JP6281354B2 - Information processing system, information processing apparatus, information processing method, and program - Google Patents

Description

  The present invention relates to information processing, and more particularly to QoS (Quality of Service) of an information processing system including an information processing device and a storage device.

  The information processing system includes one or more information processing apparatuses that are subjects of information processing, and one or more storage apparatuses that store data to be processed.

  Further, the information processing apparatus operates one or a plurality of virtual machines (VMs) on the apparatus.

  Thus, in the information processing system, a plurality of information processing apparatuses and a plurality of VMs share one or a plurality of storage devices.

  Therefore, when some information processing apparatuses or VMs issue excessive input / output (IO) requests to the storage device, the response of the storage device is degraded. As a result, the performance of the information processing system decreases.

  In order to prevent such an influence on performance, in an information processing system including an information processing device and a storage device, a technique for controlling an IO request to the storage device is widely used (for example, Patent Document 1 and Patent). Reference 2).

  One such technique is QoS that controls the bandwidth of communication between an information processing device and a storage device (see, for example, Patent Document 3).

JP 2011-243117 A JP 2012-221340 A JP-T-2006-521640

  When the storage device implements the QoS function as in the technique described in Patent Document 3, the storage device performs a predetermined process on the logical disk that is the IO request issuance destination or the information processing device that is the IO request issuance source. Performance (for example, IOPS (Input Output per Second)) can be set. However, in an information processing system using a VM, a plurality of VMs operate on one information processing apparatus. A plurality of VMs operating on one information processing apparatus may access the same logical disk. In this case, the storage device cannot set a target value of performance (for example, IOPS) corresponding to individual VMs in a plurality of VMs operating on one information processing device.

  As described above, the technique described in Patent Document 3 has a problem that a QoS function corresponding to an individual VM cannot be realized.

  In the techniques described in Patent Document 1 and Patent Document 2, the information processing apparatus performs control.

  Therefore, a method for realizing the QoS function in the information processing apparatus can be considered.

  For example, when the information processing apparatus implements the QoS function, the information processing apparatus can apply the QoS function to a VM operating on the own apparatus. That is, the information processing apparatus controls the order of VM IO processing so as to approach the target performance (for example, IOPS) set for each VM. Based on such control, the information processing apparatus can suppress the IO request of the VM that issues an excessive IO request.

  However, the information processing apparatus does not grasp the state of the storage device. For this reason, the information processing apparatus may excessively suppress IO requests with respect to the performance of the storage device. Alternatively, the information processing apparatus may transmit an IO request that exceeds the performance of the storage device.

  As described above, the techniques described in Patent Document 1 and Patent Document 2 have a problem that IO requests to the storage device cannot be appropriately controlled.

  The objective of this invention is providing the information processing system, information processing apparatus, information processing method, and program which solve an above-described subject.

  An information processing system according to an embodiment of the present invention includes a VM that is software that emulates the operation of a computer, and a first QoS function execution that executes a QoS function of an IO request from the VM based on a performance target value of the VM Means, an IO transmission processing means for receiving an IO request from the first QoS function executing means and transmitting the IO request to a storage device, and an IO request sent from the VM to the first QoS function executing means And the number of IO transmissions sent from the first QoS function executing means to the IO transmission processing means, and a request IOPS that is the number of IO requests per unit time and an IO per unit time A first IO status measuring means for calculating a transmission IOPS that is the number of transmissions; a request IOPS and a transmission IOPS calculated by the first IO status measuring means; and the VM A first IO amount storage unit that stores a performance target value; a target performance coefficient storage unit that stores a target performance coefficient that is information used when calculating a performance target value of a VM; Allocation performance receiving means for receiving the performance target value, the stored VM request IOPS, transmission IOPS and performance target value, the target performance coefficient, and the performance of the logical disk previously received by the allocation performance receiving means An information processing apparatus including first allocation performance calculation means for calculating a VM performance target value based on the target value and the performance target value received this time, a logical disk for storing and reproducing information, and the information IO reception processing means for receiving an IO request from the processing device to the logical disk, and second for executing the QoS function of the logical disk based on the performance target value of the logical disk sent to the second QoS function execution unit from the IO reception processing unit to the logical disk, and an allocation performance transmission unit that transmits the performance target value of the logical disk to the information processing device. The number of IO requests and the number of IO processes sent to the logical disk by the second QoS function execution unit are measured, and the received IOPS and the number of processed IOs are the number of received IOs per unit time. A second IO status measuring means for calculating IOPS, a second IO quantity storage means for storing the received IOPS, the processing IOPS, and a performance target value of the logical disk, and a setting for the logical disk Performance target value storage means for storing performance target values, the accepted IOPS, processing IOPS, performance target values, and performance target values stored by the performance target value storage means And a storage device including second allocation performance calculation means for calculating a performance target value of the logical disk.

  An information processing apparatus according to an aspect of the present invention includes a VM that is software that emulates the operation of a computer, and a first QoS function execution that executes a QoS function of an IO request from the VM based on a performance target value of the VM Means, an IO transmission processing means for receiving an IO request from the first QoS function executing means and transmitting the IO request to a storage device, and an IO request sent from the VM to the first QoS function executing means And the number of IO transmissions sent from the first QoS function executing means to the IO transmission processing means, and a request IOPS that is the number of IO requests per unit time and an IO per unit time A first IO status measuring means for calculating a transmission IOPS that is the number of transmissions, a request IOPS and a transmission IOPS calculated by the first IO status measuring means, and the nature of the VM From the storage device, a first IO amount storage unit that stores a target value, a target performance coefficient storage unit that stores a target performance coefficient that is information used when calculating a performance target value of a VM, and a logical disk Allocation performance receiving means for receiving a performance target value, the requested IOPS, transmission IOPS and performance target value of the stored VM, the target performance coefficient, and the performance target of the logical disk previously received by the allocation performance receiving means First allocation performance calculation means for calculating the performance target value of the VM based on the value and the performance target value received this time.

  According to an information processing method of an information system including an information processing device and a storage device according to one aspect of the present invention, a VM that is software that emulates the operation of a computer is executed in the information processing device, and the performance target value of the VM is set. Based on the QoS function of the IO request from the VM, the IO request transmitted based on the QoS function is transmitted to the storage device, and the number of IO requests from the VM and the storage device are sent to the storage device. Measuring the number of IO transmissions, calculating a request IOPS that is the number of IO requests per unit time, and a transmission IOPS that is the number of IO transmissions per unit time, and calculating the calculated request IOPS and transmission IOPS, The VM performance target value is stored, a target performance coefficient that is information used when calculating the VM performance target value is stored, The target VM performance IOPS, the transmission IOPS, the performance target value, the target performance coefficient, the performance target value of the previously received logical disk, and the performance target value received this time Based on the above, the VM performance target value is calculated, the logical disk for storing and reproducing information is executed in the storage device, the IO request to the logical disk is received from the information processing device, and the logical disk The logical disk QoS function is executed based on the performance target value of the logical disk, the performance target value of the logical disk is transmitted to the information processing apparatus, and the IO request sent from the information processing apparatus to the logical disk is transmitted. And the number of IO processes in which the QoS function of the logical disk is executed, and the received IOPS, which is the number of received IOs per unit time, and the process, which is the number of processed IOs. IOPS is calculated, the received IOPS, the processing IOPS, and the performance target value of the logical disk are stored, the performance target value set in the logical disk is stored, the received IOPS, the processing IOPS, and the performance Based on the target value and the performance target value stored by the performance target value storage means, the performance target value of the logical disk is calculated.

  A program according to an aspect of the present invention includes a process for executing a VM, which is software for emulating the operation of a computer, as an information processing apparatus of an information system including an information processing apparatus and a storage device, and a performance target value of the VM A process of executing a QoS function of an IO request from a VM based on the process, a process of transmitting an IO request transmitted based on the QoS function to the storage device, the number of IO requests from the VM, and the storage A process of measuring the number of IO transmissions sent to the device and calculating a request IOPS that is the number of IO requests per unit time and a transmission IOPS that is the number of IO transmissions per unit time; A process for storing the requested IOPS and transmission IOPS, and the performance target value of the VM, and a target performance function that is information used when calculating the performance target value of the VM , A process of receiving a performance target value of a logical disk from the storage device, a request IOPS of a stored VM, a transmission IOPS, a performance target value, the target performance coefficient, and the last received A process of calculating a VM performance target value based on the performance target value of the logical disk and the performance target value received this time, and a process of executing a logical disk for storing and reproducing information as the storage device , A process of receiving an IO request to the logical disk from the information processing apparatus, a process of executing a QoS function of the logical disk based on the performance target value of the logical disk, and a performance target value of the logical disk, The processing to be transmitted to the information processing device, the number of IO requests sent from the information processing device to the logical disk, and the QoS function of the logical disk executed The number of processes is measured, and the process of calculating the received IOPS, which is the number of received IOs per unit time, and the process IOPS, which is the number of processed IOs, the received IOPS, the processed IOPS, and the performance of the logical disk A process for storing the target value, a process for storing the performance target value set in the logical disk, the received IOPS, the process IOPS, the performance target value, and the performance stored by the performance target value storage unit Based on the target value, the computer is caused to execute processing for calculating the performance target value of the logical disk.

  According to the present invention, it is possible to provide the effect of appropriately controlling the IO request from the information processing device to the storage device.

FIG. 1 is a block diagram showing an example of the configuration of the information processing system according to the first embodiment of the present invention. FIG. 2 is a block diagram illustrating an example of the configuration of the information processing apparatus according to the first embodiment. FIG. 3 is a diagram for explaining the operation of the first embodiment. FIG. 4 is a diagram for explaining the operation of the first embodiment. FIG. 5 is a block diagram illustrating an example of the configuration of the storage device according to the first embodiment. FIG. 6 is a sequence diagram illustrating an example of the operation of the first embodiment. FIG. 7 is a sequence diagram illustrating an example of the operation of the first embodiment. FIG. 8 is a sequence diagram illustrating an example of the operation of the first embodiment. FIG. 9 is a sequence diagram illustrating an example of the operation of the first embodiment. FIG. 10 is a block diagram illustrating an example of another configuration of the information processing apparatus according to the first embodiment. FIG. 11 is a block diagram showing an example of the configuration of the information processing apparatus according to the second embodiment of the present invention. FIG. 12 is a block diagram illustrating an example of the configuration of the storage device according to the second embodiment. FIG. 13 is a block diagram illustrating an example of a configuration of a storage device according to the third embodiment.

  Next, embodiments of the present invention will be described with reference to the drawings.

  Each drawing explains an embodiment of the present invention. Therefore, the present invention is not limited to the description of each drawing. Moreover, the same number is attached | subjected to the same structure of each drawing, and the repeated description may be abbreviate | omitted.

  Further, in the drawings used for the following description, the description of the configuration of the part not related to the description of the present invention is omitted, and there are cases where it is not illustrated.

  The performance in the embodiment of the present invention is not particularly limited. For example, the embodiment of the present invention may use throughput, TAT (Turn Around Time), or IOPS as performance. In the following description, IOPS (transmission IOPS from the information processing apparatus to the storage device) will be described as an example of performance. When other performance is used, the information processing apparatus and the storage device may replace IOPS with other performance in the following description.

  Further, in the following description, a VM is used as a target of the QoS function of the information processing apparatus. However, the QoS function target of the present invention need not be limited to the VM of the information processing apparatus. For example, the QoS function of the present invention may target an application that operates on the information processing apparatus.

  Next, an information processing system in the related art will be described before describing the embodiment of the present invention.

  In order to solve the problems of Patent Literature 1 to Patent Literature 3, the information processing system of the related technology operates the QoS function on the storage device side and the QoS function on the information processing device side simultaneously. That is, the QoS function on the information processing apparatus side realizes the IOPS target value for the VM. The QoS function on the storage device side realizes the IOPS target value between the storage device and the information processing device.

  However, the information processing system using the above configuration has the following problems.

  That is, the first problem is that the performance of the IO is reduced to a VM having no problem. The reason is as follows.

  When a single logical disk is used by a plurality of VMs, the QoS function of the related-art storage device suppresses IO evenly for all VMs using this logical disk. Therefore, the QoS function of the storage device is not limited to the VM in question (a VM with an excessive IO request), but also suppresses IO of other VMs that use the target logical disk. As a result, the response time of IO of this logical disk becomes long. That is, the performance of the information processing system deteriorates.

  The second problem is that a VM (problem VM) that continues to issue excessive IO requests cannot be sufficiently suppressed. The reason is as follows.

  Not all VMs always issue IO requests up to the IOPS target value. For this reason, the value obtained by summing the IOPS target values of each VM is often set to a value that exceeds the performance of the storage device. This is because the performance of the storage device can be exhibited as high as possible. However, in the case of this operation, the information processing system using the above configuration cannot sufficiently suppress IO for a VM in which the IO load on the storage device is excessive. As a result, the information processing apparatus of the related art information processing system transmits an excessive IO request to the storage device.

  The third problem is to rapidly suppress the IO request and make the operation of the VM unstable. The reason is as follows.

  A part of the IO request is suppressed according to the IOPS target value when the QoS function operates. Therefore, a VM with a large amount of IO requests has a slow IO response time. In the storage device, the status of the VM is unknown. Therefore, the QoS function of the storage device suppresses IO in consideration of the performance of the storage device. As a result, the information processing system related to the present invention excessively suppresses the IO performance (IOPS) viewed from the VM of the information processing apparatus.

  Therefore, the information processing system according to the embodiment of the present invention solves the problems of the related techniques in addition to solving the problems in Patent Documents 1 to 3 as described below.

  Next, a first embodiment of the present invention will be described with reference to the drawings.

<First Embodiment>
As shown in FIG. 1, the information processing system 10 according to the first embodiment of the present invention includes one or more information processing devices 100, one or more storage devices 200, and a network 300. .

  The information processing apparatus 100 executes processing using information stored in the storage device 200. Therefore, the information processing apparatus 100 transmits an IO request necessary for processing to the storage device 200.

  The storage device 200 receives an IO request from the information processing device 100 and executes processing corresponding to the IO request.

  The network 300 is a communication network that connects the information processing apparatus 100 and the storage device 200. Network 300 is not particularly limited as long as it is a general communication network. For example, the network 300 may be the Internet or a telephone communication network. Therefore, detailed description of the network 300 is omitted.

  Next, configurations of the information processing apparatus 100 and the storage device 200 in the present embodiment will be described with reference to the drawings.

  First, the configuration of the information processing apparatus 100 will be described.

  FIG. 2 is a block diagram illustrating an example of the configuration of the information processing apparatus 100 according to the first embodiment.

  The information processing apparatus 100 includes a QoS function execution unit 110, an allocation performance calculation unit 121, an allocation performance reception unit 122, an IO amount storage unit 123, an IO status measurement unit 124, a target performance coefficient storage unit 125, and a VM 141. And an IO transmission processing unit 142. Here, the information processing apparatus 100 may include one or a plurality of VMs 141.

  The VM 141 is software that emulates the operation of a computer on the information processing apparatus 100.

  The QoS function execution unit 110 holds a performance target value (hereinafter referred to as “IOPS target value”) for each VM 141 calculated by the allocation performance calculation unit 121 described later. Then, the QoS function execution unit 110 receives an IO request from the VM 141. Furthermore, the QoS function execution unit 110 holds a performance history for each VM 141 (hereinafter referred to as “IOPS history”). Then, the QoS function execution unit 110 compares the IOPS history with the IOPS target value. Then, the QoS function execution unit 110 preferentially processes an IO request from the VM 141 with a low degree of achievement of the IOPS target value. That is, the QoS function execution unit 110 realizes the QoS function in the information processing apparatus 100. In this way, the QoS function execution unit 110 transmits the IO request to be processed to the IO transmission processing unit 142.

  The IO transmission processing unit 142 transmits the IO request from the QoS function execution unit 110 to the storage device 200.

  The target performance coefficient storage unit 125 is information used when the allocation performance calculation unit 121, which will be described later, calculates (determines) an IOPS target value for each VM 141 (for example, a coefficient of a calculation formula; hereinafter referred to as a target performance coefficient). Save.

  The assigned performance receiving unit 122 receives the setting of the performance target value (IOPS target value) for each logical disk 241 described later from the storage device 200. The allocation performance reception unit 122 transmits the received IOPS target value to the allocation performance calculation unit 121.

  The IO status measurement unit 124 counts the number of IO requests sent from the VM 141 to the QoS function execution unit 110 (hereinafter referred to as “requested IO count”) and the IO sent from the QoS function execution unit 110 to the IO transmission processing unit 142. (Hereinafter referred to as “the number of transmission IOs”). Then, the IO status measurement unit 124 divides the requested IO count by the measurement time, and calculates the requested IO count per unit time (hereinafter referred to as “request IOPS”). Similarly, the IO status measurement unit 124 divides the number of transmission IOs by the measurement time, and calculates the number of transmission IOs per unit time (hereinafter referred to as “transmission IOPS”). Then, the IO status measurement unit 124 stores the calculation results (request IOPS and transmission IOPS) in the IO amount storage unit 232.

  The IO amount storage unit 123 stores the performance per unit time (requested IOPS and transmission IOPS) calculated by the IO status measurement unit 124 and the performance target value (IOPS target value) calculated by the allocated performance calculation unit 121.

  The allocation performance calculation unit 121 calculates an IOPS target value (performance target value) for each VM 141.

  Next, calculation of the allocation performance calculation unit 121 will be described.

  Here, the allocation performance calculation unit 121 operates when the allocation performance reception unit 122 receives an IOPS target value for each logical disk 241 from the storage device 200.

  More specifically, the allocation performance calculation unit 121 calculates an IOPS target value for each VM 141 based on, for example, the following information. The basic information is as follows.

(1) Requested IOPS for each VM 141 stored in the IO amount storage unit 123, transmission IOPS, and IOPS target value (This IOPS target value is the IOPS target value calculated last time.)
(2) Information stored in the target performance coefficient storage unit 125 (target performance coefficient)
(3) IOPS target value for each logical disk 241 received by the allocation performance receiver 122 this time (4) IOPS target value for each logical disk 241 received by the allocation performance receiver 122 last time IOPS for each VM 141 in the allocation performance calculator 121 The target value calculation method is, for example, as follows.

  Here, in the following description, all VMs 141 that use the logical disk 241 to be calculated are represented as “VM ^ LD”.

  When the QoS function execution unit 110 does not hold the IOPS target value for each VM 141, the allocation performance calculation unit 121 determines from the target performance coefficient storage unit 125 all VMs 141 (all VMs 141 belonging to “VM ^ LD”). Get target performance coefficient. Then, the allocated performance calculation unit 121 calculates (determines) the IOPS target value for each VM 141 based on the product of the IOPS target value for each logical disk 241 and the target performance coefficient. That is, the target performance coefficient is a value corresponding to the ratio of performance assigned to the VM 141 related to the logical disk 241.

  The allocation performance calculation unit 121 notifies the QoS function execution unit 110 of the calculated IOPS target value. Furthermore, the allocation performance calculation unit 121 transmits the calculated IOPS target value to the IO amount storage unit 123.

  On the other hand, when the QoS function execution unit 110 holds the IOPS target value for each VM 141, the allocation performance calculation unit 121 sends the IOPS target value for every VM 141 belonging to “VM ^ LD” from the IO amount storage unit 123. And request IOPS and transmission IOPS. When there is a logical disk 241 whose IOPS target value has changed, the allocation performance calculation unit 121 calculates a new IOPS target value for the VM 141.

  For example, when there is a logical disk 241 whose IOPS target value is smaller than the previous value, that is, a logical disk 241 that needs to narrow down the IO request, the allocation performance calculation unit 121 calculates a new IOPS target value for the VM 141. Hereinafter, the absolute value of the difference between the IOPS target values in the logical disk 241 is “dif”.

  This calculation will be described with reference to the drawings.

  FIG. 3 is a diagram schematically showing the IOPS for each VM 141 related to the logical disk 241 that needs to narrow down the IO request used in this explanation.

  FIG. 3 shows a transmission IPOS of five VMs 141 (VM1 to VM5) as an example. That is, “VM ^ LD” includes “VM1, VM2, VM3, VM4, VM5”.

  In FIG. 3, the IOPS target value for each new VM 141 is a newly set IOPS target value (hereinafter referred to as “newd” in the description of FIG. 3).

  First, the allocation performance calculation unit 121 calculates the absolute value (dif) of the difference from the previous time in the logical disk 241. In this case, dif is the total amount of transmission IOPS to be reduced. Then, the allocation performance calculation unit 121 selects newd as the IOPS target value (newd) for each new VM 141 so that the total number of transmission IOPS exceeding newd is equal to dif. That is, the allocation performance calculation unit 121 selects the value of new so that the sum of the hatched portion (the portion exceeding new) in FIG. 3 is equal to dif. Then, the allocation performance calculation unit 121 calculates new IOPS target values (IOPS_newd) for all VMs 141 (in FIG. 3, VM1, VM2, and VM3) that belong to “VM ^ LD” that exceeds newd. Then, the allocation performance calculation unit 121 transmits the calculated new IOPS target value to the QoS function execution unit 110 and the IO amount storage unit 123.

  Further, when there is a logical disk 241 whose IOPS target value is larger than the previous time, that is, a logical disk 241 to which an IO request can be additionally issued (hereinafter, the absolute value of the difference in this case is also referred to as “dif”), allocation performance calculation The unit 121 calculates a new IOPS target value.

  This calculation will be described with reference to the drawings.

  FIG. 4 is a diagram schematically showing request IOPS for each VM 141 related to the logical disk 241 to which additional IO requests can be issued.

  FIG. 4 shows five VMs 141 (VM1 to VM5) as an example. That is, “VM ^ LD” includes “VM1, VM2, VM3, VM4, VM5”.

  In FIG. 4, the previous IOPS target value is the IOPS target value that has already been set. The IOPS target value for each new VM 141 is a newly set IOPS target value (hereinafter referred to as “newd” in the description of FIG. 4).

  First, the allocation performance calculation unit 121 calculates the absolute value (dif) of the difference from the previous time in the IOPS target value of the logical disk 241. In this case, dif is the total amount of transmission IOPS that can be added. Then, the allocation performance calculation unit 121 selects newd as the IOPS target value (newd) for each new VM 141 so that the total of requested IOPS exceeding the previous IOPS target value is equal to dif. That is, the allocation performance calculation unit 121 selects new so that the sum of the hatched portions in FIG. 4 is equal to dif. This hatched portion is a portion to be added. Then, the allocation performance calculation unit 121 calculates new IOPS target values for all the VMs 141 (in FIG. 4, VM1, VM2, and VM3) that belong to “VM ^ LD” that can be added. Here, the allocation performance calculation unit 121 sets the smaller one of the required IOPS and newd. Here, the horizontal line portion of VM1 in FIG. 4 is a request IOPS that cannot be added.

  When the total of requested IOPS exceeding the previous IOPS target value (hereinafter referred to as “sumd”) is smaller than dif, the allocation performance calculation unit 121 may recalculate the IOPS target value for each VM 141. In this case, for example, there is no horizontal line portion in FIG.

  For example, the allocation performance calculation unit 121 may acquire the target performance coefficient from the target performance coefficient storage unit 125 and set the IOPS target value using the following formula 1.

[Formula 1]
IOPS target value = (target performance coefficient) × (dif−sumd)
+ (Maximum request IOPS of all VMs 141 belonging to “VM ^ LD”)
The allocation performance calculation unit 121 calculates an IOPS target value for each VM 141 based on the above processing, and transmits the IOPS target value to the QoS function execution unit 110 and the IO amount storage unit 123.

  Next, the configuration of the storage device 200 will be described.

  FIG. 5 is a block diagram illustrating an example of the configuration of the storage device 200 according to the first embodiment.

  The storage device 200 includes a QoS function execution unit 210, an allocation performance transmission unit 221, a performance target value storage unit 222, an allocation performance calculation unit 231, an IO amount storage unit 232, an IO status measurement unit 233, a logical disk 241 and an IO reception processing unit 242. Here, the storage device 200 may include one or a plurality of logical disks 241.

  The logical disk 241 is a logical storage / reproduction device that stores and reproduces information.

  The IO reception processing unit 242 receives an IO request to the logical disk 241 from the information processing apparatus 100 and transmits it to the QoS function execution unit 210.

  The performance target value storage unit 222 stores a performance target value (IOPS target value) set for each logical disk 241. The performance target value stored by the performance target value storage unit 222 is a target setting value that is initially set in the logical disk 241. Alternatively, the performance target value stored by the performance target value storage unit 222 is a value used at the time of resetting. That is, the performance target value stored by the performance target value storage unit 222 is the initial performance target value in the logical disk 241.

  The QoS function execution unit 210 holds a performance target value (hereinafter referred to as “IOPS target value”) for each logical disk 241. The QoS function execution unit 210 receives an IO request to the logical disk 241 from the IO reception processing unit 242. Furthermore, the QoS function execution unit 210 holds an IOPS history for each logical disk 241. Then, the QoS function execution unit 210 compares the IOPS history with the IOPS target value, and preferentially processes an IO request to the logical disk 241 with a low degree of achievement of the IOPS target value. That is, the QoS function execution unit 210 realizes the QoS function in the storage device 200. As described above, the QoS function execution unit 210 passes the IO request to the logical disk 241.

  The allocation performance transmission unit 221 transmits the IOPS target value for each logical disk 241 calculated by the allocation performance calculation unit 231 to the allocation performance reception unit 122 of the information processing apparatus 100. The allocation performance transmission unit 221 may transmit the IOPS target value for each logical disk 241 to the QoS function execution unit 210. However, the allocation performance calculation unit 231 may transmit the IOPS target value to the QoS function execution unit 210. Further, before the allocation performance calculation unit 231 calculates the IOPS target value such as an initial state, the allocation performance transmission unit 221 uses the information processing apparatus 100 and the QoS function execution unit to determine the performance target value stored by the performance target value storage unit 222. You may transmit to 210.

  The IO status measurement unit 233 measures the number of IO requests sent from the IO reception processing unit 242 to the QoS function execution unit 210 for each logical disk 241 to which an IO request is requested (hereinafter referred to as “accepted IO count”). To do. Further, the IO status measurement unit 233 measures the number of IO processes sent to the logical disk 241 by the QoS function execution unit 210 (hereinafter referred to as “process IO count”). Then, the IO status measurement unit 233 calculates the number of accepted IOs and the number of processed IOs per unit time (hereinafter referred to as “accepted IOPS” and “processed IOPS”, respectively). Then, the IO status measurement unit 233 stores the accepted IOPS and the processed IOPS in the IO amount storage unit 232.

  The IO amount storage unit 232 stores, for each logical disk 241, the received IOPS calculated (measured) by the IO status measurement unit 233, the processing IOPS, and the IOPS target value of the logical disk 241 calculated by the allocation performance calculation unit 231. To do.

  The allocated performance calculation unit 231 calculates a performance target value (IOPS target value) for each logical disk 241.

  Next, the calculation of the allocation performance calculation unit 231 will be further described.

  The allocation performance calculator 231 calculates the IOPS target value of the logical disk 241 based on the following information. The information on which the calculation is based is as follows.

(1) Accepted IOPS, processing IOPS, and IOPS target value for each logical disk 241 stored in the IO amount storage unit 232 (this IOPS target value is the IOPS target value calculated last time).
(2) IOPS target value stored in performance target value storage unit 222 (this IOPS target value is an initial target value)
The allocation performance calculation unit 231 stores the calculated IOPS target value for each logical disk 241 in the IO amount storage unit 232.

  For example, the allocation performance calculation unit 231 calculates the IOPS target value for each logical disk 241 as follows.

  When the QoS function execution unit 210 does not hold the IOPS target value of the logical disk 241, the allocation performance calculation unit 231 uses the IOPS target value stored in the performance target value storage unit 222 as the IOPS target value. When the QoS function execution unit 210 does not hold the IOPS target value of the logical disk 241, the allocation performance transmission unit 221 may transmit the IOPS target value.

  On the other hand, when the QoS function execution unit 210 holds the IOPS target value for each logical disk 241, the allocation performance calculation unit 231 compares the processing IOPS stored in the IO amount storage unit 232 with the accepted IOPS.

  When there is at least one logical disk 241 whose accepted IOPS exceeds the processing IOPS, that is, when IO processing is suppressed, the allocation performance calculation unit 231 has a margin for the performance of the logical disk 241. Judge that there is no. Then, the allocation performance calculation unit 231 sets the processing IOPS of the logical disk 241 having the largest processing IOPS as a new IOPS target value for each logical disk 241.

  On the other hand, when the processing IOPS of all the logical disks 241 is equal to or lower than the accepted IOPS, that is, when there is a margin in the performance of the logical disk 241, the allocation performance calculation unit 231 determines the performance of the storage device 200. Judge that there is room. Then, the assigned performance calculation unit 231 sets the larger one of the IOPS target value and the processing IOPS stored in the performance target value storage unit 222 as the IOPS target value for each new logical disk 241.

  Next, operations of the information processing apparatus 100 and the storage device 200 will be described with reference to the drawings.

  6 to 9 are sequence diagrams showing an example of the operation of the present embodiment.

  First, the allocated performance transmission unit 221 acquires an IOPS target value for each logical disk 241 from the performance target value storage unit 222 (B101).

  The allocation performance transmission unit 221 transmits (notifies) the acquired IOPS target value as the initial IOPS target value of the logical disk 241 to the QoS function execution unit 210 (B102).

  Furthermore, the allocation performance transmission unit 221 transmits the acquired IOPS target value to the allocation performance reception unit 122 of the information processing apparatus 100 (B103).

  The allocation performance receiving unit 122 sends the received IOPS target value of the logical disk 241 to the allocation performance calculation unit 121.

  The assigned performance calculation unit 121 acquires the target performance coefficient from the target performance coefficient storage unit 125 (A104).

  Then, the allocated performance calculation unit 121 calculates a performance target value (IOPS target value) for each VM 141 based on the IOPS target value of the logical disk 241 and the target performance coefficient (A105).

  Next, the allocation performance calculation unit 121 transmits the calculated IOPS target value for each VM 141 to the QoS function execution unit 110 (A106).

  The QoS function execution unit 110 sends an IO request from the VM 141 to the IO reception processing unit 242 of the storage device 200 via the IO transmission processing unit 142 so that the IOPS target value (performance target value) for each VM 141 is not exceeded. Transmit (A107).

  Further, the IO status measurement unit 124 counts the number of IO requests (requested IO count) and the number of IO transmissions (transmission IO count) in the QoS function execution unit 110 for each VM 141. Then, the IO status measurement unit 124 calculates the number of times per unit time (performance: request IOPS and transmission IOPS) for the request IO count and the transmission IO count (A108).

  Further, the IO status measurement unit 124 stores the calculated request IOPS and transmission IOPS in the IO amount storage unit 123 (A109).

  On the other hand, the IO reception processing unit 242 of the storage device 200 receives the IO request transmitted from the IO transmission processing unit 142 (B110).

  Then, the QoS function execution unit 210 executes priority processing of the received IO request based on the IOPS target value that is held (set) and the degree of achievement of the requested IOPS. Then, the QoS function execution unit 210 transmits an IO request to the logical disk 241 (B111).

  Further, the IO status measurement unit 233 counts the number of times of receiving an IO request and the number of IO processes in the QoS function execution unit 210 for each logical disk 241. Then, the IO status measurement unit 233 calculates the number of times per unit time (request IOPS and transmission IOPS) (B112).

  The IO status measurement unit 233 stores the calculated request IOPS and transmission IOPS in the IO amount storage unit 232 (B113).

  The allocated performance calculation unit 231 acquires the requested IOPS, transmission IOPS, and IOPS target value (performance target value) for each stored logical disk 241 periodically or at a predetermined timing. Then, the allocation performance calculation unit 231 recalculates the IOPS target value for each logical disk 241 based on the acquired value (B114). Note that the assigned performance calculation unit 231 may use the performance target value stored by the performance target value storage unit 222 as necessary.

  Then, the allocation performance calculation unit 231 transmits the recalculated IOPS target value for each logical disk 241 to the allocation performance transmission unit 221 and stores it in the IO amount storage unit 232 (B115).

  Then, the allocated performance transmission unit 221 transmits the received IOPS target value of the logical disk 241 to the QoS function execution unit 210 (B116).

  Further, the allocation performance transmission unit 221 transmits (notifies) the IOPS target value of the logical disk 241 to the allocation performance reception unit 122 (B117).

  The allocation performance receiving unit 122 transmits the received IOPS target value for each logical disk 241 to the allocation performance calculation unit 121.

  The allocation performance calculation unit 121 recalculates the IOPS target value for each VM 141 based on the following information (A118). The basic information is as follows.

(1) Received IOPS target value of logical disk 241 (2) Received IOPS target value of logical disk 241 received last time (3) Requested IOPS, transmission IOPS, and IOPS target value for each VM 141 stored in the IO amount storage unit 123 (4) Target Performance Coefficient The allocation performance calculation unit 121 transmits the calculated IOPS target value for each VM 141 to the QoS function execution unit 110 (A119). Moreover, the allocation performance calculation unit 121 stores the IOPS target value in the IO amount storage unit 123.

  The information processing apparatus 100 returns to A107. Then, the QoS function execution unit 110 transmits an IO request to the storage device 200 so as not to exceed the IOPS target value. Then, the storage device 200 returns to the operation after receiving the B110 IO request.

  The information processing device 100 and the storage device 200 repeat the operations of A107-A119 and B110-B117.

  As described above, the storage device 200 calculates the IOPS target value of the logical disk 241 based on the accepted IOPS and the processing IOPS. Then, the storage device 200 implements the QoS function of the logical disk 241 based on the calculated IOPS target value. Further, the storage device 200 transmits the calculated IOPS target value for each logical disk 241 to the information processing device 100.

  The information processing apparatus 100 calculates the IOPS target value for each VM 141 using the IOPS target value for each logical disk 241 and executes the QoS function.

  That is, the information processing apparatus 100 executes the QoS function of the VM 141 based on the status of the storage device 200.

  Next, the effect of this embodiment will be described.

  The information processing apparatus 100 of the information processing system 10 according to the first embodiment of this invention can obtain the effect that the IO performance is not deteriorated in the VM 141 other than the VM 141 having an excessive load.

  The reason is as follows.

  The storage device 200 transmits the IOPS target value of the logical disk 241 to the information processing device 100. Here, the transmitted IOPS target value is a performance value that can be processed by the logical disk 241.

  The allocation performance receiving unit 122 of the information processing apparatus 100 receives the IOPS target value (the IOPS value that can be processed by the storage device 200) of the logical disk 241 from the storage device 200. The allocation performance calculation unit 121 calculates the IOPS target value of the VM 141 in consideration of the IOPS target value of the logical disk 241. Therefore, the QoS function execution unit 110 of the information processing apparatus 100 suppresses the IO request of the VM 141 that has no problem to the extent that the storage device 200 can process. As described above, the information processing apparatus 100 controls the IO request of the VM 141 within a range in which the storage device 200 can maintain the performance of the IO processing.

  In addition, the information processing apparatus 100 of the information processing system 10 according to the first embodiment is capable of selectively suppressing the problematic VM 141.

  The reason is as follows.

  The IO status measurement unit 124 of the information processing apparatus 100 monitors the status of the IO request of the VM 141 in the QoS function execution unit 110. Then, the allocation performance calculation unit 121 calculates the IOPS target value so as to selectively suppress the IO request of the VM 141 having a heavy load, that is, a problem. This is because the QoS function execution unit 110 realizes the QoS function based on the calculation result of the allocation performance calculation unit 121.

  As described above, the information processing system 10 including the information processing apparatus 100 and the storage device 200 according to the present embodiment can appropriately control the IO request to the storage device 200.

(Modification)
The information processing apparatus 100 described above is configured as follows.

  For example, each component of the information processing apparatus 100 may be configured with a hardware circuit.

  Further, the information processing apparatus 100 may be configured as a plurality of information processing apparatuses in which the respective components are connected via a network or a bus.

  Further, the information processing apparatus 100 may configure a plurality of components by a single piece of hardware.

  The information processing apparatus 100 may be realized as a computer apparatus including a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). In addition to the above configuration, the information processing apparatus 100 may be realized as a computer apparatus that further includes an input / output connection circuit (IOC) and a network interface circuit (NIC).

  FIG. 10 is a block diagram illustrating an example of the configuration of the information processing apparatus 600 according to the modification.

  The information processing apparatus 600 includes a CPU 610, a ROM 620, a RAM 630, an internal storage device 640, an IOC 650, and a NIC 680, and constitutes a computer.

  The CPU 610 reads a program from the ROM 620. The CPU 610 controls the RAM 630, the internal storage device 640, the IOC 650, and the NIC 680 based on the read program. The computer including the CPU 610 controls these configurations, and implements the functions as the information processing apparatus 100 illustrated in FIG. Here, each function is a function of the QoS function execution unit 110, the allocation performance calculation unit 121, the allocation performance reception unit 122, the IO status measurement unit 124, the IO transmission processing unit 142, and the VM 141. The CPU 610 may use the RAM 630 or the internal storage device 640 as a temporary storage of a program when realizing each function.

  Further, the CPU 610 may read a program included in the computer-readable storage medium 700 storing the program using a storage medium reading device (not shown). Alternatively, the CPU 610 may receive a program from an external device (not shown) via the NIC 680. Further, the CPU 610 may store the read program or the received program in the RAM 630 and operate based on the stored program.

  The ROM 620 stores programs executed by the CPU 610 and fixed data. The ROM 620 is, for example, a P-ROM (Programmable-ROM) or a flash ROM.

  The RAM 630 temporarily stores programs executed by the CPU 610 and data. The RAM 630 is, for example, a D-RAM (Dynamic-RAM).

  The internal storage device 640 stores data and programs that the information processing device 600 saves over a long period of time. The internal storage device 640 operates as an IO amount storage unit 123 and a target performance coefficient storage unit 125. Further, the internal storage device 640 may operate as a temporary storage device for the CPU 610. The internal storage device 640 is, for example, a hard disk device, a magneto-optical disk device, an SSD (Solid State Drive), or a disk array device.

  Here, the ROM 620 and the internal storage device 640 are nonvolatile storage media. On the other hand, the RAM 630 is a volatile storage medium. The CPU 610 can operate based on a program stored in the ROM 620, the internal storage device 640, or the RAM 630. That is, the CPU 610 can operate using a nonvolatile storage medium or a volatile storage medium.

  The IOC 650 mediates data between the CPU 610, the input device 660, and the display device 670. The IOC 650 is, for example, an IO interface card or a USB (Universal Serial Bus) card.

  The input device 660 is a device that receives an input instruction from an operator of the information processing apparatus 600. The input device 660 is, for example, a keyboard, a mouse, or a touch panel.

  The display device 670 is a device that displays information to the operator of the information processing apparatus 600. The display device 670 is a liquid crystal display, for example.

  The NIC 680 relays data exchange with an external device (not shown) via the network. The NIC 680 may realize the function as the IO transmission processing unit 142. The NIC 680 is, for example, a LAN (Local Area Network) card.

  The information processing apparatus 600 configured as described above can obtain the same effects as the information processing apparatus 100.

  This is because the CPU 610 of the information processing apparatus 600 can realize the same function as the information processing apparatus 100 based on the program.

  Further, the storage device 200 can be configured based on a hardware circuit, similarly to the information processing device 100.

  For example, each component of the storage device 200 may be configured with a hardware circuit.

  In addition, the storage device 200 may be configured as a plurality of information processing devices in which each component is connected via a network or a bus.

  In addition, the storage device 200 may be configured with a plurality of components by a single piece of hardware.

  Further, the storage device 200 may be realized as a computer device similar to the information processing device 600 illustrated in FIG. That is, the storage device 200 may be configured using a computer including the CPU 610, the ROM 620, the RAM 630, the internal storage device 640, the IOC 650, and the NIC 680. In this case, the computer that configures the storage device 200 controls these configurations and implements the functions of the storage device 200 illustrated in FIG. Here, each function is a function of the QoS function execution unit 210, the allocation performance transmission unit 221, the allocation performance calculation unit 231, the IO status measurement unit 233, the logical disk 241, and the IO reception processing unit 242. is there. The internal storage device 640 included in the storage device 200 realizes the functions of the performance target value storage unit 222, the IO amount storage unit 232, and the logical disk 241.

  The information processing apparatus 600 configured as described above can obtain the same effects as the storage apparatus 200.

  This is because the CPU 610 of the information processing device 600 can realize the same function as the storage device 200 based on the program.

<Second Embodiment>
Next, an information processing system 10 according to a second embodiment of the present invention will be described. The information processing system 10 according to the present embodiment includes one or more information processing apparatuses 101 and one or more storage apparatuses 201 as in the first embodiment.

  Next, configurations of the information processing apparatus 101 and the storage device 201 in the present embodiment will be described with reference to the drawings.

  FIG. 11 is a block diagram illustrating an example of the configuration of the information processing apparatus 101 according to the second embodiment. As illustrated in FIG. 11, the information processing apparatus 101 according to the present embodiment includes an IO status history storage unit 126 and an IO suppression level determination unit 127 in addition to the configuration of the information processing apparatus 100 according to the first embodiment.

  FIG. 12 is a block diagram illustrating an example of the configuration of the storage device 201 according to the second embodiment. As illustrated in FIG. 12, the storage device 201 of this embodiment includes an IO status history storage unit 215 and an IO suppression level determination unit 216 in addition to the configuration of the storage device 200 of the first embodiment.

  The IO status history storage unit 126 stores the history of the requested IOPS, transmission IOPS, and IOPS target value for each VM 141 measured by the IO status measurement unit 124.

  Based on the history of the request IOPS, transmission IOPS, and IOPS target value for each VM 141 stored by the IO status history storage unit 126, the IO suppression level determination unit 127 sends the QoS function execution unit 110 the degree of suppression for each VM 141 ( Degree, level). The IO suppression level determination unit 127 then instructs the QoS function execution unit 110 on the determined suppression level for each VM 141. That is, the IO suppression level determination unit 127 determines the level of the VM 141 to be suppressed in the QoS function execution unit 110 based on the performance history.

  Here, the information processing apparatus 101 may set the degree of suppression (level) when the IOPS target value is exceeded according to the information processing system 10. For example, the information processing apparatus 101 may determine the degree of suppression as follows.

  The IO suppression level determination unit 127 confirms a history of a predetermined interval (cycle). For example, if the time of one history is 10 seconds and the number of history to be used is 10, the IO suppression level determination unit 127 confirms the history for 100 seconds. Here, when the request IOPS is equal to the transmission IOPS, the IO request is not suppressed. Therefore, the IO suppression level determination unit 127 instructs the QoS function execution unit 110 to permit IO transmission (IO request) exceeding the IOPS target value within a predetermined range in the next cycle. (However, it is desirable to lower the priority of the IO request that exceeds the IOPS target value.) Based on the instruction, the QoS function execution unit 110 determines that the VM 141 whose transmission IOPS does not exceed the IOPS target value. If there is no IO request from the VM 141, the held IO request of the VM 141 (waiting for processing) is transmitted without being suppressed.

  Thus, the IO suppression level determination unit 127 determines the level of suppression based on the amount of IO requests generated within a predetermined time. Therefore, the IO suppression level determination unit 127 does not suppress the VM 141 for which the IO request has temporarily increased.

  On the other hand, even if the VM 141 constantly has many IO requests, the IO requests are suppressed even if the IO suppression level determination unit 127 uses the history. That is, the IO suppression level determination unit 127 constantly suppresses the VM 141 that has many IO requests.

  Further, the IO suppression level determination unit 127 uses a history. Therefore, the IO suppression level determination unit 127 applies the suppression of the VM 141 gradually and applies the release of the suppression gradually. That is, in the present embodiment, the suppression and cancellation of the VM 141 are gradually decreased and gradually increased. As a result, the operation of the VM 141 is stabilized.

  The IO suppression level determination unit 127 may determine the degree of suppression (level) based on the number of times the IO request has been suppressed in the past history.

  Alternatively, the IO suppression level determination unit 127 may set the maximum IOPS within a range of values between the past transmission IOPS and the IOPS target value. Then, the IO suppression level determination unit 127 may instruct the QoS function execution unit 110 not to transmit an IO request exceeding the set maximum IOPS even when there is no IO request of another VM 141.

  The IO status history storage unit 215 stores the history of received IOPS, processing IOPS, and IOPS target value for each logical disk 241 measured by the IO status measurement unit 233.

  The IO suppression level determination unit 216 performs the QoS function execution unit 210 for each logical disk 241 based on the history of received IOPS, processing IOPS, and IOPS target value for each logical disk 241 stored by the IO status history storage unit 215. The degree of suppression (degree, level) is determined. Then, the IO suppression level determination unit 216 instructs the QoS function execution unit 210 on the determined degree of suppression. That is, when any logical disk 241 exceeds the IOPS target value, the IO suppression level determination unit 216 instructs the QoS function execution unit 210 to suppress the logical disk 241 that exceeds the IOPS target value.

  The IO suppression level determination unit 216 may determine suppression based on the same method as the IO suppression level determination unit 127. Therefore, detailed description of suppression by the IO suppression level determination unit 216 is omitted.

  As described above, the IO suppression level determination unit 216 operates in the same manner as the IO suppression level determination unit 127, so that the operation of the logical disk 241 is also stabilized.

  Note that the information processing apparatus 101 and the storage apparatus 201 may be configured using a computer illustrated in FIG.

  Next, the effect of this embodiment will be described.

  The information processing apparatus 101 of the information processing system 10 according to the second embodiment can obtain the effect of causing the VM 141 to operate more stably in addition to the effect of the first embodiment.

  The reason is as follows.

  The IO status history storage unit 126 of the information processing apparatus 101 stores the history for the IO request. Then, the IO suppression level determination unit 127 determines the suppression of the VM 141 in the QoS function execution unit 110 based on the history of a predetermined time. Therefore, the IO suppression level determination unit 127 does not suppress the VM 141 for which the IO request has temporarily increased. On the other hand, the IO suppression level determination unit 127 constantly suppresses the VM 141 having many IO requests. In addition, since the IO suppression level determination unit 127 uses the history, the suppression of the VM 141 proceeds gradually. That is, the information processing apparatus 101 according to the present embodiment gradually changes the IO performance of the VM 141. As a result, the operation of the VM 141 becomes stable.

  In addition, the storage device 201 of the information processing system 10 according to the second embodiment can obtain an effect of stabilizing the operation of the logical disk 241.

  The reason is that the IO suppression level determination unit 216 executes suppression of the logical disk 241 using the history similarly to the operation of the IO suppression level determination unit 127 on the VM 141.

<Third Embodiment>
Next, an information processing system 10 according to a third embodiment of the present invention will be described. The information processing system 10 according to the present embodiment includes one or more information processing apparatuses 100 and one or more storage apparatuses 202.

  The information processing apparatus 100 in this embodiment is the same as that in the first embodiment, and thus detailed description thereof is omitted.

  The configuration of the storage device 202 in this embodiment will be described with reference to the drawings.

  FIG. 13 is a block diagram illustrating an example of the configuration of the storage device 202 according to the third embodiment.

  As shown in FIG. 13, in addition to the configuration of the storage device 200 of the first embodiment, the storage device 202 of this embodiment includes a logical configuration storage unit 235, a bottleneck performance monitoring unit 236, and an IO suppression range determination unit. 237.

  The logical configuration storage unit 235 stores information on the configuration included in the storage device 202 that may be related to the bottleneck of the device performance of the storage device 202 and the logical disk 241 that is affected by the performance of the configuration.

  The bottleneck performance monitoring unit 236 monitors the performance state of the configuration that is a bottleneck of performance in the storage device 202. And when the load becomes high in any configuration (when a bottleneck occurs), the bottleneck performance monitoring unit 236 sends the configuration (bottleneck configuration) with a high load to the IO suppression range determination unit 237. Notice.

  The IO suppression range determination unit 237 determines (determines) the logical disk 241 related to the configuration with a high load notified from the bottleneck performance monitoring unit 236 based on the information held by the logical configuration storage unit 235. Then, the IO suppression range determination unit 237 instructs the allocation performance calculation unit 231 to recalculate the IOPS target value so as to eliminate the bottleneck in all the determined logical disks 241.

  The allocation performance calculation unit 231 calculates an appropriate IOPS target value for the logical disk 241 associated with a high-load component. Then, the allocation performance calculation unit 231 transmits the calculated IOPS target value to the allocation performance transmission unit 221. The allocation performance transmission unit 221 sets the received IOPS target value in the QoS function execution unit 210. The QoS function execution unit 210 executes the QoS function of the logical disk 241 based on the IOPS target value of the recalculation result. Also, the allocation performance transmission unit 221 transmits the IOPS target value as a recalculation result to the allocation performance reception unit 122.

  The information processing apparatus 100 recalculates the IOPS target value of the VM 141 using the IOPS target value of the logical disk 241 that has received the notification.

  Note that the storage device 202 may be configured by using a hardware circuit, a plurality of devices, or the computer shown in FIG. 10, similarly to the storage device 200 of the first embodiment.

  Next, the effect of this embodiment will be described.

  The storage device 202 of the information processing system 10 according to the third embodiment can obtain an effect of improving the performance of the logical disk 241 related to the configuration that becomes the bottleneck.

  The reason is as follows.

  The bottleneck performance monitoring unit 236 of the storage device 202 monitors a configuration that becomes a bottleneck. The IO suppression range determination unit 237 determines the logical disk 241 related to the configuration that becomes the bottleneck. Then, the allocation performance calculation unit 231 recalculates the IOPS target value of the logical disk 241 related to the configuration that becomes the bottleneck so as to eliminate the bottleneck. This is because the QoS function execution unit 210 executes the QoS function based on the recalculated IOPS target value. Further, the information processing apparatus 100 sets the IOPS target value of the VM 141 based on the IOPS target value of the new logical disk 241.

  While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

DESCRIPTION OF SYMBOLS 10 Information processing system 100 Information processing apparatus 101 Information processing apparatus 110 QoS function execution part 121 Allocation performance calculation part 122 Allocation performance reception part 123 IO amount preservation | save part 124 IO status measurement part 125 Target performance coefficient preservation | save part 126 IO situation history preservation | save part 127 IO suppression level determination unit 141 VM
142 IO transmission processing unit 200 Storage device 201 Storage device 202 Storage device 210 QoS function execution unit 215 IO status history storage unit 216 IO suppression level determination unit 221 Allocation performance transmission unit 222 Performance target value storage unit 231 Allocation performance calculation unit 232 IO amount Storage unit 233 IO status measurement unit 235 Logical configuration storage unit 236 Bottleneck performance monitoring unit 237 IO suppression range determination unit 241 Logical disk 242 IO reception processing unit 300 Network 600 Information processing device 610 CPU
620 ROM
630 RAM
640 Internal storage device 650 IOC
660 Input device 670 Display device 680 NIC
700 storage media

Claims (8)

VM, which is software that emulates the operation of a computer,
First QoS function execution means for executing a QoS function of an IO request from a VM based on a performance target value of the VM;
IO transmission processing means for receiving an IO request from the first QoS function execution means and transmitting the IO request to a storage device;
The number of IO requests sent from the VM to the first QoS function execution means and the number of IO transmissions sent from the first QoS function execution means to the IO transmission processing means are measured, and unit time First IO status measuring means for calculating a request IOPS that is the number of IO requests per unit and a transmission IOPS that is the number of IO transmissions per unit time;
First IO amount storage means for storing the requested IOPS and transmission IOPS calculated by the first IO status measurement means, and the performance target value of the VM;
Target performance coefficient storage means for storing a target performance coefficient, which is information used when calculating the performance target value of the VM,
Allocation performance receiving means for receiving performance target values of logical disks from the storage device;
Based on the stored VM request IOPS, transmission IOPS, performance target value, target performance coefficient, performance target value of the logical disk previously received by the assigned performance receiving means, and performance target value received this time. First allocation performance calculation means for calculating the performance target value of the VM;
An information processing apparatus including:
A logical disk for storing and reproducing information;
IO reception processing means for receiving an IO request from the information processing apparatus to the logical disk;
Second QoS function executing means for executing the QoS function of the logical disk based on the performance target value of the logical disk;
Allocation performance transmitting means for transmitting the performance target value of the logical disk to the information processing apparatus;
Measures the number of IO requests sent from the IO reception processing means to the second QoS function execution means to the logical disk and the number of IO processes sent from the second QoS function execution means to the logical disk. A second IO status measuring means for calculating an accepted IOPS that is the number of accepted IOs per unit time and a processed IOPS that is the number of processed IOs;
Second IO amount storage means for storing the received IOPS, the processing IOPS, and the performance target value of the logical disk;
Performance target value storage means for storing performance target values set in the logical disk;
Second allocated performance calculation means for calculating the performance target value of the logical disk based on the accepted IOPS, the processing IOPS, the performance target value, and the performance target value stored in the performance target value storage means; An information processing system including a storage device. The first allocation performance calculation means includes:
The information processing system according to claim 1, wherein the performance target value of the VM is calculated when the allocated performance receiving unit receives the performance target value of the logical disk from the storage device. The information processing apparatus is
First IO status history storage means for storing a history of request IOPS, transmission IOPS and performance target value measured by the first IO status measurement means;
A first IO suppression level determination unit that determines a degree to which the first QoS function execution unit suppresses the VM based on a history stored by the first IO status history storage unit;
The first QoS function executing means includes:
The information processing system according to claim 1, wherein the VM is suppressed based on a degree of suppression determined by the first IO suppression level determination unit . The storage device is
Second IO status history storage means for storing a history of the logical disk acceptance IOPS and processing IOPS, and the performance target value of the logical disk, measured by the second IO status measurement means;
A second IO suppression level determining means for determining a degree of suppression of the logical disk by the second QoS function executing means based on the history stored by the second IO status history storing means;
The second QoS function executing means includes:
The information processing system according to any one of claims 1 to 3, wherein the logical disk is suppressed based on a degree of suppression determined by the second IO suppression level determination unit. The storage device is
Logical configuration storage means for storing information on the configuration related to the performance bottleneck and the logical disk affected by the related configuration;
Monitoring the performance status of the configuration related to the performance bottleneck, and bottleneck performance monitoring means for notifying the configuration when a bottleneck occurs in any configuration;
An IO suppression range determining means for determining logical disks affected by the configuration notified from the bottleneck performance monitoring means based on the information held by the logical configuration storage means;
The second allocation performance calculation means is
The information processing system according to claim 1, wherein the performance target value of the logical disk determined by the IO suppression range determination unit is recalculated. VM, which is software that emulates the operation of a computer,
First QoS function execution means for executing a QoS function of an IO request from a VM based on a performance target value of the VM;
IO transmission processing means for receiving an IO request from the first QoS function execution means and transmitting the IO request to a storage device;
The number of IO requests sent from the VM to the first QoS function execution means and the number of IO transmissions sent from the first QoS function execution means to the IO transmission processing means are measured, and unit time First IO status measuring means for calculating a request IOPS that is the number of IO requests per unit and a transmission IOPS that is the number of IO transmissions per unit time;
First IO amount storage means for storing the requested IOPS and transmission IOPS calculated by the first IO status measurement means, and the performance target value of the VM;
Target performance coefficient storage means for storing a target performance coefficient, which is information used when calculating the performance target value of the VM,
Allocation performance receiving means for receiving performance target values of logical disks from the storage device;
Based on the stored VM request IOPS, transmission IOPS, performance target value, target performance coefficient, performance target value of the logical disk previously received by the assigned performance receiving means, and performance target value received this time. First allocation performance calculation means for calculating the performance target value of the VM;
An information processing apparatus including: In an information processing apparatus of an information system including an information processing apparatus and a storage device, VM that is software for emulating the operation of a computer is executed.
Execute IO function QoS function from VM based on VM performance target value,
Sending an IO request sent based on the QoS function to the storage device;
The number of IO requests from the VM and the number of IO transmissions sent to the storage device are measured. The request IOPS is the number of IO requests per unit time, and the number of IO transmissions per unit time. Calculate the sending IOPS,
Storing the calculated request IOPS and transmission IOPS, and the performance target value of the VM;
Save the target performance coefficient, which is information used when calculating the performance target value of the VM,
Receiving the performance target value of the logical disk from the storage device;
Based on the saved VM request IOPS, transmission IOPS and performance target value, the target performance coefficient, the performance target value of the logical disk received last time, and the performance target value received this time, the VM performance target Calculate the value
In the storage device,
Execute a logical disk for storing and reproducing information,
Receiving an IO request to the logical disk from the information processing apparatus;
Execute the QoS function of the logical disk based on the performance target value of the logical disk;
Transmitting the performance target value of the logical disk to the information processing apparatus;
The number of IO requests sent from the information processing apparatus to the logical disk and the number of IO processes in which the QoS function of the logical disk is executed are measured, and the received IOPS is the number of received IOs per unit time. Calculate the processing IOPS, which is the number of processing IOs,
Storing the received IOPS, the processing IOPS, and the performance target value of the logical disk;
Save the performance target value set in the logical disk,
An information processing method for calculating a performance target value of the logical disk based on the received IOPS, processing IOPS, performance target value, and the stored performance target value. As an information processing device of an information system including an information processing device and a storage device,
A process of executing VM, which is software that emulates the operation of a computer;
A process of executing the QoS function of the IO request from the VM based on the performance target value of the VM;
A process of transmitting an IO request transmitted based on the QoS function to the storage device;
The number of IO requests from the VM and the number of IO transmissions sent to the storage device are measured. The request IOPS is the number of IO requests per unit time, and the number of IO transmissions per unit time. Processing to calculate the transmission IOPS;
Processing for storing the calculated request IOPS and transmission IOPS and the performance target value of the VM;
Processing for storing a target performance coefficient, which is information used when calculating a performance target value of the VM,
Processing for receiving a performance target value of a logical disk from the storage device;
Based on the saved VM request IOPS, transmission IOPS and performance target value, the target performance coefficient, the performance target value of the logical disk received last time, and the performance target value received this time, the VM performance target A process of calculating a value and as the storage device,
Processing to execute a logical disk for storing and reproducing information;
A process of receiving an IO request from the information processing apparatus to the logical disk;
A process of executing a QoS function of the logical disk based on a performance target value of the logical disk;
Processing for transmitting the performance target value of the logical disk to the information processing apparatus;
The number of IO requests sent from the information processing apparatus to the logical disk and the number of IO processes in which the QoS function of the logical disk is executed are measured, and the received IOPS is the number of received IOs per unit time. A process of calculating a process IOPS which is the number of processes IO;
Processing for storing the accepted IOPS, the processing IOPS, and a performance target value of the logical disk;
Processing for storing performance target values set in the logical disk;
A program that causes a computer to execute a process of calculating a performance target value of the logical disk based on the received IOPS, a processing IOPS, a performance target value, and the stored performance target value.

Images