WO2012046386A1 - サーバシステム、管理装置、サーバ管理方法、およびプログラム - Google Patents
サーバシステム、管理装置、サーバ管理方法、およびプログラム Download PDFInfo
- Publication number
- WO2012046386A1 WO2012046386A1 PCT/JP2011/005093 JP2011005093W WO2012046386A1 WO 2012046386 A1 WO2012046386 A1 WO 2012046386A1 JP 2011005093 W JP2011005093 W JP 2011005093W WO 2012046386 A1 WO2012046386 A1 WO 2012046386A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- server
- unit
- physical
- management device
- load data
- Prior art date
Links
Images
Classifications
-
- 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
-
- 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
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
-
- 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
- G06F11/3433—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 for load management
-
- 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
- G06F11/3442—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 planning or managing the needed capacity
-
- 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
- G06F11/3466—Performance evaluation by tracing or monitoring
-
- 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
- G06F2201/815—Virtual
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/503—Resource availability
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/508—Monitor
Definitions
- the present invention relates to a server system, a management apparatus, a server management method, and a program, and more particularly, to a server system, a management apparatus, a server management method, and a program in which a plurality of virtual servers operate on a plurality of physical servers.
- Patent Document 1 describes that in a virtual server environment in which a virtual server is operated on a plurality of physical servers, the free capacity of a physical server to which the virtual server is moved is estimated.
- load data for a certain period with periodicity that changes with time depending on the contents of the process is only determined as to whether there is capacity for the load value at a certain point in time.
- capacity is checked for load data for a certain period the number of load data increases and the amount of data becomes enormous, resulting in a problem that the calculation time for capacity determination becomes long.
- An object of the present invention is to provide a server system, a management apparatus, a server management method, and a program that solve the above-described problems.
- the server system of the present invention Multiple physical servers running multiple virtual servers, A management device connected to a plurality of the physical servers via a communication network and managing the virtual server operating on the physical server,
- the management device includes a determination unit configured to divide a process for determining whether or not a free resource capacity when the virtual server is operated satisfies a predetermined condition by performing a parallel distributed process for each predetermined unit time. .
- the management device of the present invention Connected to multiple physical servers running multiple virtual servers via a communication network,
- the physical server includes a determination unit configured to divide a process for determining whether or not a free resource capacity when the virtual server is operated satisfies a predetermined condition by performing a parallel distributed process for each predetermined unit time.
- the server management method of the present invention includes: A management device connected to a plurality of physical servers on which a plurality of virtual servers operate via a communication network and managing the virtual servers operating on the physical servers, In the physical server, the process of determining whether or not the free resource capacity when the virtual server is operated satisfies a predetermined condition is divided every predetermined unit time and is performed by parallel distributed processing.
- the program of the present invention A computer that is connected to a plurality of physical servers on which a plurality of virtual servers are operating via a communication network, and that realizes a management device that manages the virtual servers that operate on the physical servers,
- the process of determining whether or not the free resource capacity when the virtual server is operated satisfies a predetermined condition is divided every predetermined unit time, and a procedure performed by parallel distributed processing is executed. Is for.
- a plurality of components are formed as a single member, and a single component is formed of a plurality of members. It may be that a certain component is a part of another component, a part of a certain component overlaps with a part of another component, or the like.
- the plurality of procedures of the method and computer program of the present invention are not limited to being executed at different timings. For this reason, another procedure may occur during the execution of a certain procedure, or some or all of the execution timing of a certain procedure and the execution timing of another procedure may overlap.
- a server system a management apparatus, a server management method, and a program for performing resource management of a physical server when a virtual server is operated with a low load in a short time.
- FIG. 1 is a functional block diagram showing a configuration of a server system 1 according to the embodiment of the present invention.
- a server system 1 according to an embodiment of the present invention includes a plurality of physical servers 10 on which a plurality of virtual servers 20 operate, and a virtual server that is connected to the plurality of physical servers 10 via the communication network 3 and operates on the physical servers 10.
- a management apparatus 100 that manages 20. Then, the management apparatus 100 divides the processing for determining whether or not the free resource capacity when the virtual server 20 is operated in the physical server 10 satisfies a predetermined condition for each predetermined unit time. It has the determination part 104 performed by distributed processing.
- Each component of the server system 1 includes a CPU (Central Processing Unit), a memory, a program that realizes the components shown in the figure loaded in the memory, a storage unit such as a hard disk that stores the program, and a network connection interface. It is realized by any combination of computer hardware and software. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus. Each figure described below shows functional unit blocks, not hardware unit configurations. Further, in the following drawings, the configuration of parts not related to the essence of the present invention is omitted and is not shown.
- the management device 100 includes, for example, a CPU, a memory, a hard disk, and a communication device (not shown), and is connected to an input device such as a keyboard and a mouse and an output device such as a display and a printer, or a computer corresponding thereto. It is realizable with the apparatus which does. Each function of each unit can be realized by the CPU reading the program stored in the hard disk into the memory and executing it.
- the computer program according to the present embodiment performs processing for determining whether or not the free resource capacity when the virtual server 20 is operated in the physical server 10 satisfies a predetermined condition in the computer for realizing the management apparatus 100. It is described that a procedure performed by parallel distributed processing is executed by dividing every predetermined unit time.
- the computer program of the present embodiment may be recorded on a computer-readable recording medium (magnetic disk, semiconductor memory, optical disk, etc.).
- the recording medium is not particularly limited, and various forms can be considered.
- the program may be loaded from a recording medium into a computer memory, or downloaded to a computer through a network and loaded into the memory.
- the management apparatus 100 monitors the resources of the physical server 10 and the virtual server 20 that are executing business applications, and makes the virtual server 20 to another physical server 10 as necessary. Control such as migration can also be performed. At this time, the management apparatus 100 according to the present embodiment performs a determination process for finding a combination of the physical server 10 and the virtual server 20 that can efficiently use resources.
- the physical server 10 is an actual computer configured by hardware, and includes a CPU, a memory, a hard disk, and a communication control device (not shown), an input device such as a keyboard and a mouse, a display, a printer, and the like.
- a server computer a personal computer, or a device corresponding to them.
- Each function of each unit can be realized by the CPU reading the program stored in the hard disk into the memory and executing it.
- the virtual server 20 refers to a virtual computer constructed on the physical server 10 by virtualization software.
- a plurality of virtual servers 20 can be operated on the physical server 10.
- a single physical server 10 can be used simultaneously by a plurality of users, or different OSs can be executed in parallel.
- the management apparatus 100 is a computer different from the physical server 10, but can also be realized by a computer in which the physical server 10 or the virtual server 20 is operating.
- the management target system of the server system 1 is a system including a plurality of physical servers 10 on which a plurality of virtual servers 20 operate.
- a large-scale system having a large number of IT devices such as a data center. Including.
- the number of combinations of the physical server 10 and the virtual server 20 becomes enormous, the calculation time for determining the migration destination physical server 10 to which the virtual server 20 is migrated becomes long, and the processing load increases. Challenges arise. Therefore, in the server system 1 according to the embodiment of the present invention, a large-scale system having a large number of IT devices such as such a data center is obtained by dividing every predetermined unit time and performing determination processing by parallel distributed processing. It becomes possible to cope with.
- the management apparatus 100 monitors the load data for those in which load peaks of the load data of the physical server 10 and the virtual server 20 appear periodically, and efficiently. Control resources so that they can be used.
- a web server in a company has a daily periodicity in which the load increases at the time of going to and from work and lunch break.
- company payroll systems and work management systems perform data processing on a monthly basis
- the server load peaks at the end of the month or the beginning of the month, and the server load is low at other times.
- the load of the virtual server operating on the physical server changes with time depending on the processing content. Therefore, even when a load peak of load data appears periodically and the load changes within a certain period, the management apparatus 100 according to the present embodiment performs control so that resources can be used efficiently.
- Periods during which the load peaks of the physical server 10 and the virtual server 20 to be monitored appear periodically are daily, weekly, monthly, yearly, day of the week, seasonally.
- time span T Periods during which the load peaks of the physical server 10 and the virtual server 20 to be monitored appear periodically.
- the unit time ( ⁇ T) of the division processing is, for example, 1 hour, 1 day, 1 week, 1 month, 1 year, or the like.
- the resources can include the CPU, memory, network I / O (Input / Output), storage I / O, and the like of the physical server 10 or the virtual server 20.
- the server system 1 includes a management device 100 having a collection unit 102 and a determination unit 104, a load data storage unit 120 connected to the management device 100, and a determination result storage unit 130.
- the load data storage unit 120 and the determination result storage unit 130 may be included in the management apparatus 100.
- the collection unit 102 acquires the load data values of the physical server 10 and the virtual server 20 from the physical server 10 and the virtual server 20 through the communication network 3 at regular intervals, and stores them in the load data storage unit 120 as history data.
- the load data collected by the collection unit 102 includes the CPU usage rate, the memory usage rate, and the input / output performance value of the recording medium (for example, the ratio of the input / output data transfer rate to the maximum data transfer rate), It includes information on performance such as input / output performance values (transmission speed, bandwidth used, etc.) of the communication control device.
- the load data storage unit 120 stores the load data collected by the collection unit 102.
- the load data storage unit 120 can store, for each physical server 10, load data for each virtual server 20 operating on the physical server 10 as a pair for each collection time t.
- the determination unit 104 determines the storability of the virtual server 20 operating on the physical server 10 based on the load data collected by the collection unit 102. In the present embodiment, the determination unit 104 divides the load data of the physical server 10 and the virtual server 20 operating on the physical server 10 collected by the collection unit 102 for each time period, and determines the storage possibility. Further, the determination unit 104 stores the determination result of the storage possibility in the determination result storage unit 130.
- load data D1, D2,... Collected by the collection unit 102 from a plurality of physical servers 10 (PM1, PM2,..., PMm) via the communication network 3. . . , Dm are stored in the load data storage unit 120. Thereafter, for each unit time ⁇ T obtained by dividing the predetermined time span T into a plurality of data, the data d11, d21,. . . , Dm1, data d12, d22,. . . , Dm2, etc.
- the time span T is divided into four unit times ⁇ T, but the present invention is not limited to this and is shown for convenience of explanation.
- the divided data of the plurality of physical servers 10 (PM1, PM2,..., PMm) divided every unit time ⁇ T are distributed in parallel and processed.
- the processing unit U1 of the determination unit 104 physical servers PM1, PM2,. . . , PMm data d11, d21,. . . , Dm1 are processed, and the result R1 is stored in the determination result storage unit 130.
- the physical servers PM1, PM2,. . . , PMm data d12, d22,. . . , Dm2 are processed, and the result R2 is stored in the determination result storage unit 130.
- the determination process in the determination unit 104 of the management apparatus 100 of the present invention is a parallel and distributed process.
- the determination unit 104 uses a free resource on the physical server 10 in a certain period of time, for example, 24 hours, and has a certain periodic load peak. It is determined whether or not 20 can be operated, and this is shown as the possibility of storage of a certain virtual server 20 in the physical server 10.
- the determination of the storage possibility is a situation where zero or more virtual servers 20 are operating on a certain physical server 10, and when the virtual server 20 is newly operated on the physical server 10, the virtual server 20 Processing of all resources such as CPU, memory, network I / O, storage I / O, etc. without running out of capacity at all times within that period, with a period of periodicity as a unit Defined as determining whether it can be completed. However, in other embodiments, processing is performed within a certain period with periodicity without a shortage of the capacity of any predetermined resource such as CPU, memory, network I / O, or storage I / O. It can also be determined whether or not can be completed.
- the determination result storage unit 130 stores the determination result of the storage possibility performed by the determination unit 104 as “OK” or “NG”.
- the determination result is indicated as “OK” or “NG”, but the present invention is not limited to this.
- Various information such as a numerical value, a symbol, and a flag can be used as information indicating the determination result. Conceivable.
- FIG. 2 is a flowchart showing an example of the operation of the server system 1 according to the embodiment of the present invention.
- the management apparatus 100 performs a process of determining whether or not the free resource capacity when operating the virtual server in the physical server satisfies a predetermined condition.
- the data is divided for each unit time and is performed by parallel distributed processing (step S103).
- the collection unit 102 collects the load data of the physical server 10 and the virtual server 20 from each physical server 10 and stores the load data in the load data storage unit 120 (step S101). Then, the determination unit 104 refers to the load data storage unit 120 and divides the stored load data every predetermined unit time. Then, the possibility of storage is determined by parallel distributed processing, and the determination result is stored in the determination result storage unit 130 (step S103).
- the operator determines the determination result of the storage possibility when the virtual server 20 is operated on each physical server 10 stored in the determination result storage unit 130 as required by the operator. It can have a presentation unit (not shown) for presentation so that it can be browsed.
- the determination result can be displayed on an output device (not shown) such as a display, or can be printed out on an output device (not shown) such as a printer.
- a processing unit that displays an operation screen for performing an operation for searching for a determination result of storability using an input device (not shown) such as a keyboard on the display, accepts an operator's operation, and performs a search process ( (Not shown).
- the determination result may be presented as a list of determination results of all physical servers.
- the list may be sortable for each determination result.
- the list may include other performance information (for example, CPU performance, memory capacity, etc.) of the physical server.
- the load data of the physical server 10 and the virtual server 20 is stored in time.
- parallel distributed processing becomes possible. Therefore, according to the present invention, it is possible to select the physical server 10 suitable for the destination of the virtual server 20 with a low load in a short time.
- resource management of a physical server when a virtual server is operated can be performed in a short time with a low load. According to the present invention, it is possible to avoid insufficient capacity of the virtual server 20 and generation of useless free resources, and resources can be used efficiently.
- the present invention it is possible to verify in a short time that the capacity necessary for the execution of the virtual server 20 is sufficiently secured especially for load data of a certain period with periodicity. This is because the possibility of storage can be determined for the entire cyclic load data of the virtual server.
- the server system 1 of the present invention even when the number of physical servers and virtual servers increases, there is an effect that it is possible to quickly determine a physical server having sufficient capacity to operate the virtual server. . As a result, it is possible to avoid the capacity shortage of the virtual server and the generation of useless free resources. This is because the storage possibility determination process is realized as a process that can be executed in parallel and distributed.
- FIG. 3 is a functional block diagram showing the configuration of the server system 1 according to the embodiment of the present invention.
- the management device 200 of the server system 1 according to the present embodiment has a specific configuration that performs processing based on the load data collected by the collection unit 102 in the management device 100 according to the above-described embodiment.
- the server system 1 will be described as an example in which the number of physical servers 10 and virtual servers 20 is not so large and the size of load data (the number of load data to be monitored) is medium.
- the management device 200 further includes a generation unit 202 and a calculation unit 204 in addition to the configuration of the management device 100 of the above-described embodiment of FIG.
- the generation unit 202 uses the identifier (PM_ID) of each physical server 10 and the collection time t as keys from the load data of the physical server 10 and the virtual server 20 collected from each physical server 10 by the collection unit 102. Generate data for value pairs. Then, the generation unit 202 generates paired data with the unit time ⁇ T of the division processing as a key and the log at time t included in the time as value. Each data generated by the generation unit 202 can be stored in the load data storage unit 120.
- the calculation unit 204 calculates the surplus resource amount of the physical server 10 when the virtual server 20 is operated for all ⁇ T included in the time span T.
- FIG. 4 is a flowchart showing an example of the operation of the server system 1 according to the embodiment of the present invention.
- FIG. 5 is a table showing examples of data formats handled in a series of processes in the server system 1. This will be described below with reference to FIGS.
- the collection unit 102 of the management device 200 acquires the load data of each physical server 10 at regular intervals via the communication network 3 and stores it in the load data storage unit 120 (step S201).
- the data format at this time is as shown in step S201 in Table 1 of FIG.
- the load data includes, for example, the collection time t (2010/01/27 13:05:02), identification information (PM1, VM11) of the physical server 10 and the virtual server 20, and the virtual server 20 Includes CPU utilization (CPU 32%), network I / O performance value (NW send 253Mbps), and so on.
- the generation unit 202 of the management device 200 refers to the load data storage unit 120, and from each load data, the identifier (PM_ID) of the physical server 10 and the time t are keys, and the load data (“load_info_1 in Table 1 in FIG. (Shown as VM_ID + CPU + NW_send + ...) generates value pair data (step S203)
- the data format at this time is as shown in step S203 of Table 1 of FIG.
- the generation unit 202 of the management apparatus 200 refers to the load data storage unit 120, and the log of the time t included in the unit time ⁇ T of the division processing is a key (in Table 1, FIG. Pair data having a value “load_info_2 (list of PM_ID + t + load_info_1)”) is generated (step S205).
- the data format at this time is as shown in step S205 in Table 1 of FIG.
- the determination unit 104 of the management apparatus 200 refers to the load data storage unit 120, generates a pair of PM_ID and virtual server identifier (VM_ID) from the log included in the same ⁇ T, and determines the storage possibility. Then, the determination unit 104 of the management apparatus 200 stores the determination result (corresponding to “result_2 (PM_ID, VM_ID, OK / NG)” in Table 1 of FIG. 5) in the determination result storage unit 130 (step S207). . Further, the calculation unit 204 of the management apparatus 200 calculates the surplus resource amount (corresponding to “result_2 (PM_ID, VM_ID, surplus resource)” in Table 1 of FIG. 5) with reference to the load data storage unit 120, It memorize
- the determination unit 104 of the management apparatus 200 refers to the determination result storage unit 130 to determine the storage possibility for all ⁇ T included in the time span T. Then, the determination unit 104 of the management apparatus 200 stores a determination result (corresponding to “result_1 (OK / NG)” in Table 1 of FIG. 5) corresponding to the PM_ID and VM_ID pair data in the determination result storage unit 130. (Step S211). Then, the calculation unit 204 of the management device 200 refers to the determination result storage unit 130 and refers to the surplus resource of the physical server 10 corresponding to the data of the PM_ID and VM_ID pair (“result_1 (surplus resource) in Table 1 in FIG. 5). Is calculated) and stored in the determination result storage unit 130 (step S213). The data format at this time is as shown in steps S211 and S213 of Table 1 in FIG.
- the determination result of the storage possibility when the virtual server 20 is operated on each physical server 10 stored in the determination result storage unit 130 in this manner is the same as in the above embodiment, in response to a search request from the operator. Search results can be presented to the operator. This makes it possible for the operator to determine in a short time the selection of the physical server 10 suitable for the destination of the virtual server 20. In addition, by presenting the surplus resource amounts so that they can be compared, it is possible to make the operator determine the selection of the physical server 10 with less waste.
- FIG. 6 is a functional block diagram showing the configuration of the server system 1 according to the embodiment of the present invention.
- the management apparatus 300 of the server system 1 of the present embodiment is different from the above-described embodiment of FIG. 3 in that the determination processing is performed by shortening the unit time of the division processing of the determination unit 104 when the number of load data is large. It is different in that it reduces the load.
- the management apparatus 300 includes, in addition to the configuration of the management apparatus 200 according to the above-described embodiment in FIG. 3, load data that is processed within a unit time is less than a predetermined number. Further, a determination unit 302 that determines the minimum value of the unit time is further provided. Furthermore, the determination unit 104 of the management apparatus 300 divides the unit time into the determined minimum value and performs a determination process using the minimum unit time. In the present embodiment, an example in which the server system 1 has a large number of physical servers 10 and virtual servers 20 and a large load data size (number of load data to be monitored) will be described.
- the management apparatus 300 includes a generation unit 304 instead of the generation unit 202 of the management apparatus 200 according to the above-described embodiment illustrated in FIG. It has the determination part 104 and the calculation part 204, and also has the determination part 302.
- the minimum value of the unit time may be determined so that the number of pairs of the physical server 10 and the virtual server 20 corresponding to the load data processed within the unit time is less than the predetermined number within the time. it can. It is preferable to perform the main process of the determination unit 302 when the number of processes per unit time of the load data collected by the collection unit 102 is large and the determination process load is higher than a predetermined reference.
- the generation unit 304 uses the physical server 10 and the virtual server 20 load data collected by the collection unit 102 from each physical server 10,
- the identifier (PM_ID) and the collection time t are keys, and the load data is value pair data. Then, the generation unit 202 generates paired data with the unit time ⁇ T of the division processing as a key and the log at time t included in the time as value.
- the generation unit 304 uses the minimum value ⁇ Tmin of the unit time ⁇ T as a key, and sets a pair of logs whose value is a log at time t included in that time. Generate data.
- Each of the data generated by the generation unit 304 can be stored in the load data storage unit 120.
- FIG. 7 is a flowchart showing an example of the operation of the management apparatus 300 of the server system 1 according to the embodiment of the present invention.
- FIG. 8 is a table showing examples of data formats handled in a series of processing in the server system 1.
- management apparatus 300 of the present embodiment performs processing in the same procedure as the flowchart showing the operation of the management apparatus 200 of the above-described embodiment of FIG. 4, and between steps S205 and S207 of FIG. Steps S301 to S307 shown in FIG. This will be described below with reference to FIGS. 4 and 6 to 8.
- the collection unit 102 of the management apparatus 300 acquires the load data of each physical server 10 via the communication network 3 at regular intervals and stores it in the load data storage unit 120 (see FIG. 4 step S201).
- the data format at this time is as shown in step S201 in Table 2 of FIG.
- the generation unit 304 of the management device 300 refers to the load data storage unit 120, and from each load data, the identifier (PM_ID) of each physical server 10 and the collection time t are the keys, and the load data (in Table 2 of FIG. “Load_info_1 (indicated as VM_ID + CPU + NW_send +...)” Generates value pair data (step S203 in FIG. 4) The data format at this time is as shown in step S203 of Table 2 in FIG. .
- the generation unit 304 of the management apparatus 300 refers to the load data storage unit 120, and the unit time ⁇ T of the division process is a key, and the log of the time t included in the time (in Table 2, FIG. 8, “load_info_2 ( A list of “PM_ID + t + load_info_1”) is generated as a value (step S205 in FIG. 4).
- the data format at this time is as shown in step S205 in Table 2 of FIG.
- the management apparatus 300 transfers a process to step S301 of FIG.
- the determination unit 302 of the management apparatus 300 refers to the load data storage unit 120 and determines whether or not the number of pairs of data generated in step S205 is greater than a predetermined number (step S301 in FIG. 7). If the number of data pairs is greater than the predetermined number (YES in step S301 in FIG. 7), the determination unit 302 of the management apparatus 300 refers to the load data storage unit 120 and loads processed within a unit time The minimum value ⁇ Tmin of the unit time ⁇ T is determined so that less than a predetermined number of data is included (step S302 in FIG. 7). When the number of data pairs is smaller than the predetermined number (NO in step S301 in FIG. 7), the management apparatus 300 returns to the flow in FIG. 4 and continues the process in step S207 in FIG.
- the generation unit 304 of the management apparatus 300 refers to the load data storage unit 120, and the log of the time t included in the time with the minimum value ⁇ Tmin of the unit time ⁇ T as a key (in Table 2 of FIG. 8, Pair data with “load_info_3 (load_info_2 list)” as a value is generated (step S303 in FIG. 7).
- the data format at this time is as shown in step S303 in Table 2 of FIG.
- the determination unit 104 of the management apparatus 300 refers to the load data storage unit 120, generates PM_ID and VM_ID pair data from the log included in the same ⁇ Tmin, and determines the storage possibility. Then, the determination unit 104 of the management apparatus 300 stores the determination result (corresponding to “result_3 (PM_ID, VM_ID, OK / NG)” in Table 2 of FIG. 8) in the determination result storage unit 130 (FIG. 7). Step S305). Further, the calculation unit 204 of the management apparatus 300 refers to the load data storage unit 120 and calculates a surplus resource amount (corresponding to “result_3 (PM_ID, VM_ID, surplus resource)” in Table 2 of FIG. 8).
- step S307 of FIG. 7 It memorize
- the data format at this time is as shown in steps S305 and S307 in Table 2 of FIG. Then, the management apparatus 300 returns to the flow of FIG. 4 and continues the process of step S207 of FIG.
- the determination unit 104 of the management apparatus 300 refers to the determination result storage unit 130 and determines the storage possibility for all ⁇ Tmin included in ⁇ T. Then, the determination unit 104 of the management apparatus 300 stores the determination result (corresponding to “result_2 (PM_ID, VM_ID, OK / NG)” in Table 2 of FIG. 8) in the determination result storage unit 130 (FIG. 4). Step S207). Further, the calculation unit 204 of the management apparatus 300 refers to the determination result storage unit 130 and calculates a surplus resource amount (corresponding to “result_2 (PM_ID, VM_ID, surplus resource)” in Table 2 of FIG. 8). The result is stored in the determination result storage unit 130 (step S209 in FIG. 4). The data format at this time is as shown in steps S207 and S209 of Table 2 in FIG.
- the determination unit 104 of the management device 300 refers to the determination result storage unit 130 to determine the storage possibility for all ⁇ T included in the time span T. Then, the determination unit 104 of the management apparatus 300 stores the determination result (corresponding to “result_1 (OK / NG)” in Table 2 of FIG. 8) corresponding to the data of the PM_ID and VM_ID pair in the determination result storage unit 130. (Step S211 in FIG. 4). Then, the calculation unit 204 of the management apparatus 300 refers to the determination result storage unit 130, and the surplus resource of the physical server 10 corresponding to the PM_ID and VM_ID pair data ("result_1 (surplus resource)" in Table 2 in FIG. 8). Is calculated) and stored in the determination result storage unit 130 (step S213 in FIG. 4). The data format at this time is as shown in steps S211 and S213 in Table 2 of FIG.
- the determination result of the storage possibility when the virtual server 20 is operated on each physical server 10 stored in the determination result storage unit 130 in this manner is the same as in the above embodiment, in response to a search request from the operator. Search results can be presented to the operator. This makes it possible for the operator to determine in a short time the selection of a physical server suitable for the destination of the virtual server 20. In addition, by presenting the surplus resource amounts so that they can be compared, it is possible to make the operator determine the selection of the physical server 10 with less waste.
- the same effects as the above embodiment can be obtained, and load data of the physical server 10 and the virtual server 20 can be divided and stored for each minimum value ⁇ Tmin of the division processing unit time.
- parallel distributed processing with smaller granularity becomes possible. Therefore, according to the present invention, even when the amount of data becomes large, the physical server 10 suitable for the destination of the virtual server 20 can be selected in a short time, and the capacity of the virtual server 20 is insufficient or the physical server 10 Generation of useless free resources can be avoided.
- FIG. 9 is a functional block diagram showing the configuration of the server system 1 according to the embodiment of the present invention.
- the management apparatus 400 of the server system 1 according to the present embodiment groups servers having load data having the same load pattern as the above-described embodiments of FIGS. 3 and 6, and selects a representative server from the group.
- the determination process is different in that the load of the determination process is reduced.
- the management device 400 of the server system 1 of the present embodiment is particularly useful in the case of a system in which many servers with similar load patterns exist. It is also useful in the case of a system in which there are many servers with low load and almost no change in load.
- the management device 400 classifies the load data collected by the collection unit 102 into groups based on the range of unit loads, and receives a predetermined number of load data from each group.
- a selection unit 402 for selecting is further provided.
- the determination unit 104 of the management apparatus 400 performs a determination process based on the selected load data.
- the server system 1 has an example in which the number of physical servers 10 and virtual servers 20 is very large and the size of load data (the number of load data to be monitored) is extremely large. explain.
- the management apparatus 400 includes a generation unit 406 instead of the generation unit 202 of the management apparatus 200 or the generation unit 304 of the management apparatus 300 according to the above-described embodiment illustrated in FIG. Similar to the management device 200 and the management device 300, the collection unit 102, the determination unit 104, and the calculation unit 204 are included. Instead of the determination unit 302 of the management device 300, the determination unit 404 is included. Part 402.
- the selection unit 402 classifies the load data collected by the collection unit 102 into groups based on the unit load range, and selects a predetermined number of load data from each group. As will be described later, the selection unit 402 loads the load when the number of pairs of load data of the physical server 10 and the virtual server 20 included in the unit time ⁇ T or the minimum value ⁇ Tmin of the unit time ⁇ T is larger than a predetermined number of pairs. Data selection processing can be performed. Alternatively, the selection process may be performed when there are more data having a similar load pattern than a predetermined number of data.
- the range of the unit load is, for example, a range such as an average load of the physical server 10 per unit time, a peak load, or a load fluctuation range.
- the selection unit 402 divides the physical servers 10 having a predetermined range of average load, peak load, or load fluctuation range into groups, and selects a representative server from the group.
- the determination unit 404 determines the minimum value of unit time so that less than a predetermined number of load data to be processed within the unit time is included. Note that the management apparatus 400 according to the present embodiment includes the determination unit 404, but is not necessarily required, and there is no configuration for determining the minimum unit time as in the management apparatus 200 of FIG. Also good.
- the generation unit 406 Similar to the generation unit 202 of the management apparatus 200 of the above-described embodiment of FIG. 3, the generation unit 406 generates each physical server 10 from the load data of the physical server 10 and the virtual server 20 collected from each physical server 10 by the collection unit 102.
- the identifier (PM_ID) and the collection time t are keys, and the load data is value pair data. Then, the generation unit 202 generates paired data with the unit time ⁇ T of the division processing as a key and the log at time t included in the time as value.
- the generation unit 406 like the generation unit 304 of the management apparatus 300 of the above embodiment of FIG. 6, determines that the minimum value ⁇ Tmin of the unit time ⁇ T is a key when the number of generated data pairs is greater than a predetermined number. Thus, pair data with the value of the log at time t included in the time is generated.
- Each data generated by the generation unit 406 can be stored in the load data storage unit 120.
- FIG. 10 is a flowchart showing an example of the operation of the management apparatus 400 of the server system 1 according to the embodiment of the present invention.
- FIG. 11 is a table showing examples of data formats handled in a series of processes in the server system 1. Note that the management apparatus 400 according to the present embodiment performs processing in the same procedure as the flowchart showing the operation of the management apparatus 200 according to the above-described embodiment in FIG. 4, and between steps S205 and S207 in FIG. Steps S301 to S307 shown in FIG. 10 are performed, and further, Steps S401 to S405 in FIG. 10 are performed between Steps S303 and S305 in FIG. Hereinafter, description will be made with reference to FIGS. 4, 7, and 9 to 11.
- the collection unit 102 of the management apparatus 400 acquires the load data of each physical server 10 at regular intervals via the communication network 3 and stores the load data in the load data storage unit 120 (see FIG. 4 step S201).
- the data format at this time is as shown in step S201 in Table 3 of FIG.
- the generation unit 406 of the management device 400 refers to the load data storage unit 120, and from each load data, the identifier (PM_ID) of each physical server 10 and the collection time t are keys, and the load data (in Table 3 of FIG. “Load_info_1 (indicated as VM_ID + CPU + NW_send +...)” Generates value pair data (step S203 in FIG. 4) The data format at this time is as shown in step S203 in Table 3 of FIG. .
- the generation unit 406 of the management apparatus 400 refers to the load data storage unit 120, the unit time ⁇ T of the division processing is a key, and the log of the time t included in the time (in Table 3, FIG. 11, “load_info_2 ( A list of “PM_ID + t + load_info_1”) is generated as a value (step S205 in FIG. 4).
- the data format at this time is as shown in step S205 in Table 3 of FIG.
- the management apparatus 400 transfers a process to step S301 of FIG.
- the determination unit 404 of the management device 400 refers to the load data storage unit 120 and determines whether or not the number of data pairs generated in step S205 is greater than a predetermined number (step S301 in FIG. 7). If the number of data pairs is large (YES in step S301 in FIG. 7), the determination unit 404 of the management apparatus 400 refers to the load data storage unit 120 and loads data to be processed within a unit time is predetermined. The minimum value ⁇ Tmin of the unit time ⁇ T is determined so that it is less than the number (step S302 in FIG. 7).
- the generation unit 304 of the management device 400 refers to the load data storage unit 120 and uses the minimum value ⁇ Tmin of the unit time ⁇ T as a key, and the log of the time t included in the time (in Table 3 of FIG. 11, Pair data with “load_info_3 (load_info_2 list)” as a value is generated (step S303 in FIG. 7).
- the data format at this time is as shown in step S303 in Table 3 of FIG.
- the management apparatus 400 transfers a process to step S401 of FIG.
- the selection unit 402 of the management device 400 refers to the load data storage unit 120 and determines whether or not the number of data pairs included in ⁇ Tmin generated in step S303 is greater than a predetermined number (step S401 in FIG. 10). . If the number of data pairs included in ⁇ Tmin is greater than the predetermined number (YES in step S401 in FIG. 10), the data pairs are grouped for each unit load ⁇ L (step S403 in FIG. 10). The data format at this time is as shown in step S403 in Table 3 of FIG. When the number of data pairs included in ⁇ Tmin is smaller than the predetermined number (NO in step S401 in FIG. 10), the management apparatus 400 returns to the flow in FIG. 7 and continues the process in step S305 in FIG.
- the selection unit 402 of the management apparatus 400 refers to the load data storage unit 120, selects r representative data pairs among the data pairs included in ⁇ L, and replaces the data pairs included in ⁇ Tmin (see FIG. 10 step S405).
- the data format at this time is as shown in step S405 in Table 3 of FIG.
- the management apparatus 400 returns to the flow of FIG. 7 and continues the process of step S305 of FIG.
- the determination unit 104 of the management device 400 refers to the load data storage unit 120, generates PM_ID and VM_ID pair data from the log included in the same ⁇ Tmin, and determines the storage possibility. Then, the determination unit 104 of the management apparatus 400 stores the determination result (corresponding to “result_3 (PM_ID, VM_ID, OK / NG)” in Table 3 of FIG. 11) in the determination result storage unit 130 (FIG. 7). Step S305). Further, the calculation unit 204 of the management device 400 calculates the surplus resource amount (corresponding to “result_3 (PM_ID, VM_ID, surplus resource)” in Table 3 of FIG.
- step S307 of FIG. 7 The data format at this time is as shown in steps S305 and S307 in Table 3 of FIG. Then, the management apparatus 400 returns to the flow of FIG. 4 and continues the process of step S207 of FIG.
- the determination unit 104 of the management apparatus 400 refers to the determination result storage unit 130 and determines the storability of all ⁇ Tmin included in ⁇ T. Then, the determination unit 104 of the management apparatus 400 stores the determination result (corresponding to “result_3 (PM_ID, VM_ID, OK / NG)” in Table 2 of FIG. 8) in the determination result storage unit 130 (FIG. 4). Step S207). Furthermore, the calculation unit 204 of the management apparatus 400 calculates the surplus resource amount (corresponding to “result_2 (PM_ID, VM_ID, surplus resource)” in Table 3 of FIG. 11) with reference to the determination result storage unit 130, The result is stored in the determination result storage unit 130 (step S209 in FIG. 4). The data format at this time is as shown in steps S207 and S209 of Table 3 in FIG.
- the determination unit 104 of the management device 400 refers to the determination result storage unit 130 to determine the storage possibility for all ⁇ T included in the time span T. Then, the determination unit 104 of the management apparatus 400 stores the determination result (corresponding to “result_1 (OK / NG)” in Table 3 of FIG. 11) corresponding to the PM_ID and VM_ID pair data in the determination result storage unit 130. (Step S211 in FIG. 4). Then, the calculation unit 204 of the management apparatus 400 refers to the determination result storage unit 130 and refers to the surplus resource of the physical server 10 corresponding to the PM_ID and VM_ID pair data (“result_1 (surplus resource) in Table 3 of FIG. 11). Is calculated) and stored in the determination result storage unit 130 (step S213 in FIG. 4). The data format at this time is as shown in steps S211 and S213 in Table 3 of FIG.
- the determination result of the storage possibility when the virtual server 20 is operated on each physical server 10 stored in the determination result storage unit 130 in this manner is the same as in the above embodiment, in response to a search request from the operator. Search results can be presented to the operator. This makes it possible for the operator to determine in a short time the selection of the physical server 10 suitable for the destination of the virtual server 20. In addition, by presenting the surplus resource amounts so that they can be compared, it is possible to make the operator determine the selection of the physical server 10 with less waste.
- the same effects as those of the above-described embodiment can be obtained, and the load data of the physical server 10 and the virtual server 20 is divided for each minimum value ⁇ Tmin of the unit time of the division processing, and further the unit By narrowing down the representative value for each range of the load ⁇ L and executing the storage possibility determination process, parallel distributed processing with smaller granularity and fewer combinations becomes possible. Therefore, according to the present invention, even when the amount of data becomes extremely large, the physical server 10 suitable for the destination of the virtual server 20 can be selected in a short time, and the capacity of the virtual server 20 is insufficient or the physical server 10 Generation of useless free resources can be avoided.
- the storage possibility determination process in the determination unit of the management device may be configured to perform the following determination.
- the first determination method is a method that places importance on safety, and determines whether the free resource capacity satisfies the requested resource amount over the entire period. This is a determination method for online processing.
- the second determination method it is determined whether the average free resource capacity satisfies the requested resource amount.
- This second determination method is a method suitable for a system in which the resource capacity may temporarily reach the upper limit, and the processing only needs to be completed within a predetermined period (within one cycle) (for example, within 24 hours).
- the determination unit of the management apparatus periodically shows a free resource capacity and a load data load peak in the determination process. It is determined whether or not the required condition is satisfied over a predetermined period. If this requirement is satisfied, the determination unit can determine that there is a possibility of storage.
- the determination unit of the management apparatus has a free space within a predetermined period in which load data load peaks appear periodically in the determination process. It is determined whether the average value of the resource capacity satisfies the average value requirement condition. If this requirement is satisfied, the determination unit can determine that there is a possibility of storage.
- the management apparatus can further include a resource presentation unit (not shown) that presents the calculated surplus resource amount to the user. Furthermore, you may further provide the determination result presentation part (not shown) which presents the determination result by the determination part of a management apparatus to a user.
- FIG. 12 is a diagram showing an example of a screen displayed on a display device (not shown) by the resource presentation unit of the server system 1 according to the embodiment of the present invention.
- the screen 500 includes, as a list 502, surplus resources 516 such as a CPU, a memory, or a network for data of a pair of a virtual server 512 (VM) and a destination physical server 514 (PM).
- the resource selection list 518 can be sorted in ascending order of the surplus resource amount of resources selected by the operator from the CPU, memory, or network using the operation unit (not shown).
- the surplus resource amount of the physical server when the virtual server is moved can be compared.
- the list 502 includes only the items determined as “OK” in the determination of the storage possibility.
- a list of all the physical server 10 and virtual server 20 pairs may be included.
- information (“OK” or “NG” or the like) indicating the determination result of the storage possibility is included in the list 502. It is desirable to display.
- the list 502 may include a move button 520 that receives an instruction to move the virtual server 512 to the destination physical server 514 for each pair of the virtual server 512 and the destination physical server 514.
- the operator can select a pair of the physical server 10 and the virtual server 20 with the move button 520 with reference to the list 502, and can accept an instruction to execute migration with the OK button 522.
- the OK button 522 When the operator operates the OK button 522, a pair of the physical server 10 and the virtual server 20 selected by the move button 520 in the list 502 is accepted as an instruction to migrate the virtual server 512 to the destination physical server 514. Can do. If the operator determines that the migration is not necessary with reference to the list 502, the screen 500 is closed or the selection by the move button 520 is canceled by operating the cancel button 524. Also good.
- the management apparatus 100 may further include a moving unit that migrates the received virtual server to the received physical server when receiving an instruction to migrate the virtual server to the physical server. Alternatively, migration may be instructed to another management apparatus or physical server.
- the present invention can also include the following aspects.
- (Appendix 1) Connected to multiple physical servers running multiple virtual servers via a communication network,
- the physical server includes a determination unit that performs a process of determining whether or not a free resource capacity when operating the virtual server satisfies a predetermined condition by dividing the process every predetermined unit time and performing parallel distributed processing.
- it comprises a collecting means for collecting load data of the plurality of virtual servers and the plurality of physical servers via the communication network, in which predetermined processing is executed
- the determination unit is a management device that performs the determination process based on the load data collected by the collection unit.
- a determination unit for determining a minimum value of the unit time so that less than a predetermined number of the load data to be processed within the unit time are included;
- the determination unit is a management device that divides the unit time into the determined minimum value and performs the determination process with the unit time of the minimum value.
- the load data collected by the collection means is further classified into groups based on a range of unit loads, and further comprises selection means for selecting a predetermined number of load data from each of the groups,
- the determination unit is a management device that performs the determination process based on the selected load data.
- the determination unit of the management apparatus determines whether the free resource capacity satisfies a required condition over a predetermined period in which a load peak of the load data periodically appears. .
- the determination unit of the management apparatus determines whether an average value of the free resource capacity within a predetermined period in which a load peak of the load data periodically appears satisfies a request condition for an average value
- a management device that determines whether or not.
- the management device Calculation means for calculating the surplus resource amount of the physical server based on the load data; Resource presenting means for presenting the calculated surplus resource amount to the user; A management device further comprising:
- a management apparatus further comprising determination result presentation means for presenting a determination result by the determination means of the management apparatus to a user.
- a management device connected to a plurality of physical servers on which a plurality of virtual servers operate via a communication network and managing the virtual servers operating on the physical servers,
- the process of determining whether or not the free resource capacity when operating the virtual server satisfies a predetermined condition is divided every predetermined unit time, and is performed by parallel distributed processing, Further, load data of the plurality of virtual servers and the plurality of physical servers on which predetermined processing is executed are collected via the communication network, A server management method for performing the determination process based on the collected load data.
- the management device is Determining a minimum value of the unit time so that the load data to be processed within the unit time includes less than a predetermined number;
- the server management method which divides
- the management device is Based on the collected load data and unit load range, classify into groups, select a predetermined number of load data from each group, A server management method for performing the determination processing based on the selected load data.
- the management device is In the determination process, a server management method for determining whether or not the free resource capacity satisfies a required condition over a predetermined period in which a load peak of the load data periodically appears.
- the management device is In the determination process, a server management method for determining whether or not an average value of the free resource capacity within a predetermined period in which a load peak of the load data periodically appears satisfies a request condition for the average value.
- the management device is Based on the load data, calculate the surplus resource amount of the physical server, A server management method for presenting the calculated surplus resource amount to a user.
- a management device connected to a plurality of the physical servers via a communication network and managing the virtual server operating on the physical server,
- the management device includes a determination unit that performs a process of determining whether or not a free resource capacity when operating the virtual server satisfies a predetermined condition by dividing the process every predetermined unit time and performing parallel distributed processing.
- a server system further comprising determination result presentation means for presenting a determination result by the determination means of the management device to a user.
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Environmental & Geological Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
複数の仮想サーバが稼動する複数の物理サーバと、
複数の前記物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置と、を備え、
前記管理装置は、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う判定手段を備える。
複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う判定手段を備える。
複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置が、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う。
複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置を実現するコンピュータに、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う手順を実行させるためのものである。
図1は、本発明の実施の形態に係るサーバシステム1の構成を示す機能ブロック図である。
本発明の実施の形態に係るサーバシステム1は、複数の仮想サーバ20が稼動する複数の物理サーバ10と、複数の物理サーバ10に通信網3で接続され、物理サーバ10上で稼働する仮想サーバ20を管理する管理装置100と、を備える。そして、管理装置100は、物理サーバ10において、仮想サーバ20を稼働させたときの空きリソース容量が、所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う判定部104を有する。図1では、物理サーバPM1で稼動する仮想サーバに符号VM11、...、VM1n(nは自然数)を付与し、物理サーバPMmで稼動する仮想サーバに符号VMm1、...、VMmi(mとiは自然数)を付与する。以下の説明では、これらの符号を用いてそれぞれを区別する。
また、以下の各図において、本発明の本質に関わらない部分の構成については省略してあり、図示されていない。
図1では、管理装置100は、物理サーバ10とは異なるコンピュータとなっているが、物理サーバ10または仮想サーバ20が稼働しているコンピュータにより実現することもできる。
また、リソースは、物理サーバ10または仮想サーバ20のCPU、メモリ、ネットワークI/O(Input/Output)、およびストレージI/Oなどを含むことができる。
なお、負荷データ記憶部120および判定結果記憶部130は、管理装置100に含まれてもよい。
たとえば、図13に示すように、まず、複数の物理サーバ10(PM1、PM2、...、PMm)から通信網3を介して収集部102が収集した負荷データD1、D2、...、Dmは、負荷データ記憶部120に格納する。その後、所定のタイムスパンTを複数に分割した単位時間△Tごとに、それぞれ、時刻t1からt2の間のデータd11、d21、...、dm1、時刻t2からt3の間のデータd12、d22、...、dm2などに分割される。なお、本図では、タイムスパンTが4つの単位時間△Tに分割されているが、これに限定されるものではなく、説明上便宜的に示しているだけである。そして、判定部104において、複数の処理部U1、U2、U3、...が、この単位時間△Tごとに分割された複数の物理サーバ10(PM1、PM2、...、PMm)の分割データをそれぞれ並列に分散して処理する。そして、各処理部から処理された複数の結果R1、R2、R3、...が判定結果記憶部130に格納される。
ただし、他の実施形態では、CPU、メモリ、ネットワークI/O、またはストレージI/Oなどのうち予め決められたいずれかのリソースのキャパシティが不足することなく周期性のある一定期間内に処理を完了できるかどうかを判定することもできる。
図2は、本発明の実施の形態に係るサーバシステム1の動作の例を示すフローチャートである。以下、図1および図2を用いて説明する。
本発明の実施の形態に係るサーバ管理方法は、管理装置100が、物理サーバにおいて、仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う(ステップS103)。
オペレータが、キーボードなどの入力装置(不図示)を用いて、格納可能性の判定結果を検索する操作を行う操作画面をディスプレイに表示し、オペレータの操作を受け付けて、検索処理を行う処理部(不図示)を有することができる。
図3は、本発明の実施の形態に係るサーバシステム1の構成を示す機能ブロック図である。
本実施形態のサーバシステム1の管理装置200は、上記実施の形態の管理装置100における、収集部102が収集した負荷データに基づく処理を行う具体的な構成を有する。
なお、本実施形態において、サーバシステム1は、物理サーバ10および仮想サーバ20の台数がそれほど多くなく、負荷データのサイズ(監視対象となる負荷データ件数)が中程度の場合の例について説明する。
生成部202は、はじめに、収集部102が各物理サーバ10から収集した物理サーバ10および仮想サーバ20の負荷データから、各物理サーバ10の識別子(PM_ID)と収集時刻tがキーで、負荷データがバリューのペアのデータを生成する。そして、生成部202は、分割処理の単位時間△Tがキーで、その時間内に含まれる時刻tのログをバリューとするペアのデータを生成する。
生成部202によって生成された上記各データは、負荷データ記憶部120に記憶することができる。
図4は、本発明の実施の形態に係るサーバシステム1の動作の例を示すフローチャートである。また、図5は、サーバシステム1における一連の処理で扱うデータ形式の例を一覧表で示す図である。以下、図3乃至図5を用いて説明する。
これにより、仮想サーバ20の移動先に適した物理サーバ10の選択をオペレータに短時間で判断させることが可能になる。また、余剰リソース量を比較できるように提示することにより、無駄の少ない物理サーバ10の選択をオペレータに判断させることが可能になる。
図6は、本発明の実施の形態に係るサーバシステム1の構成を示す機能ブロック図である。
本実施形態のサーバシステム1の管理装置300は、図3の上記実施形態とは、負荷データの件数が多い場合などに、判定部104の分割処理の単位時間をより短くすることで、判定処理の負荷を低減する点で相違する。
なお、本実施形態において、サーバシステム1は、物理サーバ10および仮想サーバ20の台数が多く、負荷データのサイズ(監視対象となる負荷データ件数)が大きい場合の例について説明する。
生成部304によって生成された上記各データは、負荷データ記憶部120に記憶することができる。
図7は、本発明の実施の形態に係るサーバシステム1の管理装置300の動作の例を示すフローチャートである。また、図8は、サーバシステム1における一連の処理で扱うデータ形式の例を一覧表で示す図である。
これにより、仮想サーバ20の移動先に適した物理サーバの選択をオペレータに短時間で判断させることが可能になる。また、余剰リソース量を比較できるように提示することにより、無駄の少ない物理サーバ10の選択をオペレータに判断させることが可能になる。
図9は、本発明の実施の形態に係るサーバシステム1の構成を示す機能ブロック図である。
本実施形態のサーバシステム1の管理装置400は、図3および図6の上記実施形態とは、同様な負荷パターンの負荷データを有するサーバをグループ化し、グループの中から代表のサーバを選択して判定処理を行うことで、判定処理の負荷を低減する点で相違する。
なお、本実施形態において、サーバシステム1は、物理サーバ10および仮想サーバ20の台数が非常に多く、かつ、負荷データのサイズ(監視対象となる負荷データ件数)が極端に大きいの場合の例について説明する。
なお、本実施形態の管理装置400は、決定部404を有しているが、必ずしも必要はなく、図3の管理装置200と同様に、単位時間の最小値を決定する構成は有さなくてもよい。
生成部406によって生成された上記各データは、負荷データ記憶部120に記憶することができる。
図10は、本発明の実施の形態に係るサーバシステム1の管理装置400の動作の例を示すフローチャートである。また、図11は、サーバシステム1における一連の処理で扱うデータ形式の例を一覧表で示す図である。
なお、本実施形態の管理装置400は、図4の上記実施形態の管理装置200の動作を示すフローチャートと同様な手順で処理を行うとともに、図4のステップS205とステップS207の間に、図7に示すステップS301乃至ステップS307の処理を行い、さらに、図7のステップS303とステップS305の間に、図10のステップS401乃至ステップS405の処理を行う。以下、図4、図7、図9乃至図11を用いて説明する。
これにより、仮想サーバ20の移動先に適した物理サーバ10の選択をオペレータに短時間で判断させることが可能になる。また、余剰リソース量を比較できるように提示することにより、無駄の少ない物理サーバ10の選択をオペレータに判断させることが可能になる。
たとえば、本発明の他の実施形態のサーバシステム1において、管理装置の判定部における格納可能性の判定処理は、以下の判定を行う構成としてもよい。第1の判定方法は、安全性を重視した方法であり、空きリソース容量が全期間にわたり要求されたリソース量を満たしているかを判定する。オンライン処理向けの判定方法である。第2の判定方法は、空きリソース容量の平均が要求されたリソース量を満たしているかを判定する。この第2の判定方法では、一時的にリソース容量が上限に達してもよく、所定期間内(一周期以内)(たとえば、24時間以内)に処理が終了できればよいシステムに適した方法であり、バッチ処理向けの判定方法である。
これら2つの判定方法は、オペレータのニーズに応じて、いずれかを選択して行ってもよいし、両方行ってもよい。
同図に示すように、画面500は、仮想サーバ512(VM)と移動先物理サーバ514(PM)のペアのデータに対するCPU、メモリ、またはネットワークなどの余剰リソース516をリスト502として含む。リスト502において、リソース選択リスト518によってCPU、メモリ、またはネットワークの中からオペレータが操作部(不図示)を用いて選択したリソースの余剰リソース量の少ない順にソートすることができる。このリスト502をオペレータが参照することで、仮想サーバを移動したときの物理サーバの余剰リソース量を比較できる。
(付記1)
複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う判定手段を備え、
さらに、所定の処理が実行されている、複数の前記仮想サーバおよび複数の前記物理サーバの負荷データを前記通信網を介して収集する収集手段を備え、
前記判定手段は、前記収集手段が収集した前記負荷データに基づいて、前記判定する処理を行う管理装置。
(付記1)に記載の管理装置において、
前記単位時間内に処理される前記負荷データが所定数未満含まれるように前記単位時間の最小値を決定する決定手段をさらに備え、
前記判定手段は、決定された前記最小値に前記単位時間を分割し、前記最小値の前記単位時間で前記判定する処理を行う管理装置。
(付記1)または(付記2)に記載の管理装置において、
前記収集手段が収集した前記負荷データを、単位負荷の範囲に基づいて、グループに分類し、各前記グループから所定数の負荷データを選択する選択手段をさらに備え、
前記判定手段は、選択された前記負荷データに基づいて、前記判定する処理を行う管理装置。
(付記1)乃至(付記3)いずれかに記載の管理装置において、
前記管理装置の前記判定手段は、前記判定する処理において、前記空きリソース容量が、前記負荷データの負荷のピークが周期的に現れる所定期間にわたり、要求条件を満たしているか否かを判定する管理装置。
(付記1)乃至(付記4)いずれかに記載の管理装置において、
前記管理装置の前記判定手段は、前記判定する処理において、前記負荷データの負荷のピークが周期的に現れる所定期間内の前記空きリソース容量の平均値が、平均値の要求条件を満たしているか否かを判定する管理装置。
(付記1)乃至(付記5)いずれかに記載の管理装置において、
前記管理装置は、
前記負荷データに基づいて、前記物理サーバの余剰リソース量を算出する算出手段と、
算出された前記余剰リソース量をユーザに提示するリソース提示手段と、
をさらに備える管理装置。
(付記1)乃至(付記6)いずれかに記載の管理装置において、
前記管理装置の前記判定手段による判定結果をユーザに提示する判定結果提示手段をさらに備える管理装置。
複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置が、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行い、
さらに、所定の処理が実行されている、複数の前記仮想サーバおよび複数の前記物理サーバの負荷データを前記通信網を介して収集し、
収集した前記負荷データに基づいて、前記判定する処理を行うサーバ管理方法。
(付記8)に記載のサーバ管理方法において、
前記管理装置が、
前記単位時間内に処理される前記負荷データが所定数未満含まれるように前記単位時間の最小値を決定し、
決定された前記最小値に前記単位時間を分割し、前記最小値の前記単位時間で前記判定する処理を行うサーバ管理方法。
(付記8)または(付記9)に記載のサーバ管理方法において、
前記管理装置が、
収集した前記負荷データ、単位負荷の範囲に基づいて、グループに分類し、各前記グループから所定数の負荷データを選択し、
選択された前記負荷データに基づいて、前記判定する処理を行うサーバ管理方法。
(付記8)乃至(付記10)いずれかに記載のサーバ管理方法において、
前記管理装置が、
前記判定する処理において、前記空きリソース容量が、前記負荷データの負荷のピークが周期的に現れる所定期間にわたり、要求条件を満たしているか否かを判定するサーバ管理方法。
(付記8)乃至(付記11)いずれかに記載のサーバ管理方法において、
前記管理装置が、
前記判定する処理において、前記負荷データの負荷のピークが周期的に現れる所定期間内の前記空きリソース容量の平均値が、平均値の要求条件を満たしているか否かを判定するサーバ管理方法。
(付記8)乃至(付記12)いずれかに記載のサーバ管理方法において、
前記管理装置が、
前記負荷データに基づいて、前記物理サーバの余剰リソース量を算出し、
算出された前記余剰リソース量をユーザに提示するサーバ管理方法。
(付記8)乃至(付記13)いずれかに記載のサーバ管理方法において、
前記管理装置が、前記判定結果をユーザに提示するサーバ管理方法。
複数の仮想サーバが稼動する複数の物理サーバと、
複数の前記物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置と、を備え、
前記管理装置は、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う判定手段を備え、
前記管理装置の前記判定手段による判定結果をユーザに提示する判定結果提示手段をさらに備えるサーバシステム。
Claims (10)
- 複数の仮想サーバが稼動する複数の物理サーバと、
複数の前記物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置と、を備え、
前記管理装置は、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う判定手段を備えるサーバシステム。 - 請求項1に記載のサーバシステムにおいて、
前記管理装置は、
所定の処理が実行されている、複数の前記仮想サーバおよび複数の前記物理サーバの負荷データを前記通信網を介して収集する収集手段をさらに備え、
前記管理装置の前記判定手段は、前記収集手段が収集した前記負荷データに基づいて、前記判定する処理を行うサーバシステム。 - 請求項2に記載のサーバシステムにおいて、
前記管理装置は、
前記単位時間内に処理される前記負荷データが所定数未満含まれるように前記単位時間の最小値を決定する決定手段をさらに備え、
前記管理装置の前記判定手段は、決定された前記最小値に前記単位時間を分割し、前記最小値の前記単位時間で前記判定する処理を行うサーバシステム。 - 請求項2または3に記載のサーバシステムにおいて、
前記管理装置は、
前記収集手段が収集した前記負荷データを、単位負荷の範囲に基づいて、グループに分類し、各前記グループから所定数の負荷データを選択する選択手段をさらに備え、
前記管理装置の前記判定手段は、選択された前記負荷データに基づいて、前記判定する処理を行うサーバシステム。 - 請求項2乃至4いずれか一項に記載のサーバシステムにおいて、
前記管理装置の前記判定手段は、前記判定する処理において、前記空きリソース容量が、前記負荷データの負荷のピークが周期的に現れる所定期間にわたり、要求条件を満たしているか否かを判定するサーバシステム。 - 請求項2乃至5いずれか一項に記載のサーバシステムにおいて、
前記管理装置の前記判定手段は、前記判定する処理において、前記負荷データの負荷のピークが周期的に現れる所定期間内の前記空きリソース容量の平均値が、平均値の要求条件を満たしているか否かを判定するサーバシステム。 - 請求項2乃至6いずれか一項に記載のサーバシステムにおいて、
前記管理装置は、
前記負荷データに基づいて、前記物理サーバの余剰リソース量を算出する算出手段と、
算出された前記余剰リソース量をユーザに提示するリソース提示手段と、
をさらに備えるサーバシステム。 - 複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う判定手段を備える管理装置。 - 複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置が、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行うサーバ管理方法。 - 複数の仮想サーバが稼動する複数の物理サーバに通信網で接続され、前記物理サーバ上で稼働する前記仮想サーバを管理する管理装置を実現するコンピュータに、
前記物理サーバにおいて、前記仮想サーバを稼働させたときの空きリソース容量が所定の条件を満たすか否かを判定する処理を、所定の単位時間ごとに分割し、並列分散処理によって行う手順を実行させるためのプログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/877,949 US9319291B2 (en) | 2010-10-07 | 2011-09-09 | Server system, management device, server management method, and program |
JP2012537560A JP5870927B2 (ja) | 2010-10-07 | 2011-09-09 | サーバシステム、管理装置、サーバ管理方法、およびプログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2010227824 | 2010-10-07 | ||
JP2010-227824 | 2010-10-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012046386A1 true WO2012046386A1 (ja) | 2012-04-12 |
Family
ID=45927395
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2011/005093 WO2012046386A1 (ja) | 2010-10-07 | 2011-09-09 | サーバシステム、管理装置、サーバ管理方法、およびプログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US9319291B2 (ja) |
JP (1) | JP5870927B2 (ja) |
WO (1) | WO2012046386A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014122743A1 (ja) * | 2013-02-06 | 2014-08-14 | 株式会社日立製作所 | 仮想ゲスト管理システム及び仮想ゲスト管理方法 |
WO2015151290A1 (ja) * | 2014-04-04 | 2015-10-08 | 株式会社日立製作所 | 管理計算機と計算機制御方法及び計算機システム |
JP2016085687A (ja) * | 2014-10-28 | 2016-05-19 | 株式会社三菱東京Ufj銀行 | プログラムインストールコスト算出装置、プログラムインストール装置及び記録媒体 |
JP2016522947A (ja) * | 2013-05-21 | 2016-08-04 | アマゾン・テクノロジーズ・インコーポレーテッド | コンピュータリソースサービスの性能能力の決定及び監視 |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6718844B2 (ja) * | 2017-06-28 | 2020-07-08 | 日本電信電話株式会社 | 輻輳制御装置及び輻輳制御方法 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001101149A (ja) * | 1999-09-30 | 2001-04-13 | Nec Corp | 分散並列型データ処理装置及び分散並列型データ処理プログラムを記録した記録媒体並びに分散並列型データ処理システム |
JP2006228115A (ja) * | 2005-02-21 | 2006-08-31 | Ntt Comware Corp | 負荷分散予測装置及び負荷分散制御方法並びにそのプログラム、負荷分散制御システムおよび当該システムにおける負荷分散制御方法 |
JP2008217302A (ja) * | 2007-03-02 | 2008-09-18 | Nec Corp | 仮想マシンシステム、管理サーバ、仮想マシン移行方法及びプログラム |
JP2008276320A (ja) * | 2007-04-25 | 2008-11-13 | Nec Corp | 仮想システム制御方法およびコンピュータシステム |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3712654B2 (ja) * | 2000-10-30 | 2005-11-02 | 三菱重工業株式会社 | 気体状況予測方法 |
JP3861087B2 (ja) | 2003-10-08 | 2006-12-20 | 株式会社エヌ・ティ・ティ・データ | 仮想マシン管理装置及びプログラム |
US7836181B2 (en) * | 2004-09-21 | 2010-11-16 | Alcatel-Lucent Usa Inc. | Fairness and load balancing in wireless LANs using association control |
US8539075B2 (en) * | 2006-04-21 | 2013-09-17 | International Business Machines Corporation | On-demand global server load balancing system and method of use |
US20080082647A1 (en) * | 2006-09-29 | 2008-04-03 | Dennis Baker | Connection reservation in dynamic node allocation computing environments |
CN101212389B (zh) * | 2006-12-30 | 2010-06-23 | 华为技术有限公司 | 一种突发汇聚控制方法及相应的装置、通信设备 |
JP4980792B2 (ja) | 2007-05-22 | 2012-07-18 | 株式会社日立製作所 | 仮想計算機の性能監視方法及びその方法を用いた装置 |
US7801994B2 (en) | 2007-11-29 | 2010-09-21 | Hitachi, Ltd. | Method and apparatus for locating candidate data centers for application migration |
JP2010146420A (ja) * | 2008-12-22 | 2010-07-01 | Hitachi Ltd | 余剰資源管理システム、その管理方法、及びサーバ装置 |
-
2011
- 2011-09-09 WO PCT/JP2011/005093 patent/WO2012046386A1/ja active Application Filing
- 2011-09-09 JP JP2012537560A patent/JP5870927B2/ja not_active Expired - Fee Related
- 2011-09-09 US US13/877,949 patent/US9319291B2/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001101149A (ja) * | 1999-09-30 | 2001-04-13 | Nec Corp | 分散並列型データ処理装置及び分散並列型データ処理プログラムを記録した記録媒体並びに分散並列型データ処理システム |
JP2006228115A (ja) * | 2005-02-21 | 2006-08-31 | Ntt Comware Corp | 負荷分散予測装置及び負荷分散制御方法並びにそのプログラム、負荷分散制御システムおよび当該システムにおける負荷分散制御方法 |
JP2008217302A (ja) * | 2007-03-02 | 2008-09-18 | Nec Corp | 仮想マシンシステム、管理サーバ、仮想マシン移行方法及びプログラム |
JP2008276320A (ja) * | 2007-04-25 | 2008-11-13 | Nec Corp | 仮想システム制御方法およびコンピュータシステム |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014122743A1 (ja) * | 2013-02-06 | 2014-08-14 | 株式会社日立製作所 | 仮想ゲスト管理システム及び仮想ゲスト管理方法 |
US9762506B2 (en) | 2013-02-06 | 2017-09-12 | Hitachi, Ltd. | Virtual guest management system and virtual guest management method |
JP2016522947A (ja) * | 2013-05-21 | 2016-08-04 | アマゾン・テクノロジーズ・インコーポレーテッド | コンピュータリソースサービスの性能能力の決定及び監視 |
WO2015151290A1 (ja) * | 2014-04-04 | 2015-10-08 | 株式会社日立製作所 | 管理計算機と計算機制御方法及び計算機システム |
JP2016085687A (ja) * | 2014-10-28 | 2016-05-19 | 株式会社三菱東京Ufj銀行 | プログラムインストールコスト算出装置、プログラムインストール装置及び記録媒体 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2012046386A1 (ja) | 2014-02-24 |
US9319291B2 (en) | 2016-04-19 |
JP5870927B2 (ja) | 2016-03-01 |
US20130205024A1 (en) | 2013-08-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Usama et al. | Job schedulers for Big data processing in Hadoop environment: testing real-life schedulers using benchmark programs | |
Vasile et al. | Resource-aware hybrid scheduling algorithm in heterogeneous distributed computing | |
US11755452B2 (en) | Log data collection method based on log data generated by container in application container environment, log data collection device, storage medium, and log data collection system | |
US9852035B2 (en) | High availability dynamic restart priority calculator | |
Joseph et al. | IntMA: Dynamic interaction-aware resource allocation for containerized microservices in cloud environments | |
Xu et al. | Adaptive task scheduling strategy based on dynamic workload adjustment for heterogeneous Hadoop clusters | |
Bi et al. | SLA-based optimisation of virtualised resource for multi-tier web applications in cloud data centres | |
JP5870927B2 (ja) | サーバシステム、管理装置、サーバ管理方法、およびプログラム | |
WO2014045549A1 (ja) | マシン配置計画を生成する情報処理装置及びマシン配置計画生成方法 | |
Gill et al. | Tails in the cloud: a survey and taxonomy of straggler management within large-scale cloud data centres | |
CN109614227A (zh) | 任务资源调配方法、装置、电子设备及计算机可读介质 | |
Ashraf et al. | Distributed virtual machine consolidation: A systematic mapping study | |
Dias et al. | A systematic literature review on virtual machine consolidation | |
Babu et al. | Interference aware prediction mechanism for auto scaling in cloud | |
Deng et al. | Resource utilization analysis of Alibaba cloud | |
JP2011192187A (ja) | 管理装置、管理方法およびプログラム | |
Ouyang et al. | Mitigating stragglers to avoid QoS violation for time-critical applications through dynamic server blacklisting | |
Barletta et al. | Criticality-aware monitoring and orchestration for containerized industry 4.0 environments | |
Surya et al. | Prediction of resource contention in cloud using second order Markov model | |
Heidari et al. | A cost-efficient auto-scaling algorithm for large-scale graph processing in cloud environments with heterogeneous resources | |
JP6184116B2 (ja) | 移行アドバイザリー装置 | |
Long et al. | An improved topology schedule algorithm for storm system | |
He et al. | A Review of Resource Scheduling in Large-Scale Server Cluster | |
JP2021043893A (ja) | リソースの利用効率の最適化を支援するシステム及び方法 | |
Patel et al. | An efficient selection procedure with an enhanced load-balancing scheme for gridsim |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11830321 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2012537560 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13877949 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 11830321 Country of ref document: EP Kind code of ref document: A1 |