Operating mode setting device, operating mode setting method, operating mode setting program, and system
The operating mode setting device optimizes server power efficiency and performance by dynamically adjusting hardware settings based on load, addressing inefficiencies in existing virtualization systems.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- NIPPON TELEGRAPH & TELEPHONE CORP
- Filing Date
- 2022-09-07
- Publication Date
- 2026-05-15
AI Technical Summary
Server power efficiency varies with load, being low at low loads due to base power consumption and inefficient under high loads when power-saving settings degrade performance, and existing virtualization infrastructures fail to dynamically adjust hardware settings to meet performance requirements.
An operating mode setting device that monitors hardware load and dynamically selects optimal combinations of settings such as CPU cores, frequencies, and FAN settings based on power efficiency and performance metrics, ensuring compliance with SLA requirements.
Improves power efficiency and meets performance demands by dynamically adjusting hardware settings during operation, enhancing energy utilization while maintaining service quality.
Smart Images
Figure 0007859510000001 
Figure 0007859510000002 
Figure 0007859510000003
Abstract
Description
Technical Field
[0001] The present invention relates to an operation mode setting device, an operation mode setting method, an operation mode setting program, and a system.
Background Art
[0002] In recent years, there has been concern about the increasing power consumption of IT devices, and in server farms such as data centers, in addition to the performance of devices, power efficiency has come to be regarded. Power efficiency shows different tendencies depending on the hardware architecture. Also, even for the same hardware, by changing settings such as the number of CPU cores to be enabled, the maximum / minimum frequency, C-state, DVFS (Dynamic Voltage and Frequency Scaling), FAN setting, etc., the tendency of power efficiency changes.
[0003] A virtualization platform uses virtualization technology to abstract and conceal the hardware that constitutes a server or a network, and constructs a virtual machine or a container that is an execution environment. The virtualization platform is a virtual environment prepared as a common platform for a plurality of applications and services, and a system that manages those virtual environments. In the open-source virtualization platforms in the market, in many cases, de facto standard OSS (Open Source Software) such as OpenStack and Kubernetes is used. OpenStack is software for constructing a cloud environment, and mainly manages and operates physical machines and virtual machines. Kubernetes is software for operating, managing, and automating containerized workloads and services.
[0004] Virtualization infrastructures optimize hardware resource utilization and power consumption through dynamic scaling and migration, for example (see, for example, Non-Patent Document 1). Commercial virtualization infrastructures generally feature auto-scaling capabilities, automatically adjusting the number of virtual machines and containers based on system usage. Furthermore, virtualization infrastructure employs optimizations such as DVFS technology for frequency and voltage (see, for example, Non-Patent Document 2). DVFS technology reduces server power consumption by dynamically changing the CPU clock frequency and CPU core voltage according to the server load. [Prior art documents] [Non-patent literature]
[0005] [Non-Patent Document 1] Than, Moh Moh, and Thandar Thein. “Energy-Saving Resource Allocation in Cloud Data Centers” 2020 IEEE Conference on Computer Applications (ICCA). IEEE, 2020. [Non-Patent Document 2] Kuehn, PJ and Mashaly, M. “DVFS-power management and performance engineering of data center server clusters” 2019 15th Annual Conference on Wireless Ondemand Network Systems and Services (WONS), IEEE, pp. 91-98 (2019). [Overview of the project] [Problems that the invention aims to solve]
[0006] Server power efficiency varies depending on the load on the server, but especially at low loads, power efficiency tends to be low due to base power (power consumed regardless of load). In this case, it is possible to reduce power consumption and improve power efficiency by changing hardware settings (number of CPU cores to enable, maximum / minimum frequency, C-state, DVFS, FAN settings, etc.). On the other hand, while power efficiency improves under high server load compared to low load, settings designed to conserve power may degrade server performance (throughput, latency, etc.). In this case, the server may not be able to meet the performance requirements stipulated in the SLA (Service Level Agreement).
[0007] In typical virtualization infrastructures, hardware configuration changes are generally not made while the server is running. As mentioned above, virtualization infrastructures generally employ autoscaling, which controls the number and placement of virtual machines or containers according to the server load, and optimizes the number of hardware devices running on them. However, commercially available virtualization infrastructure does not take into account the dynamic modification of individual hardware settings. Therefore, when running CPU-intensive applications, especially in situations where the load is low immediately after adding hardware through scaling, the hardware is likely to be operating in an inefficient state. Similarly, when running memory-intensive applications, the CPU load is low regardless of scaling, so the hardware is likely to be operating in an inefficient state.
[0008] Therefore, it is necessary to dynamically configure the hardware's operating mode (a combination of factors that affect the hardware's power consumption) while the hardware is running, thereby improving power efficiency while meeting the performance requirements of the hardware. [Means for solving the problem]
[0009] The operating mode setting device according to the present invention is an operating mode setting device provided in a system comprising hardware and a virtualization infrastructure that allocates resources of the hardware to construct a virtual machine or container, and is characterized by comprising: a storage unit that stores a plurality of operating modes consisting of a combination of items that can be dynamically set on the hardware in relation to the power consumption of the hardware, and power characteristic data including power efficiency and performance index values obtained for each of the plurality of operating modes with different load rates of the hardware; a monitoring unit that monitors the hardware in operation and obtains metric data including the load rate of the hardware; and an operating mode setting unit that refers to the power characteristic data and selects operating modes from the plurality of operating modes in order of the highest power efficiency at the load rate of the metric data, and operates the hardware in the selected operating mode if the performance index value at the load rate of the selected operating mode satisfies a predetermined performance requirement of the hardware. [Effects of the Invention]
[0010] According to the present invention, the operating mode of the hardware (a combination of items that affect the power consumption of the hardware) can be dynamically set during the operation of the hardware, thereby improving power efficiency while meeting the performance requirements of the hardware. [Brief explanation of the drawing]
[0011] [Figure 1] This is a block diagram showing an example of the configuration of a system including an operating mode setting device according to this embodiment. [Figure 2] This figure shows the configuration of the system according to this embodiment. [Figure 3] This figure shows an example of hardware operating mode settings. [Figure 4] This figure shows an example of power consumption data for each operating mode. [Figure 5] This figure shows an example of power efficiency data for each operating mode. [Figure 6]It is a diagram showing an example of data of performance index values (latency) for each operation mode. [Figure 7] It is a flowchart for explaining the processing flow of the operation mode setting device according to the present embodiment. [Figure 8] It is a hardware configuration diagram showing an example of a computer that realizes the functions of the operation mode setting device according to the present embodiment. [Figure 9] It is a block diagram showing a configuration example of a system including an operation mode setting device according to Modification 1.
Mode for Carrying Out the Invention
[0012] Next, a mode for carrying out the present invention (hereinafter referred to as "the present embodiment") will be described with reference to the drawings. FIG. 1 is a diagram showing a configuration example of a system 1 including an operation mode setting device 4 according to the present embodiment. FIG. 2 is a diagram showing the configuration of the system 1 according to the present embodiment. In the present embodiment, an example will be described in which the operation mode setting device 4 is configured as a controller 2 of the system 1 together with an in-city virtualization infrastructure 3.
[0013] As shown in FIG. 1, the system 1 according to the present embodiment includes a plurality of hardware 5a, 5b, 5c, 5d... constituting a server or a network, and a controller 2 that controls the plurality of hardware 5a, 5b, 5c, 5d.... The hardware 5a, 5b, 5c, 5d... can include accelerator resources such as, for example, a GPU (Graphics Processing Unit) and an FPGA (field-programmable gate array).
[0014] The controller 2 has a virtualization infrastructure 3 and an operation mode setting device 4. As shown in FIG. 2, the virtualization infrastructure 3 uses virtualization technology to abstract and conceal a plurality of hardware components 5a, 5b, 5c, 5d... that make up the server and network, and manages virtual machines (VMs) 6a, 6b, 6c... that operate on any of the hardware components 5a, 5b, 5c, 5d... A load balancer 7 is connected to each virtual machine 6a, 6b, 6c..., and traffic is distributed from the Internet 9 via the local network 8. The load balancer 7 can also be constructed as a virtual machine. Each hardware component 5a, 5b, 5c, 5d... includes an application processing unit 51 (FIG. 1) that executes an application assigned to each virtual machine.
[0015] In the following description, when the hardware components 5a, 5b, 5c, 5d... are not particularly distinguished, they are simply referred to as hardware 5. Also, when the virtual machines 6a, 6b, 6c... are not particularly distinguished, they are simply referred to as virtual machine 6. Although FIG. 1 illustrates an example using a virtual machine 6 as the execution environment, a container may also be used as the execution environment.
[0016] <Virtualization Infrastructure> As shown in FIG. 1, the virtualization infrastructure 3 includes a resource management unit 31, a virtual machine control unit 32, a hardware startup / stop control unit 33, and a storage unit 34. The hardware startup / stop control unit 33 is hereinafter referred to as the "HW startup / stop control unit 33".
[0017] The resource management unit 31 manages the resources of the hardware 5 assigned to each virtual machine 6. The resource management unit 31 further monitors the load of the resources assigned to each virtual machine 6 and periodically collects metric data such as the load rate and performance metric values.
[0018] The virtual machine control unit 32 controls the virtual machine 6 based on the metrics data collected by the resource management unit 31. As an example of controlling the virtual machine 6, the virtual machine control unit 32 performs scaling processing. Scaling operations specifically include changing resource allocations for virtual machine 6, deleting virtual machine 6 (scale-in), and adding virtual machine 6 (scale-out). The virtual machine control unit 32 also changes the number of operating hardware 5 units as needed. When changing the number of operating hardware 5 units, the virtual machine control unit 32 inputs instructions to the hardware startup / shutdown control unit 33. The hardware startup / shutdown control unit 33 controls the startup or shutdown of the target hardware 5 unit. The virtual machine control unit 32 also performs scaling processing in the same manner when it receives a scaling processing instruction from the operation mode setting device 4, which will be described later.
[0019] The memory unit 34 stores various types of information necessary for processing the virtualization infrastructure 3. The memory unit 34 stores, for example, resource information for hardware 5, resource allocation information for virtual machine 6, and so on.
[0020] Furthermore, if containers are used instead of virtual machines 6 as the execution environment, the resource management unit 31 manages the hardware resources 5 allocated to each container. In addition, the virtualization infrastructure 3 can be provided with a container control unit instead of a virtual machine control unit 32. The container control unit can perform scaling operations such as increasing or decreasing the number of containers.
[0021] <Operation Mode Setting Device> As shown in Figure 1, the operating mode setting device 4 includes a power characteristic data calculation unit 41, a monitoring unit 42, a hardware setting control unit 43 (operating mode setting unit), and a storage unit 44. The "hardware setting control unit 43" will be referred to as the "HW setting control unit 43" from now on. As described above, the virtualization infrastructure 3 primarily manages virtual machines 6a, 6b, 6c, etc., which operate on one of the hardware components 5a, 5b, 5c, 5d, etc. On the other hand, the operation mode setting device 4 is a device that manages the operation mode of each hardware component 5 that makes up system 1.
[0022] The operating mode is a combination of items that can be dynamically configured for hardware 5 and are related to the power consumption of hardware 5. Items related to the power consumption of hardware 5 refer to items that affect the power consumption of hardware 5, for example. Examples of settings that can be configured in operation mode include the number of CPU cores to enable, maximum / minimum CPU frequency, maximum / minimum GPU frequency, C-state settings, DVFS settings, and FAN settings. The operating mode setting device 4 provides multiple operating modes for the hardware 5. These pre-created operating modes are listed in the operating mode list 441 and stored in the memory unit 44. The same operating mode list 441 may be used for each piece of hardware 5 constituting the system 1. If the system 1 contains hardware 5 with different specifications, multiple operating mode lists 441 corresponding to the specifications of each piece of hardware 5 may be created.
[0023] Figure 3 shows an example of the settings for the operating mode of hardware 5. Figure 3 shows five different operating modes, OM1 to OM5. Figure 3 illustrates how the number of CPU cores to enable, the maximum CPU frequency, and the maximum GPU frequency are settings for each of the operating modes OM1 to OM5. For example, in operating mode OM1, the number of enabled CPU cores is set to a low 2, while the maximum CPU frequency is set to 2000MHz and the maximum GPU frequency to 1000MHz. Similarly, in operating mode OM5, the number of enabled CPU cores is set to a high 6, while the maximum CPU frequency is set to 1200MHz and the maximum GPU frequency to 800MHz. Each of the other operating modes also has its own set of values.
[0024] The power characteristics data calculation unit 41 operates each of the hardware components 5 of the system 1 in one of the multiple operating modes listed in the operating mode list 441, and calculates power characteristics data 442 for each operating mode. The power characteristics data calculation unit 41 can, for example, perform a test operation of the hardware components 5 before their official operation and calculate the power characteristics data 442. The power characteristics data 442 is used by the HW setting control unit 43, described later, as a criterion for setting the operating mode of the hardware components 5.
[0025] The power characteristics data calculation unit 41 acquires the data necessary for calculating power characteristics data 442 for each operating mode while varying the load factor LF of the hardware 5. The load factor LF refers to the usage rate (Usage) or request processing rate (RFS: Requests Per Second) of the hardware 5. The power characteristics data calculation unit 41 measures data on the power consumption, throughput (number of data processed), and performance indicator value PIV (e.g., latency) of the hardware 5 while varying the load factor LF of the hardware 5.
[0026] The power characteristics data calculation unit 41 further calculates the power efficiency PE of the hardware 5 from the relationship between the measured power consumption and throughput. Power efficiency PE represents "the power efficiency change with usage rate, with power efficiency at 100% usage rate set to 1.0". When using power consumption and throughput, power efficiency PE is calculated as the number of processes per watt. The power characteristic data calculation unit 41 stores the performance index value PIV and the calculated power efficiency PE as power characteristic data 442 in the storage unit 44. The power characteristic data 442 is stored in association with each operating mode listed in the operating mode list 441.
[0027] Figure 4 shows an example of power consumption data for each operating mode. Figure 5 shows an example of power efficiency PE data for each operating mode. Figure 6 shows an example of PIV (latency) data for each operating mode. As shown in Figures 4 to 6, the multiple operating modes OM1 to OM5 exhibit different trends in power consumption, power efficiency PE, and performance index PIV depending on the combination of their respective settings.
[0028] As shown in Figure 4, the power consumption of hardware 5 increases in proportion to the increase in load factor LF, but the rate of increase in power consumption differs depending on the operating mode. As shown in Figure 5, at low load factors, the power efficiency PE tends to be lower due to the base power (power consumed regardless of load), and the power efficiency PE increases as the load factor LF increases. Here, operating modes OM4 and OM5 have higher power efficiency PE at high load factors compared to operating modes OM1 to OM3, but as shown in Figure 6, the performance index value PIV (latency) at high load factors is also larger.
[0029] Returning to Figure 1, the monitoring unit 42 monitors each piece of hardware 5 while it is operating and periodically collects metric data. The monitoring unit 42 collects data such as the load ratio LF (utilization rate or number of requests processed) and performance indicator values PIV such as latency as metric data. The metric data is temporarily stored in the storage unit 44. The monitoring unit 42 also collects load factor LF data when the HW setting control unit 43, described later, changes the operating mode of the hardware 5.
[0030] The hardware setting control unit 43 dynamically selects and sets an operating mode for each hardware 5 in operation from among multiple operating modes. Specifically, the HW setting control unit 43 refers to the power characteristic data 442 (Figure 5) and selects from among several operating modes an operating mode in which the power efficiency PE at the load factor LF indicated by the metrics data is higher than that of the other operating modes. The HW setting control unit 43 refers to the power characteristic data 442 (Figure 6) and, if the performance index value PIV at the load factor LF of the selected operating mode satisfies the performance requirements of the hardware 5 based on the SLA (Service Level Agreement) (hereinafter referred to as "predetermined performance requirements"), it operates the hardware 5 in the selected operating mode.
[0031] More specifically, the HW setting control unit 43 sorts the multiple operating modes in the operating mode list 441 in descending order of power efficiency PE at the load factor LF indicated by the metrics data. The HW setting control unit 43 then selects the first operating mode from the sorted operating mode list 441 and determines whether the performance index value PIV at the load factor LF of the selected operating mode satisfies predetermined performance requirements. In other words, the HW setting control unit 43 selects from among multiple operating modes an operating mode in which the power efficiency PE of the operating hardware 5 is high at load factor LF and which satisfies predetermined performance requirements.
[0032] An SLA (Service Level Agreement) is an agreement between the provider of System 1 and the user regarding the level of service. The hardware 5 that constitutes System 1 is required to ensure the performance metrics specified in the SLA. Therefore, as a predetermined performance requirement, it is required that the hardware be below the performance metrics specified in the SLA (e.g., latency).
[0033] The HW setting control unit 43, in determining predetermined performance requirements, specifically refers to the power characteristic data 442 and performs a comparison process between the performance index value PIV of the selected operating mode and the pre-set performance target value SLO. The performance target value SLO is set based on the performance index value stipulated in the SLA. The performance target value SLO (predetermined performance requirement) may be the same as the performance index value stipulated in the SLA, or it may be a value that provides a margin over the performance index value stipulated in the SLA. For example, if the performance index value is latency, the performance target value SLO may be a value lower than the latency stipulated in the SLA.
[0034] The hardware setting control unit 43 sets the selected operating mode as the operating mode of the hardware 5 if the performance index value PIV at the load factor LF of the selected operating mode satisfies the predetermined performance requirements. If the performance index value PIV at the load factor LF of the selected operating mode does not meet the predetermined performance requirements, the HW setting control unit 43 selects the next operating mode with the highest power efficiency PE from the operating mode list 441 and sequentially determines whether it meets the predetermined performance requirements. Furthermore, if there is no operating mode in the operating mode list 441 that satisfies the predetermined performance requirements, the HW configuration control unit 43 outputs a scaling instruction to the virtualization infrastructure 3. For example, the HW configuration control unit 43 outputs an instruction to scale out by adding 6 virtual machines / containers or an instruction to increase the number of hardware units 5.
[0035] The hardware configuration control unit 43 sets an operating mode that satisfies predetermined performance requirements, and then determines whether scaling processing is necessary. As described above, the monitoring unit 42 collects load ratio LF data (utilization rate or number of requests processed) from the hardware 5 whose operating mode has been changed. The hardware configuration control unit 43 outputs a scaling instruction to the virtualization infrastructure 3 if the load ratio LF is lower than a preset threshold TH (a predetermined threshold). As an example, the hardware configuration control unit 43 outputs an instruction to scale in by deleting virtual machines 6 / containers or an instruction to reduce the number of hardware 5 units. Thus, if there is an increase or decrease in the load on hardware 5 that cannot be adequately addressed by changing the operating mode, the issue can be resolved by issuing a scaling instruction to the virtualization infrastructure 3.
[0036] As described above, the HW setting control unit 43 dynamically sets the operating mode according to the load ratio LF of the hardware 5 measured by the monitoring unit 42 while the hardware 5 is operating. On the other hand, when hardware 5 starts operating, the load factor LF of hardware 5 is expected to be low. Therefore, when hardware 5 starts operating, the HW setting control unit 43 refers to the power characteristic data 442 (Figure 5) and selects from among multiple operating modes the operating mode with the highest power efficiency PE at a low load factor (load factor below a predetermined value) and sets it as the operating mode of hardware 5. The predetermined value can be set from the load factor expected when hardware 5 starts operating.
[0037] Figure 7 is a flowchart illustrating the processing flow of the operating mode setting device 4. Figure 7 shows the processing of the hardware 5 of the operating mode setting device 4 at the start of operation and during operation. As described above, the power characteristic data calculation unit 41 calculates power characteristic data 442 for each of the multiple operating modes through prior test runs and stores it in the storage unit 44.
[0038] As shown in Figure 7, the HW setting control unit 43 refers to the power characteristic data 442 (Figure 5) in the storage unit 44 and sorts the operating modes listed in the operating mode list 441 in descending order of power efficiency PE at low load rates (load rates below a predetermined value) (step S01).
[0039] The HW setting control unit 43 selects the first operating mode from the top of the operating mode list 441 (the operating mode with the highest power efficiency PE) and sets it as the operating mode for the hardware 5 (step S02). The HW setting control unit 43 operates the hardware 5 according to the settings of the selected operating mode.
[0040] While hardware 5 is operating, the monitoring unit 42 periodically collects metrics data (load ratio LF, etc.) from hardware 5 (step S03). The HW setting control unit 43 obtains the load factor LF of the operating hardware 5 from the metrics data. The HW setting control unit 43 refers to the power characteristic data 442 (Figure 5) and sorts the operating modes listed in the operating mode list 441 in descending order of power efficiency PE at load factor LF (step S04).
[0041] The HW setting control unit 43 sets a=1 (step S05) and selects the a-th operating mode from the top of the operating mode list 441 (step S06). The hardware 5 setting control unit refers to the power characteristic data 442 of the selected operating mode and determines whether the performance index value PIV satisfies the predetermined performance requirements (step S07). Specifically, the HW setting control unit 43 performs a comparison process between the performance index value PIV at the load factor LF of the power characteristic data 442 (Figure 6) and the performance target value SLO, which is based on the performance index value stipulated in the SLA. For example, if the performance index value PIV is latency, the HW setting control unit 43 determines that the predetermined performance requirements are met if the latency of the metrics data is less than the performance target value SLO.
[0042] If the HW setting control unit 43 determines that the performance index value PIV of the selected operating mode satisfies a predetermined performance requirement (step S07: Yes), it proceeds to step S10. If the performance index value PIV of the selected operating mode does not meet the predetermined performance requirements (step S07: No), the HW setting control unit 43 sets a = a + 1 (step S08). Then, the HW setting control unit 43 determines whether or not an a-th operating mode exists in the operating mode list 441 (step S09). If an a-th operating mode exists in the operating mode list 441 (step S09: Yes), the process returns to step S06, and the HW setting control unit 43 determines whether the a-th operating mode meets the predetermined performance requirements. In this way, the HW setting control unit 43 sequentially determines the predetermined performance requirements for the operating modes listed in the operating mode list 441, in order of decreasing power efficiency PE.
[0043] If the a-th operating mode does not exist in the operating mode list 441 (step S09: No), that is, if all the operating modes listed in the operating mode list 441 do not meet the predetermined performance requirements, the HW setting control unit 43 proceeds to step S14 and instructs the virtual machine control unit 32 of the virtualization infrastructure 3 to perform scaling. In this case, since an increase in the load rate that cannot be handled by changing the operating mode is expected in the hardware 5, the HW setting control unit 43 can, for example, instruct the virtualization infrastructure 3 to scale out by increasing the number of virtual machines 6 / containers, or to increase the number of hardware units 5.
[0044] In step S10, the HW setting control unit 43 determines whether the selected operating mode matches the operating mode currently set for the hardware 5. If the operating modes do not match (step S10: No), the HW setting control unit 43 changes the operating mode of the hardware 5 to the selected operating mode (step S11). The HW setting control unit 43 then operates the hardware 5 according to the settings of the changed operating mode. In step S10, if the selected operating mode matches the currently set operating mode, the HW setting control unit 43 proceeds to step S15.
[0045] In step S12, the monitoring unit 42 obtains load ratio LF (utilization rate or number of requests processed) data from the hardware 5 whose operating mode has been changed. The HW setting control unit 43 determines whether the load ratio LF acquired by the monitoring unit 42 is lower than the threshold TH (a predetermined threshold) (step S13). If the load ratio LF is equal to or greater than the threshold TH (step S13: No), the process proceeds to step S15.
[0046] If the load ratio LF acquired by the monitoring unit 42 is lower than the threshold TH (step S13: Yes), the HW setting control unit 43 proceeds to step S14 and instructs the virtualization infrastructure 3 to scale. In this case, since it is assumed that the load ratio of hardware 5 will remain low even if the operating mode is changed, the HW setting control unit 43 can, for example, issue instructions to scale in to reduce the number of virtual machines 6 / containers or to reduce the number of hardware units 5.
[0047] After waiting for a predetermined time in step S15, the operating mode setting device 4 returns to step S03. This allows the operating mode setting device 4 to periodically acquire hardware metric data while the hardware 5 is operating and to dynamically set the operating mode. Specifically, the operating mode setting device 4 revises the operating mode in response to changes in the load factor LF while the hardware 5 is operating, taking into account both power efficiency PE and predetermined performance requirements. Furthermore, for changes in the load factor LF that cannot be addressed by changing the operating mode, the operating mode setting device 4 instructs the virtualization infrastructure 3 to scale.
[0048] <Hardware Configuration> The operating mode setting device 4 according to this embodiment is implemented by a computer 900, for example, as shown in Figure 8. Figure 8 is a hardware configuration diagram showing an example of a computer 900 that implements the functions of the operation mode setting device 4 according to this embodiment. The computer 900 has a CPU (Central Processing Unit) 901, ROM (Read Only Memory) 902, RAM 903, HDD (Hard Disk Drive) 904, input / output I / F (Interface) 905, communication I / F 906, and media I / F 907.
[0049] The CPU 901 operates based on a program (data acquisition program) stored in the ROM 902 or HDD 904, and controls the operation mode setting device 4 shown in Figure 1. The ROM 902 stores the boot program executed by the CPU 901 when the computer 900 starts up, as well as programs related to the computer 900's hardware 5, etc.
[0050] The CPU 901 controls input devices 910, such as a mouse or keyboard, and output devices 911, such as a display, via the input / output interface 905. The CPU 901 acquires data from the input devices 910 and outputs the generated data to the output devices 911 via the input / output interface 905. In addition to the CPU 901, a GPU (Graphics Processing Unit) or the like may also be used as a processor.
[0051] HDD904 stores programs executed by CPU901 and data used by those programs. Communication I / F906 receives data from other devices via a communication network (e.g., NW(Network)920) and outputs it to CPU901, and also transmits data generated by CPU901 to other devices via the communication network.
[0052] The media interface 907 reads a program or data stored in the recording medium 912 and outputs it to the CPU 901 via the RAM 903. The CPU 901 loads the program related to the desired processing from the recording medium 912 onto the RAM 903 via the media interface 907 and executes the loaded program. The recording medium 912 can be an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto Optical Disk), a magnetic recording medium, a conductive memory tape medium, or a semiconductor memory.
[0053] For example, when computer 900 functions as the operation mode setting device 4 according to this embodiment, the CPU 901 of computer 900 realizes the function of the operation mode setting device 4 by executing a program loaded on RAM 903. The HDD 904 stores the data in RAM 903. The CPU 901 reads and executes the program related to the desired processing from the recording medium 912. Alternatively, the CPU 901 may read the program related to the desired processing from another device via a communication network (NW 920).
[0054] <Configuration of the above embodiment and its effects> (1) The operating mode setting device 4 is provided in a system 1 which includes hardware 5 and a virtualization infrastructure 3 that allocates the resources of hardware 5 to build a virtual machine 6 or container. The operation mode setting device 4 includes a storage unit 44, a monitoring unit 42, and a hardware setting control unit 43 (operation mode setting unit). The storage unit 44 stores an operating mode list 441 that describes multiple operating modes and power characteristic data 442. The multiple operating modes are combinations of items that are related to the power consumption of the hardware 5 and can be dynamically set on the hardware 5, and are stored in the storage unit 44 as the operating mode list 441. The power characteristic data 442 is obtained by the power characteristic data calculation unit 41 for each of the multiple operating modes, with a different load factor LF of the hardware 5, and includes data on power efficiency PE and performance index value PIV. The monitoring unit 42 monitors the operating hardware 5 and acquires metric data, including the load ratio LF of the hardware 5. The HW setting control unit 43 refers to the power characteristic data 442 and selects from among multiple operating modes in order of highest power efficiency PE at the load factor LF of the metrics data. If the performance index value PIV at the load factor LF of the selected operating mode satisfies the predetermined performance requirements of the hardware 5, the HW setting control unit 43 operates the hardware 5 in the selected operating mode.
[0055] According to the present invention, during the operation of the hardware 5, the operating mode of the hardware 5 (a combination of items that affect the power consumption of the hardware) can be dynamically set, thereby improving power efficiency PE while satisfying the performance requirements (predetermined performance requirements) required for the hardware 5. Specifically, the HW setting control unit 43 can select an operating mode with high power efficiency PE at the load factor LF of the operating hardware 5 by referring to the power characteristic data 442. The HW setting control unit 43 further determines whether the performance index value PIV of the selected operating mode meets predetermined performance requirements. In this way, the operating mode setting device 4 can improve the power efficiency of the hardware 5 while meeting predetermined performance requirements.
[0056] (2) When the hardware 5 starts operating, the HW setting control unit 43 refers to the power characteristic data 442 and selects the operating mode in which the power efficiency PE at a load factor below a predetermined value, i.e., a low load factor, is highest, and operates the hardware 5 in the selected operating mode.
[0057] When hardware 5 starts operating, the load is generally expected to be low. Therefore, the HW setting control unit 43 can improve power efficiency at startup by selecting an operating mode with high power efficiency PE at a preset low load rate.
[0058] (3) If the load ratio LF obtained in the selected operating mode is lower than the threshold TH (a predetermined threshold), the HW setting control unit 43 instructs the virtualization infrastructure 3 to perform scaling of the virtual machine 6 or container.
[0059] As a result, if changing the operating mode of hardware 5 is insufficient to adequately handle increases or decreases in load, power efficiency can be improved by scaling to increase or decrease the number of virtual machines 6 / containers or the number of hardware 5 units. Thus, in system 1 of this embodiment, power efficiency can be further improved by combining dynamic setting of the operating mode of hardware 5 with scaling.
[0060] (4) The operating mode setting device 4 includes a power characteristic data calculation unit 41 that calculates power characteristic data 442 including the power efficiency and performance index value PIV of the hardware 5 while varying the load factor LF of the hardware 5 for each of the multiple operating modes.
[0061] This allows us to test-run hardware 5 before official deployment and obtain highly accurate power characteristic data.
[0062] The effects described above can also be applied to the operation mode setting method performed by the operation mode setting device 4, and to the operation mode setting program that causes the computer 900 to function as the operation mode setting device 4. Furthermore, the effects described above can also be applied to a system 1 equipped with an operating mode setting device 4. In this embodiment, an example was described in which the operating mode setting device 4 is provided together with the virtualization infrastructure 3 as a controller 2 for the hardware 5. The operating mode setting device 4 centrally manages the operating modes of multiple hardware 5 at the controller 2. This configuration is particularly effective in a virtualization environment composed of identical hardware 5. Furthermore, when scaling on the virtualization infrastructure 3, it is necessary to determine that the system is a cluster composed of multiple hardware components 5. By having the operation mode setting device 4 collect and centrally manage information on multiple hardware components 5, the virtualization infrastructure 3 can make decisions more easily, and the number of transactions between the virtualization infrastructure 3 and the individual hardware components 5 can be reduced.
[0063] <Example 1> Figure 9 shows an example of the configuration of the operating mode setting device 4 according to Modification 1. In the above-described embodiment, an example was explained in which the operation mode setting device 4 is configured as a controller of system 1 together with the virtualization infrastructure 3, but the embodiment is not limited to this. As shown in Figure 9, in Modification 1, the operating mode setting device 4 is provided on each of the individual hardware components 5a, 5b, 5c, 5d, etc. that make up the system 1. The operating mode setting device 4 in Modification 1 has the same configuration as the operating mode setting device 4 in the embodiment described above, and performs generally the same processing, so a detailed explanation will be omitted.
[0064] In the modified example 1, the monitoring unit 42 of the operating mode setting device 4 collects metric data from the hardware 5 on which each operating mode setting device 4 is installed. The HW setting control unit 43 sets the operating mode of the hardware 5, which is equipped with an operating mode setting device 4. In other words, in the modified example 1, the operating mode setting device 4 provided on each piece of hardware 5 individually sets the operating mode of the hardware 5. Similar to the embodiment described above, the HW setting control unit 43 issues a scaling instruction to the virtualization infrastructure 3 if there is an increase or decrease in the load factor LF that cannot be adequately addressed by changing the operating mode. The virtualization infrastructure 3 aggregates the scaling instructions input from the operating mode setting device 4 of each hardware 5. Based on the aggregated scaling instruction information, the virtualization infrastructure 3 can increase or decrease the number of virtual machines 6 / containers or the number of hardware 5 units.
[0065] As described above, the operating mode setting device 4 according to the modified example 1 is provided as a controller 2 in the hardware 5 that constitutes the system 1. The embodiment of Modification 1 is particularly effective in a virtualization environment where hardware 5 with different specifications are mixed. That is, different operating modes can be created for each piece of hardware 5 according to its specifications, and the operating modes can be individually set in the operating mode setting device 4 provided for each piece of hardware 5. Furthermore, in Modification 1, since the operating mode setting process can be performed quickly after metric data is collected by each hardware 5, it becomes possible to control the hardware 5 in short spans in response to changes in the load ratio LF.
[0066] It should be noted that the present invention is not limited to the embodiments described above, and many modifications are possible within the technical concept of the present invention by those with ordinary skill in the art. [Explanation of Symbols]
[0067] 1 System 2 Controllers 3. Virtualization infrastructure 4. Operating mode setting device 5 Hardware 6 Virtual Machines 7 Load balancer 31 Resource Management Department 32 Virtual Machine Control Unit 33 Hardware Startup / Shutdown Control Unit 34 Storage section 41 Power Characteristics Data Calculation Unit 42 Monitoring Department 43 Hardware configuration control unit (operation mode setting unit) 44 Memory section 441 Operating Mode List 442 Power Characteristics Data
Claims
1. An operating mode setting device provided in a system comprising hardware and a virtualization infrastructure that allocates the resources of the hardware to construct a virtual machine or container, A storage unit that stores a plurality of operating modes, each consisting of a combination of dynamically configurable items for the hardware, related to the power consumption of the hardware, and power characteristic data, including power efficiency and performance index values, obtained for each of the plurality of operating modes with different load rates for the hardware. A monitoring unit that monitors the hardware in operation and acquires metric data including the load rate of the hardware, An operating mode setting unit that, by referring to the power characteristics data, selects from the plurality of operating modes in order of the highest power efficiency at the load factor of the metrics data, and if the performance index value of the selected operating mode at that load factor satisfies the predetermined performance requirements of the hardware, operates the hardware using the selected operating mode. An operating mode setting device characterized by comprising the following features.
2. The operating mode setting unit, when the hardware starts operating, refers to the power characteristic data and selects the operating mode that has the highest power efficiency at a load factor below a predetermined value, and operates the hardware in the selected operating mode. The operating mode setting device according to claim 1, characterized by the following:
3. The operation mode setting unit, when the load rate obtained in the selected operation mode is lower than a predetermined threshold, instructs the virtualization infrastructure to perform scaling of the virtual machine or the container. The operating mode setting device according to claim 1, characterized by the following:
4. For each of the aforementioned multiple operating modes, the system includes a power characteristics data calculation unit that calculates power characteristics data, including the power efficiency and performance index values of the hardware, while varying the load rate of the hardware. The operating mode setting device according to claim 1, characterized by the following:
5. A method for setting the operating mode of an operating mode setting device provided in a system comprising hardware and a virtualization infrastructure that allocates the resources of the hardware to build a virtual machine or container, The aforementioned operating mode setting device is The system has a storage unit that stores a plurality of operating modes consisting of a combination of items that can be dynamically set for the hardware, in relation to the power consumption of the hardware, and power characteristic data including power efficiency and performance index values obtained for each of the plurality of operating modes with different load rates for the hardware. The steps include monitoring the hardware in operation and acquiring metric data including the load rate of the hardware, The steps include: referring to the power characteristics data, selecting from the plurality of operating modes in order of the highest power efficiency at the load factor of the metrics data, and operating the hardware in the selected operating mode if the performance index value at the load factor of the selected operating mode satisfies the predetermined performance requirements of the hardware; A method for setting an operating mode, characterized by performing the following actions.
6. An operating mode setting program for causing a computer to function as an operating mode setting device according to any one of claims 1 to 4.
7. Hardware and A virtualization platform that allocates the aforementioned hardware resources to build virtual machines or containers, A system comprising the virtualization infrastructure and an operating mode setting device provided as a controller for the hardware, The aforementioned operating mode setting device is A storage unit that stores a plurality of operating modes, each consisting of a combination of dynamically configurable items for the hardware, related to the power consumption of the hardware, and power characteristic data, including power efficiency and performance index values, obtained for each of the plurality of operating modes with different load rates for the hardware. A monitoring unit that monitors the hardware in operation and acquires metric data including the load rate of the hardware, An operating mode setting unit that, by referring to the power characteristics data, selects from the plurality of operating modes in order of the highest power efficiency at the load factor of the metrics data, and if the performance index value of the selected operating mode at that load factor satisfies the predetermined performance requirements of the hardware, operates the hardware using the selected operating mode. A system characterized by comprising the following features.
8. Hardware and A system comprising a virtualization platform that allocates the aforementioned hardware resources to build virtual machines or containers, The aforementioned hardware includes an operating mode setting device provided as a controller, The aforementioned operating mode setting device is A storage unit that stores power characteristic data, including power efficiency and performance index values, obtained for each of the above-mentioned operating modes, with different load rates for the hardware, and a plurality of operating modes consisting of a plurality of operating modes consisting of a plurality of operating modes consisting of a plurality of combinations of items that can be dynamically set for the hardware. A monitoring unit that monitors the hardware in operation and acquires metric data including the load rate of the hardware, An operating mode setting unit that, by referring to the power characteristics data, selects from the plurality of operating modes in order of the highest power efficiency at the load factor of the metrics data, and if the performance index value of the selected operating mode at that load factor satisfies the predetermined performance requirements of the hardware, operates the hardware using the selected operating mode. A system characterized by comprising the following features.