaining regression calculation in (a) and (b).
- FIG. 8 is a diagram for explaining the reason for limiting the target of regression calculation to business hours.
- FIG. 9 is a diagram showing a processing flow of reliability calculation processing.
- FIG. 10 is a diagram showing a processing flow of delay time correction processing according to reliability.
- FIGS. 11 (a) to 11 (c) are diagrams for explaining a specific example of a delay time correction process according to reliability.
- FIG. 12 is a diagram showing a processing flow of a delay time change estimation process when the request frequency varies.
- FIG. 13 is a diagram showing a processing flow of a delay time variation estimation process when the number of CPUs varies.
- FIG. 14 is a diagram showing a processing flow of a delay time change estimation process when the number of servers fluctuates.
- FIG. 15 is a diagram illustrating an example of a table of processing results.
- FIG. 16 is a diagram showing an example of a graph of processing results.
- FIG. 17 is a functional block diagram of a computer.
- Fig. 1 the average delay time for a single server S with multiple CPUs.
- the server S shown in Fig. 1 has C CPUs from CPU-1 to CPU-C, and requests entered from the outside with request frequency (req / sec) are placed in the queue Sw. Processed with C CPUs. In this case, the CPU usage rate is p (%) .
- the average stay time T (C, ⁇ ,) of requests in server s is as follows.
- the average stay time T (C, ⁇ , in the server S has the following relationship: ⁇ represents the proportion of requests that reach the server S. .
- the average delay time of requests in a specific single layer in multiple layers is obtained using a delay model in a single server.
- the prerequisite system model is shown in Figure 2. There are M servers S, S,... S in the first tier, and in the second tier
- ⁇ is the request to reach the ⁇ -th layer.
- the request of ⁇ / ⁇ is input, the request that also breaks down the first layer force is (1—H) ⁇ is all all, and each server of the second layer is input with the request of ⁇ ⁇ , From 2 layers
- the request to leave is ( ⁇ -a ⁇ N-l
- the request output from the layer is ⁇ ⁇ . Note that l ⁇ n ⁇ N and l ⁇ m ⁇ M.
- Each layer has different roles such as a Web server used as a front end for a user and an application server for dynamically processing a request. Assigned c
- the average delay time can be expressed as T (C, ⁇ , ⁇ ).
- the ⁇ layer
- the average delay time W of all requests in the nth layer is the average of the average delay times of all servers in the nth layer.
- H n is defined as follows.
- the delay time in the entire system Do model ⁇ is as follows.
- the average delay L of requests that leave the system after using the servers from the first layer to the ⁇ th layer is as follows.
- Average delay time per request For requests that leave the system after using the servers from the first layer to the i-th layer, express them as the product of those delays and the ratio to the total requests. It can be done as follows.
- H n represents the delay that occurs in each layer. It can be said that it represents time.
- Fig. 3 shows an overview of the system including the monitored system 100 and the delay analyzer 120.
- the monitoring target system 100 is connected to a network and has a configuration of ?? layers (in FIG. 3, two layers are used for simplicity) as shown in FIG.
- Each layer is provided with load balancers 101 and 102, and the load balancer includes server groups S and S of each layer.
- the server log 11 la is provided in the first layer server, and the log data is stored when processing the request.
- Each server is provided with CPU (Central Processing Unit) usage rate acquisition units 112a and 112b.
- the CPU usage rate is acquired in units of%.
- These CPU usage rate acquisition units 112a and 112b are general tools that are executed by commands such as sar, mpstat, and iostat in the case of UNIX (registered trademark) OS (Operating System). Many have the same function.
- the delay analyzer 120 is connected to the monitoring target system 100, and performs processing using the log data and CPU usage rate stored in the server log 11la. In this way, unlike the conventional case, since no special mechanism is incorporated in the monitored system 100, the delay analysis device 120 can be easily introduced, and all packets processed in the monitored system 100 are analyzed. There is no need to use large-capacity storage, so there is a security problem.
- the delay analyzer 120 is connected to a display device, a mouse, a keyboard, and other input / output units 121.
- FIG. 4A and FIG. 4B show functional block diagrams of the delay analyzer 120.
- the delay analyzer 120 includes a request frequency acquisition unit 1201, a CPU usage rate acquisition unit 1202, a log data storage unit 1203, a request frequency storage unit 1204, a measured delay value storage unit 1205, and a CPU usage rate storage unit 1206.
- the request frequency acquisition unit 1201 receives log data from the server log 11 la of the monitored system 100 and stores it in the log data storage unit 1203, and also stores it in the log data storage unit 1203 according to the input data from the input / output unit 121.
- the stored log data is processed to calculate the request frequency (req / sec) and stored in the request frequency storage unit 1204. Further, the log data stored in the log data storage unit 1203 is processed to calculate an average delay actual measurement value, and the average delay actual measurement value is stored in the delay actual measurement value storage unit 1205.
- the CPU usage rate acquisition unit 1202 acquires the CPU usage rate data from the CPU usage rate acquisition unit 112 of the monitoring target system 100 and stores the data in the CPU usage rate storage unit 1206.
- the CPU time calculation unit calculates the consumed CPU time per request with reference to the request frequency storage unit 1204, the CPU usage rate storage unit 1206, and the system configuration data storage unit 1207, and calculates the calculated data Stored in the CPU time storage unit 1209.
- the server delay time calculation unit 1 210 calculates the delay time for each server with reference to the CPU time storage unit 1209, the G table storage unit 1211, and the CPU usage rate storage unit 1206, and the calculated data is used as the server delay time. Store in the storage unit 1214.
- the server delay time calculation unit 1210 may refer to the request frequency storage unit 1204 and the system configuration data storage unit 1207 when the G table storage unit 1211 is not referred to.
- the layer delay time calculation unit 1215 refers to the server delay time storage unit 1214 and the system configuration data storage unit 1207 to calculate the delay time for each layer, and uses the calculated data as the layer delay time.
- the system delay time calculation unit 1217 refers to the layer delay time storage unit 1216 and the system configuration data storage unit 1207, calculates the delay time of the entire system, and stores the calculated data in the system delay time storage unit 1218.
- the residual delay time calculation unit 1219 refers to the actual delay value storage unit 1205 and the system delay time storage unit 1218 to calculate the residual delay time consumed by other devices other than the server, and calculates the calculated data. Stored in the remaining delay time storage unit 1220.
- the reliability calculation unit 1221 includes a residual delay time storage unit 1220 and a system configuration data size. Refer to the storage unit 1207, the actual delay storage unit 1205, the request frequency storage unit 1204, the CPU usage storage unit 1206, and the layer delay time storage unit 1216, and the remaining delay time consumed by devices other than the server If is less than 0, the reliability is calculated for the delay time of each layer, and the calculated reliability data is stored in the reliability storage unit 1222.
- the delay time correction unit 1223 corrects the delay time for each layer with reference to the layer delay time storage unit 1216 and the reliability storage unit 1222, and stores the corrected delay time data in the correction delay time storage unit 1224. .
- the performance prediction processing unit 1213 performs processing using the CPU usage rate storage unit 1206, the system configuration data storage unit 1207, the CPU time storage unit 1209, and the request frequency storage unit 1204.
- the input / output unit 121 can output data in the storage unit in the delay analyzer 120 to a display device or the like.
- the request frequency acquisition unit 1201 acquires log data from the server log 11 la of the monitored system 100 and stores it in the log data storage unit 1203.
- the CPU usage rate acquisition unit 1202 acquires the CPU usage rate of the monitored system 100.
- the CPU usage rate data is received from the unit 112 and stored in the CPU usage rate storage unit 1206 (FIG. 5: step S 1).
- log data stored in the log data storage unit 1203 is shown below.
- the server log is stored as 11 la under the / var / log / httpd / directory!
- This first item “192.168.164.108” represents the IP address of the accessing client. Paragraphs 2 and 3 are omitted. Section 4 [[14 / Sep / 2004: 12: 27: 50 +0900]] represents the access time. Item 5 GET / ⁇ ! “ioge / SSSS / SSSS—20040816.pdf HTTP / 1. ⁇ ” indicates access contents. The sixth term “200” represents the status (normal here). Section 7 ⁇ 147067 '' is sent Represents the number of received bytes. Section 8 “'“ ”represents the requested URL nose.
- Section 9 "Mozilla / 4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.1.4322)"” refers to the browser used by the accessing client.
- the input / output unit 121 receives an input of setting of the analysis target period and the business time zone, and stores it in a storage device such as a main memory (step S3).
- the business time zone refers to a time zone in which the server spends less CPU time for processing other than requests from users. By specifying the business time zone, it is possible to reduce the estimation error caused by a large amount of CPU time being consumed by the server when there are few requests such as at night.
- the request frequency acquisition unit 1201 reads the log data in the specified analysis target period and business time zone from the log data storage unit 1203, and counts how many requests are processed every hour, for example.
- the request frequency acquisition unit 1201 calculates an average delay actual measurement value by adding the time taken to handle all requests every hour and dividing by the number of requests, for example, and the delay actual measurement value storage unit Store in 1205.
- the CPU usage rate storage unit 1206 calculates the average CPU usage rate of each server S every hour based on the CPU usage rate data stored in the CPU usage rate storage unit 1206, and the CPU usage rate storage unit 1206 Stored in
- Step S5 When one server has multiple CPUs, the average CPU usage rate of the multiple CPUs is calculated as the CPU usage rate of the server.
- the average CPU usage rate p i, i represents the i-th unit time (here, every hour).
- the letter “average” may be omitted.
- the CPU time calculation unit 1208 refers to the request frequency storage unit 1204, the CPU usage rate storage unit 1206, and the system configuration data storage unit 1207 to calculate the CPU time consumed per request, Store in the time storage unit 1209 (step S7). Occurs on each server To calculate the delay time, first, the external force incoming Rikuesutoe i (re q / sec) for the entire system, is consumed CPU time how much per request at each server It is necessary to ask. But simply CP of server S at time i
- a server usually consumes some CPU time due to system maintenance in addition to request processing.
- the request frequency is extremely small, the ratio of CPU time is relatively large, so the CPU time consumed per request is overestimated, which may cause errors.
- the horizontal axis is the request frequency and the vertical axis is the CPU usage rate, if the equation (12) is interpreted as it is, the CPU usage rate should be 0 if there is no request. Therefore, if the slope of the straight line connecting the origin and each measurement point is the consumed CPU time per request, a large variation occurs.
- the server delay time calculation unit 1210 next refers to the CPU usage rate storage unit 1 206, the system configuration data storage unit 1207, the CPU time storage unit 1209, and the G table storage unit 1211.
- the average delay time per request generated in each server is calculated, and the calculated value is stored in the server delay time storage unit 1214 (step S9).
- the average delay time T 1 per request generated in each server is
- G (C, p) is calculated based on the number of CPUs and the CPU usage rate of the server as shown in equation (3). However, if the equation (3) is calculated as it is, it takes a relatively long time. If the granularity of the analysis is fixed, it can be calculated in advance by changing the number of CPUs and the CPU usage rate of the server. is there. For example, if the analysis granularity is sufficient in units of 1% in terms of CPU usage, and the expected number of CPUs per server is 50 or less, the CPU usage is 0 to 99% (1 For each server with 1 to 50 CPUs, G (C,) is calculated in advance and stored in the G table storage unit 1211 as a 100 X 50 matrix. Then, the number of CPUs can be acquired from the system configuration data storage unit 1207, the CPU usage rate can be acquired from the CPU usage rate storage unit 1206, and the value of G (C, can be acquired from the G table storage unit 1211.
- an average delay time T 1 per request generated in each server according to equation (14) (hereinafter also referred to as the average delay time of each server for short) is calculated, and the server delay time is calculated.
- the layer delay time calculation unit 1215 refers to the server delay time storage unit 1214 and the system configuration data storage unit 1207, calculates the delay time L 1 in each layer, and stores the layer delay time. Store in the storage unit 1216 (step SI 1).
- the delay time in each layer is the sum of the average delay times of the servers in each layer and is expressed as follows. M is acquired from the system configuration data storage unit 1207.
- the system delay time calculating unit 1217 refers to the layer delay time storage unit 1216 and system configuration data storage unit 1207, calculates the delay time D 1 of the entire system, the system delay time storage unit 1218 (Step S13).
- Delay time D 1 of the whole system is the sum of the delay time L 1 in each layer n, is expressed as follows. N is acquired from the system configuration data storage unit 1207.
- the residual delay time calculation unit 1219 refers to the measured delay value storage unit 1205 and the system delay time storage unit 1218, calculates the delay time ⁇ 1 that is used in a place other than the server, and the residual delay time Store in the storage unit 1220 (step S15).
- Delay time E 1 is a difference in delay time D 1 and the delay measured value A 1 of the entire system, is calculated as follows.
- the reliability calculation unit 1221 includes a residual delay time storage unit 1220, a layer delay time storage unit 1216, and a system configuration data storage unit 1207.
- the request frequency storage unit 1204, the CPU usage rate storage unit 1206, and the measured delay value storage unit 1205 are used to calculate the reliability of the average delay time of each layer.
- the result is stored in the reliability storage unit 1222 (step S17). This process will be described with reference to FIG. First, the reliability calculation unit 1221 calculates the correlation coefficient between the sum p and the request frequency of consumption CPU time of the n layer Well as the initial confidence R 1 in the average delay time of each layer n, reliability storage unit Store in 1222 (step S31). When correl is a function for obtaining a correlation coefficient, the reliability R 1 is calculated according to the following formula.
- the first term of the correl function in Eq. (15) is the total CPU time consumed in the nth layer. Note that the correlation coefficient is also used in later calculations, and is retained for each layer.
- the delay time correction unit 1223 refers to the reliability storage unit 1222 and the layer delay time storage unit 1216 and corrects the delay time according to the reliability, thereby correcting the delay. Stored in the time storage unit 1224 (step S19). If ⁇ , this step is skipped. This process will be described with reference to FIG. First, the delay time correction unit 1223 refers to the layer delay time storage unit 1216 and the reliability storage unit 1222, sorts the delay times of each layer in descending order of reliability, and stores them in the correction delay time storage unit 1224. (Step S4 Do) Note that if there are multiple layers with a confidence level of 0, sort them in descending order of their correlation coefficient.
- step S 43 the delay times of the layers are added in the descending order of reliability according to the sorting result, and the order of reliability B is specified such that the added value is the maximum less than the average delay actual measurement value (step S). 43).
- ⁇ ⁇ ⁇
- the high reliability of the ⁇ -th layer is the ⁇ th top force.
- Ri> always holds.
- step S43 the maximum that satisfies
- the reliability B + delay time of the first layer is corrected as follows (step S45). That is, the estimated delay time L 1 of the P layer is corrected and the result
- This equation shows the reliability B + 1st layer so that the actual measured delay value is equal to the sum of delay time (estimated average value) up to the first direction B + high reliability among the reliability of each layer. Correct the delay time.
- reliability B + The reliability of the first layer is corrected as follows (step S47). That is, P
- the reliability ⁇ + the delay time and reliability of the second and subsequent layers are corrected as follows (step S49).
- Fig. 11 (a) shows the delay time estimation results and actual measurement results.
- the first layer of the monitoring target system 100 in this example is a Web server
- the second layer is an application server
- the third layer is a DB server.
- the estimated delay time of the first layer is 150 msec
- the correlation coefficient is 0.9
- the reliability is 0.
- the estimated delay time of the second layer is 60 msec
- the correlation coefficient is 0.85
- the reliability is 0.85.
- the estimated delay time of the third layer is 30 msec
- the correlation coefficient is 0.6
- the reliability is 0.6. Note that the measured average delay is 100 msec.
- step S41 when sorting in step S41, as shown in Fig. 11 (b), the layers are arranged in the order of the second layer, the third layer, and the first layer, and the delay time of the entire system is apparent. The average delay measurement value is exceeded, and the estimated delay time of the entire system is exceeded in the middle of the first layer.
- the delay time and reliability are used as they are, and the estimated delay time of the first layer is the average delay measured value and It is reduced to the difference from the sum of the delay times of the second and third layers, resulting in 10 msec.
- the input / output unit 121 performs output processing (step S 21).
- the data output by the input / output unit 121 includes (1) the estimated delay time T 1 , (
- the value itself may be output, or the reliability may be classified into the following three levels, for example, and the classification result may be output. That is, if> 0.7, the reliability is “high”, if 0.7 ⁇ Ri> 0.3, the reliability is “medium”, and if 0.3 ⁇ Ri, the reliability is “low”.
- the classification of high, medium, and low reliability as described above is a value generally used for determining the strength of correlation in the correlation coefficient. That is, generally, if the absolute value of the correlation coefficient is 0.7 or more, it is determined that there is a strong correlation between the two parameters, and 0.3 to 0.7. If it is, it is judged that there is a weak correlation, and if it is 0.3 or less, it is considered that there is almost no correlation. This is due to the fact that the square of the correlation coefficient is the explanation rate of fluctuation. When the correlation coefficient is 0.7, the explanation rate is 0.49 (about 50%). In other words, about half of the changes in the dependent variable can be explained by the explanatory variables. When the number of correlations is 0.3, the explanation rate is 0.1 (about 10%), and only about 10% of changes in the dependent variable are attributed to explanation variables. There is almost no correlation between the variable and the dependent variable.
- the reliability is considered to be high.
- the prediction error is within ⁇ 50% if the reliability is high, and the prediction error is ⁇ if the reliability is medium. If it is within 100% and the reliability is “low”, there is a high possibility of ⁇ 100% or more.
- this result is only a guide based on the experimental results, and does not guarantee the accuracy (error range) described above.
- the delay time of each server, each layer, and the entire system can be calculated using elements already existing in the monitored system 100.
- the delay time can be corrected from the relationship with the actual measured delay value, and the reliability can be presented to the user.
- the estimation of the change in the delay time when the request frequency varies will be described with reference to FIG.
- the request frequency is given at a certain time point i
- the estimated average delay time when the request frequency changes to ⁇ force ⁇ ′ is calculated. That is, ⁇ ′ is input from the input / output unit 121 and is received by the performance prediction processing unit 1213 of the delay analyzer 120 (step S50). Then, the performance prediction processing unit 1213 changes the CPU usage rate p for all the servers S according to the change in the request frequency, and stores the CPU usage rate.
- step S51 CPU usage rate p is changed to p 'as follows.
- the server delay time calculation unit 1210 calculates the delay time of each server after the change using the CPU usage rate p ′ after the change, and stores it in the server delay time storage unit 1214. (Step S53). The calculations in steps S51 and S53 are as follows.
- the layer delay time calculation unit 1215 calculates the delay time in each layer after the change using the delay time T ”of each server after the change stored in the server delay time storage unit 1214.
- step S55 Calculate and store in the layer delay time storage unit 1216 (step S55). Further, the system delay time calculation unit 1217 calculates the delay time of the entire system after the change using the delay time in each layer after the change stored in the layer delay time storage unit 1216, and stores it in the system delay time storage unit 1218. Store (step S57).
- the input / output unit 121 outputs the delay times before and after the change (steps).
- the CPU number C ′ is input from the unit 121, and the performance prediction processing unit 1213 of the delay analyzer 120
- the performance prediction processing unit 1213 changes the CPU usage rate p according to the change in the number of CPUs and stores it in the CPU usage rate storage unit 1206 (step S63).
- the CPU usage rate p is changed to P 'as follows only for the server whose CPU number has changed.
- the server delay time calculation unit 1210 calculates the delay time T ”of each server whose number of CPUs has been changed, using the CPU usage rate p ′ after the change.
- step S65 Stored in the server delay time storage unit 1214 (step S65).
- the calculations in steps S63 and S65 are as follows.
- the layer delay time calculation unit 1215 calculates the delay time in the layer related to the change using the server delay time after the change stored in the server delay time storage unit 1214.
- the system delay time calculation unit 1217 calculates the delay time of the entire system after the change using the delay time in each layer stored in the layer delay time storage unit 1216, and stores it in the system delay time storage unit 1218. (Step S68).
- the input / output unit 121 outputs each delay time before and after the change (step S69). This allows the user to consider changes in delay time according to changes in the number of CPUs. For example, consider the effect of increasing the number of CPUs using this result.
- the estimated delay time when the number of servers in the n-th layer is changed to M force M ' is obtained.
- the number M ′ of servers in the n-th layer is input from the input / output unit 121 and received by the performance prediction processing unit 1213 of the delay analysis apparatus 120 (step S71).
- the performance prediction processing unit 1213 corrects the CPU time consumed per request according to the change in the number of servers, and stores it in the CPU time storage unit 1209 (step S73).
- ⁇ is the intercept obtained when calculating 1 ⁇ , and CPU time storage
- the server delay time calculation unit 1210 includes the CPU usage rate p ′ after the change stored in the CPU usage rate storage unit 1206 and the CPU consumption per request after the change stored in the CPU time storage unit 1209.
- the server delay time after the change is calculated using the time 1Z ′ (n, m), and stored in the server delay time storage unit 1214 (step S77).
- the changed server delay time T ", n, m) is expressed as follows.
- the layer delay time calculation unit 1215 calculates the delay time in each layer using the changed server delay time stored in the server delay time storage unit 1214, and the layer delay time is calculated.
- step S79 Stored in the extended time storage unit 1216 (step S79). In this step as well, the following calculation is performed using M 'from the performance prediction processing unit 1213.
- L ′ 1 is also expressed as follows from equation (16).
- system delay time calculation unit 1217 calculates the delay time of the entire system after the change using the delay time in each layer stored in the layer delay time storage unit 1216, and the system delay time storage unit 1218 (Step S81).
- the input / output unit 121 outputs each delay time before and after the change (step S83).
- the user can consider changes in delay time according to changes in the number of servers. For example, consider the effect of increasing the number of servers using this result.
- FIGS. 4A and 4B are examples, and may not necessarily correspond to the actual program configuration.
- a table as shown in Fig. 15 can be used instead of just displaying the numerical values as they are (consumed CPU time per request, CPU usage rate, average delay time for each server, average delay time for each layer, actual delay measurement) Value, estimated delay time other than server, and reliability of delay time of each layer for each unit time i) and graphs as shown in Fig. 16 (horizontal axis represents time, vertical axis represents delay time) , Web server (first tier), application server (second tier), DB server (third tier), etc.
- the delay analysis device 120 described above is a computer device and is shown in FIG. Is it the same as memory 2501 (memory unit)? 112503 (Processing unit) and hard disk 'drive 03 ⁇ 400) 2 505 and display device 2509 connected to display device 2509 and drive device 2513 for removable disk 2511 and input device 2515 and communication control unit 2517 for connecting to network Connected with Cass 2519!
- the operating system (OS) and the application program for executing the processing in this embodiment are stored in the HDD 2505, and when executed by the CPU 2503, it is read out from the HDD 2505 power to the memory 2501. .
- the CPU 2503 controls the display control unit 2507, the communication control unit 2517, and the drive device 2513 as necessary to perform necessary operations.
- the data being processed is stored in the memory 2501, and stored in the HDD 2505 if necessary.
- an application program for executing the above-described processing is stored and distributed on a removable disk 2511 and installed in the HDD 2505 from the drive device 2513. It may be installed in HDD2505 via network such as the Internet and communication control unit 2517.
- Such a computer device performs various functions as described above by organically cooperating the hardware such as CPU2503 and memory 2501 described above with the OS and the necessary application program. Realize.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Environmental & Geological Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2006542174A JP4180638B2 (ja) | 2004-10-28 | 2004-10-28 | 分析方法及び装置 |
EP04793159.7A EP1806658B1 (en) | 2004-10-28 | 2004-10-28 | Analyzing method and device |
PCT/JP2004/016051 WO2006046297A1 (ja) | 2004-10-28 | 2004-10-28 | 分析方法及び装置 |
US11/739,946 US8560667B2 (en) | 2004-10-28 | 2007-04-25 | Analysis method and apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2004/016051 WO2006046297A1 (ja) | 2004-10-28 | 2004-10-28 | 分析方法及び装置 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/739,946 Continuation US20100231444A1 (en) | 2007-10-26 | 2007-04-26 | Positioning receiver and positioning method |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006046297A1 true WO2006046297A1 (ja) | 2006-05-04 |
Family
ID=36227549
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2004/016051 WO2006046297A1 (ja) | 2004-10-28 | 2004-10-28 | 分析方法及び装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US8560667B2 (ja) |
EP (1) | EP1806658B1 (ja) |
JP (1) | JP4180638B2 (ja) |
WO (1) | WO2006046297A1 (ja) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007310836A (ja) * | 2006-05-22 | 2007-11-29 | Daiwa Securities Group Inc | 評価装置及び設計装置、評価方法及び設計方法、評価用プログラム及び設計用プログラム並びに情報記録媒体 |
JP2011154483A (ja) * | 2010-01-26 | 2011-08-11 | Fujitsu Ltd | 異常検出装置、プログラム、及び異常検出方法 |
JP4900881B2 (ja) * | 2006-07-10 | 2012-03-21 | 日本電気株式会社 | コンピュータシステム、管理装置、及びコンピュータプログラム |
WO2012086443A1 (ja) * | 2010-12-24 | 2012-06-28 | 日本電気株式会社 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
WO2012086444A1 (ja) * | 2010-12-24 | 2012-06-28 | 日本電気株式会社 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
US8924551B2 (en) | 2012-03-15 | 2014-12-30 | Fujitsu, Limited | Analysis method and information processing apparatus |
US9921861B2 (en) | 2013-06-26 | 2018-03-20 | Fujitsu Limited | Virtual machine management method and information processing apparatus |
JP2018156553A (ja) * | 2017-03-21 | 2018-10-04 | 株式会社エヌ・ティ・ティ・データ | 識別情報付与システム及び識別情報付与方法 |
US10169059B2 (en) | 2013-06-26 | 2019-01-01 | Fujitsu Limited | Analysis support method, analysis supporting device, and recording medium |
Families Citing this family (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4711672B2 (ja) * | 2004-12-24 | 2011-06-29 | 富士通株式会社 | 情報処理方法及びプログラム |
US20070136302A1 (en) * | 2005-12-12 | 2007-06-14 | Microsoft Corporation | Automated device blog creation |
US8914774B1 (en) | 2007-11-15 | 2014-12-16 | Appcelerator, Inc. | System and method for tagging code to determine where the code runs |
US8954989B1 (en) | 2007-11-19 | 2015-02-10 | Appcelerator, Inc. | Flexible, event-driven JavaScript server architecture |
US8260845B1 (en) | 2007-11-21 | 2012-09-04 | Appcelerator, Inc. | System and method for auto-generating JavaScript proxies and meta-proxies |
US8566807B1 (en) | 2007-11-23 | 2013-10-22 | Appcelerator, Inc. | System and method for accessibility of document object model and JavaScript by other platforms |
US8719451B1 (en) | 2007-11-23 | 2014-05-06 | Appcelerator, Inc. | System and method for on-the-fly, post-processing document object model manipulation |
US8806431B1 (en) | 2007-12-03 | 2014-08-12 | Appecelerator, Inc. | Aspect oriented programming |
US8819539B1 (en) | 2007-12-03 | 2014-08-26 | Appcelerator, Inc. | On-the-fly rewriting of uniform resource locators in a web-page |
US8756579B1 (en) | 2007-12-03 | 2014-06-17 | Appcelerator, Inc. | Client-side and server-side unified validation |
US8938491B1 (en) | 2007-12-04 | 2015-01-20 | Appcelerator, Inc. | System and method for secure binding of client calls and server functions |
US8527860B1 (en) | 2007-12-04 | 2013-09-03 | Appcelerator, Inc. | System and method for exposing the dynamic web server-side |
US8639743B1 (en) | 2007-12-05 | 2014-01-28 | Appcelerator, Inc. | System and method for on-the-fly rewriting of JavaScript |
US8335982B1 (en) | 2007-12-05 | 2012-12-18 | Appcelerator, Inc. | System and method for binding a document object model through JavaScript callbacks |
US8285813B1 (en) | 2007-12-05 | 2012-10-09 | Appcelerator, Inc. | System and method for emulating different user agents on a server |
US20090217282A1 (en) * | 2008-02-26 | 2009-08-27 | Vikram Rai | Predicting cpu availability for short to medium time frames on time shared systems |
US8224624B2 (en) * | 2008-04-25 | 2012-07-17 | Hewlett-Packard Development Company, L.P. | Using application performance signatures for characterizing application updates |
US8291079B1 (en) | 2008-06-04 | 2012-10-16 | Appcelerator, Inc. | System and method for developing, deploying, managing and monitoring a web application in a single environment |
US8880678B1 (en) * | 2008-06-05 | 2014-11-04 | Appcelerator, Inc. | System and method for managing and monitoring a web application using multiple cloud providers |
US20090307347A1 (en) * | 2008-06-08 | 2009-12-10 | Ludmila Cherkasova | Using Transaction Latency Profiles For Characterizing Application Updates |
US7596620B1 (en) | 2008-11-04 | 2009-09-29 | Aptana, Inc. | System and method for developing, deploying, managing and monitoring a web application in a single environment |
US8019858B2 (en) * | 2008-09-09 | 2011-09-13 | International Business Machines Corporation | System and method for utilizing system lag to send facts to an end user |
US7925748B2 (en) * | 2008-10-24 | 2011-04-12 | International Business Machines Corporation | System and method for managing system resources in a network environment |
US7930395B2 (en) * | 2008-12-01 | 2011-04-19 | International Business Machines Corporation | System and method for managing system resources in a network environment |
US20110153828A1 (en) * | 2009-12-17 | 2011-06-23 | Korea Advanced Institute Of Science And Technology | Load balancing apparatus and method for regulating load using the same |
PL3367252T3 (pl) | 2010-07-26 | 2020-05-18 | Seven Networks, Llc | Zarządzanie ruchem z uwzględnieniem kontekstu w celu zaoszczędzenia zasobów w sieci bezprzewodowej |
EP3651028A1 (en) * | 2010-07-26 | 2020-05-13 | Seven Networks, LLC | Mobile network traffic coordination across multiple applications |
CN103404193B (zh) | 2010-11-22 | 2018-06-05 | 七网络有限责任公司 | 调校数据传输以优化为通过无线网络的传输建立的连接 |
CN103116524A (zh) * | 2011-11-16 | 2013-05-22 | 鸿富锦精密工业(深圳)有限公司 | Cpu使用率调整系统及方法 |
US9537740B2 (en) | 2014-07-31 | 2017-01-03 | International Business Machines Corporation | Monitoring device usage |
US9998347B2 (en) | 2014-07-31 | 2018-06-12 | International Business Machines Corporation | Monitoring device usage |
US10169182B2 (en) | 2014-07-31 | 2019-01-01 | International Business Machines Corporation | Monitoring levels of utilization of device |
US10102103B2 (en) | 2015-11-11 | 2018-10-16 | International Business Machines Corporation | System resource component utilization |
US10649876B2 (en) | 2017-04-20 | 2020-05-12 | International Business Machines Corporation | Maintaining manageable utilization in a system to prevent excessive queuing of system requests |
US10394643B2 (en) * | 2017-08-16 | 2019-08-27 | National Instruments Corporation | Distributed run-time auto-calculation of measurement uncertainty |
US11023280B2 (en) | 2017-09-15 | 2021-06-01 | Splunk Inc. | Processing data streams received from instrumented software using incremental finite window double exponential smoothing |
Citations (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6237763A (ja) * | 1985-08-12 | 1987-02-18 | Fujitsu Ltd | キヤパシテイプランニングにおける捕捉率自動算出方式 |
JPH05324358A (ja) * | 1992-05-20 | 1993-12-07 | Hitachi Ltd | 性能予測装置 |
JPH05334102A (ja) * | 1992-05-28 | 1993-12-17 | Nec Corp | ジョブ実行状況予測装置 |
JPH0695931A (ja) * | 1992-09-14 | 1994-04-08 | Toshiba Corp | システム実行性能評価支援装置 |
WO1994009429A1 (en) | 1992-10-13 | 1994-04-28 | Zitel Corporation | Method and system for computer performance evaluation |
JPH11282695A (ja) * | 1998-03-11 | 1999-10-15 | Internatl Business Mach Corp <Ibm> | 多重システム・クラスタ内のサ―バの数を制御する方法及び装置 |
JP2000092052A (ja) * | 1998-09-08 | 2000-03-31 | Fujitsu Ltd | ネットワークサーバ負荷検出装置、割当装置および方法 |
JP2000172537A (ja) * | 1998-12-04 | 2000-06-23 | Nippon Steel Corp | 性能予測装置および方法、記録媒体 |
JP2000298593A (ja) * | 1999-04-14 | 2000-10-24 | Nec Corp | マルチタスクシステムの性能予測システム及び予測方法並びにその方法プログラムを記録した記録媒体 |
JP2002099448A (ja) * | 2000-09-21 | 2002-04-05 | Ntt Data Corp | 性能監視装置、及びその方法 |
US20020082807A1 (en) | 2000-12-23 | 2002-06-27 | Turicchi Thomas Edwin | Method for service level estimation in an operating computer system |
JP2002268922A (ja) | 2001-03-09 | 2002-09-20 | Ntt Data Corp | Wwwサイトの性能監視装置 |
JP2002342182A (ja) * | 2001-05-21 | 2002-11-29 | Hitachi Ltd | ネットワークシステムにおける運用管理の支援システム |
JP2003178040A (ja) * | 2001-12-10 | 2003-06-27 | Hitachi Information Systems Ltd | ウェブサイトの構成決定支援方法 |
JP2003263342A (ja) * | 2002-03-07 | 2003-09-19 | Telecommunication Advancement Organization Of Japan | 情報処理装置の監視装置および監視方法並びにそのプログラム |
US20030236878A1 (en) | 2002-06-19 | 2003-12-25 | Masashi Egi | Statistical method for estimating the performances of computer systems |
JP2004005135A (ja) * | 2002-05-31 | 2004-01-08 | Nippon Telegr & Teleph Corp <Ntt> | サーバ性能推定方法およびサーバ性能推定装置 |
JP2004046734A (ja) * | 2002-07-15 | 2004-02-12 | Fuji Electric Holdings Co Ltd | Webシステム性能予測装置、Webシステム性能予測方法、及びプログラム |
JP2004193816A (ja) * | 2002-12-10 | 2004-07-08 | Hitachi Ltd | ネットワーク評価システム |
JP2004272582A (ja) * | 2003-03-07 | 2004-09-30 | Toshiba Corp | 計算機システムの性能予測プログラムおよび設計支援システム |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5255350A (en) | 1975-10-30 | 1977-05-06 | Nec Corp | System recomposition control unit |
JPS5851362A (ja) | 1981-09-24 | 1983-03-26 | Hitachi Ltd | 計算機システムの性能予測方式 |
JPH09305417A (ja) | 1996-05-13 | 1997-11-28 | Nec Corp | Cpu使用率最適化方式 |
US6243105B1 (en) * | 1998-11-19 | 2001-06-05 | Ncr Corporation | Drill-down method to historical data in a performance monitor using a platform independent program |
US6557035B1 (en) * | 1999-03-30 | 2003-04-29 | International Business Machines Corporation | Rules-based method of and system for optimizing server hardware capacity and performance |
JP2001109638A (ja) | 1999-10-06 | 2001-04-20 | Nec Corp | 推定伸長率に基づくトランザクション負荷分散方法及び方式並びにコンピュータ可読記録媒体 |
JP2002132543A (ja) | 2000-10-25 | 2002-05-10 | Hitachi Ltd | 計算機システムの管理方法 |
JP2003032306A (ja) | 2001-07-19 | 2003-01-31 | Hitachi Electronics Service Co Ltd | Webサーバの回線毎の応答性能予測システム |
US6792393B1 (en) * | 2001-12-03 | 2004-09-14 | At&T Corp. | System and method for diagnosing computer system operational behavior |
US7222245B2 (en) * | 2002-04-26 | 2007-05-22 | Hewlett-Packard Development Company, L.P. | Managing system power based on utilization statistics |
-
2004
- 2004-10-28 JP JP2006542174A patent/JP4180638B2/ja not_active Expired - Fee Related
- 2004-10-28 EP EP04793159.7A patent/EP1806658B1/en not_active Ceased
- 2004-10-28 WO PCT/JP2004/016051 patent/WO2006046297A1/ja active Application Filing
-
2007
- 2007-04-25 US US11/739,946 patent/US8560667B2/en not_active Expired - Fee Related
Patent Citations (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6237763A (ja) * | 1985-08-12 | 1987-02-18 | Fujitsu Ltd | キヤパシテイプランニングにおける捕捉率自動算出方式 |
JPH05324358A (ja) * | 1992-05-20 | 1993-12-07 | Hitachi Ltd | 性能予測装置 |
JPH05334102A (ja) * | 1992-05-28 | 1993-12-17 | Nec Corp | ジョブ実行状況予測装置 |
JPH0695931A (ja) * | 1992-09-14 | 1994-04-08 | Toshiba Corp | システム実行性能評価支援装置 |
WO1994009429A1 (en) | 1992-10-13 | 1994-04-28 | Zitel Corporation | Method and system for computer performance evaluation |
JPH11282695A (ja) * | 1998-03-11 | 1999-10-15 | Internatl Business Mach Corp <Ibm> | 多重システム・クラスタ内のサ―バの数を制御する方法及び装置 |
JP2000092052A (ja) * | 1998-09-08 | 2000-03-31 | Fujitsu Ltd | ネットワークサーバ負荷検出装置、割当装置および方法 |
JP2000172537A (ja) * | 1998-12-04 | 2000-06-23 | Nippon Steel Corp | 性能予測装置および方法、記録媒体 |
JP2000298593A (ja) * | 1999-04-14 | 2000-10-24 | Nec Corp | マルチタスクシステムの性能予測システム及び予測方法並びにその方法プログラムを記録した記録媒体 |
JP2002099448A (ja) * | 2000-09-21 | 2002-04-05 | Ntt Data Corp | 性能監視装置、及びその方法 |
US20020082807A1 (en) | 2000-12-23 | 2002-06-27 | Turicchi Thomas Edwin | Method for service level estimation in an operating computer system |
JP2002268922A (ja) | 2001-03-09 | 2002-09-20 | Ntt Data Corp | Wwwサイトの性能監視装置 |
JP2002342182A (ja) * | 2001-05-21 | 2002-11-29 | Hitachi Ltd | ネットワークシステムにおける運用管理の支援システム |
JP2003178040A (ja) * | 2001-12-10 | 2003-06-27 | Hitachi Information Systems Ltd | ウェブサイトの構成決定支援方法 |
JP2003263342A (ja) * | 2002-03-07 | 2003-09-19 | Telecommunication Advancement Organization Of Japan | 情報処理装置の監視装置および監視方法並びにそのプログラム |
JP2004005135A (ja) * | 2002-05-31 | 2004-01-08 | Nippon Telegr & Teleph Corp <Ntt> | サーバ性能推定方法およびサーバ性能推定装置 |
US20030236878A1 (en) | 2002-06-19 | 2003-12-25 | Masashi Egi | Statistical method for estimating the performances of computer systems |
JP2004021756A (ja) * | 2002-06-19 | 2004-01-22 | Hitachi Ltd | 情報システム性能の統計的予測方法 |
JP2004046734A (ja) * | 2002-07-15 | 2004-02-12 | Fuji Electric Holdings Co Ltd | Webシステム性能予測装置、Webシステム性能予測方法、及びプログラム |
JP2004193816A (ja) * | 2002-12-10 | 2004-07-08 | Hitachi Ltd | ネットワーク評価システム |
JP2004272582A (ja) * | 2003-03-07 | 2004-09-30 | Toshiba Corp | 計算機システムの性能予測プログラムおよび設計支援システム |
Non-Patent Citations (1)
Title |
---|
See also references of EP1806658A4 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007310836A (ja) * | 2006-05-22 | 2007-11-29 | Daiwa Securities Group Inc | 評価装置及び設計装置、評価方法及び設計方法、評価用プログラム及び設計用プログラム並びに情報記録媒体 |
JP4900881B2 (ja) * | 2006-07-10 | 2012-03-21 | 日本電気株式会社 | コンピュータシステム、管理装置、及びコンピュータプログラム |
US8171133B2 (en) | 2006-07-10 | 2012-05-01 | Nec Corporation | Management apparatus and management method for computer system |
JP2011154483A (ja) * | 2010-01-26 | 2011-08-11 | Fujitsu Ltd | 異常検出装置、プログラム、及び異常検出方法 |
JPWO2012086443A1 (ja) * | 2010-12-24 | 2014-05-22 | 日本電気株式会社 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
WO2012086444A1 (ja) * | 2010-12-24 | 2012-06-28 | 日本電気株式会社 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
WO2012086443A1 (ja) * | 2010-12-24 | 2012-06-28 | 日本電気株式会社 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
US9111227B2 (en) | 2010-12-24 | 2015-08-18 | Nec Corporation | Monitoring data analyzing apparatus, monitoring data analyzing method, and monitoring data analyzing program |
JP5871193B2 (ja) * | 2010-12-24 | 2016-03-01 | 日本電気株式会社 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
JP5871192B2 (ja) * | 2010-12-24 | 2016-03-01 | 日本電気株式会社 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
US9465713B2 (en) | 2010-12-24 | 2016-10-11 | Nec Corporation | Monitoring data analyzing apparatus, monitoring data analyzing method, and monitoring data analyzing program |
US8924551B2 (en) | 2012-03-15 | 2014-12-30 | Fujitsu, Limited | Analysis method and information processing apparatus |
US9921861B2 (en) | 2013-06-26 | 2018-03-20 | Fujitsu Limited | Virtual machine management method and information processing apparatus |
US10169059B2 (en) | 2013-06-26 | 2019-01-01 | Fujitsu Limited | Analysis support method, analysis supporting device, and recording medium |
JP2018156553A (ja) * | 2017-03-21 | 2018-10-04 | 株式会社エヌ・ティ・ティ・データ | 識別情報付与システム及び識別情報付与方法 |
Also Published As
Publication number | Publication date |
---|---|
EP1806658B1 (en) | 2016-04-13 |
US20070214261A1 (en) | 2007-09-13 |
EP1806658A4 (en) | 2009-07-29 |
JPWO2006046297A1 (ja) | 2008-05-22 |
JP4180638B2 (ja) | 2008-11-12 |
EP1806658A1 (en) | 2007-07-11 |
US8560667B2 (en) | 2013-10-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2006046297A1 (ja) | 分析方法及び装置 | |
US7133808B2 (en) | System and method for analyzing data streams | |
US8037042B2 (en) | Automated analysis of user search behavior | |
US8078913B2 (en) | Automated identification of performance crisis | |
EP2278502A1 (en) | Deleting data stream overload | |
US8494806B2 (en) | Method, program and apparatus for optimizing configuration parameter set of system | |
EP2177993A1 (en) | Computer system, management device, and computer system management method | |
CN110083518B (zh) | 一种基于AdaBoost-Elman的虚拟机软件老化预测方法 | |
US10956214B2 (en) | Time frame bounded execution of computational algorithms | |
JP2003178040A (ja) | ウェブサイトの構成決定支援方法 | |
Davis et al. | Failuresim: a system for predicting hardware failures in cloud data centers using neural networks | |
WO2022222906A1 (zh) | 带宽预测器的设置、服务调整方法及装置 | |
WO2017131669A1 (en) | Recommendations based on the impact of code changes | |
CN112527448A (zh) | 基于openstack的动态负载调整方法及其系统 | |
US20050096877A1 (en) | System and method for determination of load monitoring condition and load monitoring program | |
US8214693B2 (en) | Damaged software system detection | |
US20090313282A1 (en) | Automatic request categorization for internet applications | |
US11157348B1 (en) | Cognitive control of runtime resource monitoring scope | |
JP2022028611A (ja) | 情報処理に用いられる方法、装置、デバイス及び記憶媒体 | |
JP4843379B2 (ja) | 計算機システムの開発プログラム | |
CN111274112A (zh) | 应用程序压测方法、装置、计算机设备和存储介质 | |
CN115860856A (zh) | 一种数据处理方法、装置、电子设备及存储介质 | |
Rahman et al. | Predicting the performance of GridFTP transfers | |
Subair et al. | Receiver Operating Characteristic Curves in Binary Classification of Protein Secondary Structure Data | |
Casale et al. | Versatile models of systems using map queueing networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BW BY BZ CA CH CN CO CR CU CZ DK DM DZ EC EE EG ES FI GB GD GE GM HR HU ID IL IN IS JP KE KG KP KZ LC LK LR LS LT LU LV MA MD MK MN MW MX MZ NA NI NO NZ PG PH PL PT RO RU SC SD SE SG SK SY TJ TM TN TR TT TZ UA UG US UZ VN YU ZA ZM |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ NA SD SZ TZ UG ZM ZW AM AZ BY KG MD RU TJ TM AT BE BG CH CY DE DK EE ES FI FR GB GR HU IE IT MC NL PL PT RO SE SI SK TR BF CF CG CI CM GA GN GQ GW ML MR SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2006542174 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 11739946 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2004793159 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWP | Wipo information: published in national office |
Ref document number: 2004793159 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 11739946 Country of ref document: US |