System and method for utilizing heat generated by running at least one virtual operating system instance
By configuring controllers in the data center to optimize the arrangement of virtual operating system instances on the host computer, the problem of low heat utilization efficiency in the prior art is solved, and efficient and safe heat recovery and hardware management are achieved.
Patent Information
- Application Number
- CN202480024909.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-02-10
- Filing Date
- 2024-02-12
- Publication Date
- 2025-11-07
AI Technical Summary
In data centers, when virtual operating system instances are orchestrated as visitors to host computers, existing technologies struggle to effectively utilize the heat generated by multiple host computers, especially during load changes, resulting in low thermal efficiency and an increased risk of hardware overheating.
By configuring one of at least two host computers as a controller, the heat dissipation rate of each virtual operating system instance is estimated, and the optimal arrangement is calculated to maximize heat generation. The operation of the host computers is controlled to achieve the target heat dissipation rate, including migration and energy-saving modes to optimize heat utilization.
It improves heat utilization efficiency, reduces the number of mainframes required, increases waste heat levels, ensures hardware safety, adapts to changes in computing resource demands, and achieves efficient heat recovery.
Smart Images

Figure CN120917430A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present invention relates to a system for utilizing heat generated by running at least one virtual operating system instance. The present invention also relates to a method for utilizing heat generated by running at least one virtual operating system instance. BACKGROUND
[0002] Sustainability is becoming an increasingly high priority in the data center industry nowadays. For example, some known developments aim at utilizing heat generated by computers running in data centers (also referred to as recovered heat), and then supplying the utilized heat to external facilities. The utilized heat can be used to heat buildings that need heat (e.g. apartment buildings, public swimming pool facilities) or food production processes (e.g. greenhouses).
[0003] There are various known solutions for utilizing heat from computers used in data centers. For example, known heat utilization systems are based on direct contact evaporative cooling systems (e.g. ZutaCore Hypercool 2TM ), in which at least one heat generating chip (e.g. a central processing unit chip) is equipped with a means for transferring heat from the chip to a coolant (e.g. 3M TM Novec TM 7000 engineering fluid). The heat generated by the heat transfer is then transferred to an external facility and used at the external facility. Other heat utilization systems are also known.
[0004] In this context, it is observed that when at least two host computers are used as heat sources for a heat utilization system, it is challenging to orchestrate instances of virtual operating systems as guests of the at least two host computers.
[0005] Running instances of virtual operating systems as guests of host computers allows the hardware resources of the host computers to be shared with the instances. Known methods for running instances of virtual operating systems include hypervisor software (e.g. KVM), which runs at least one virtual machine instance on a host computer, and container engine software (e.g. Docker), which runs at least one container instance on a host computer. Since the demand for computing resources by each instance can vary over time, the overall demand for hardware resources by the host computers also varies over time.
[0006] In the known art, "orchestration" refers to a method of monitoring and managing on which of at least two host computers each instance of a virtual operating system is run. Several factors can be taken into account when orchestrating (virtual machine or container) instances. Typically, the known orchestration methods aim at cooling the hardware as efficiently as possible, while achieving a uniform and low computational load on all host computers. The known orchestration methods can aim at balancing the server load in order to avoid the risk of hardware overheating that can lead to damage and / or malfunction. With respect to the consumption of hardware resources by the instances in the at least two host computers, the known orchestration methods typically achieve a low rate of heat utilization. SUMMARY
[0007] The invention will now be disclosed and the object of the invention is to remedy or reduce at least one of the drawbacks of the prior art, or to at least provide a useful alternative to the prior art. This object is achieved by the features specified in the following description and in the attached claims. The invention is defined by the independent claims. The dependent claims define advantageous embodiments of the invention.
[0008] According to a first aspect of the invention, there is provided a system for utilizing heat generated by running at least one virtual operating system instance. The system comprises: - at least two host computers for running at least one virtual operating system instance, the at least two host computers being networked; and - a subsystem for utilizing heat generated by the at least two host computers.
[0009] One or more of the at least two host computers are configured to run as a controller of the at least two host computers. The controller is configured to perform the following steps: - for each virtual operating system instance, estimate a heat dissipation rate that would be generated by running the instance as a guest of a host computer; - calculate an arrangement of the at least one virtual operating system instance on the at least two host computers such that the heat dissipation rate achieved by the at least two host computers is maximized; and - control the at least two host computers to run the at least one virtual operating system instance as defined in the calculated arrangement.
[0010] Thus, the system achieves a maximization of the heat dissipation rate generated by the at least two host computers in use, with respect to the number of virtual operating system instances that are running. Thus, the level of residual heat of each host computer is increased, which makes the at least two host computers more useful for the purpose of heat utilization. Furthermore, the system has the advantage of reducing the number of host computers needed for the same workload (caused by the virtual operating system instances).
[0011] Optionally, for each host computer, the controller is configured with a target minimum heat dissipation rate to be achieved by the host computer, and the step of computing the arrangement comprises the step of computing the arrangement of the at least one virtual operating system instance on the at least two host computers so that the number of host computers achieving the respective target minimum heat dissipation rate is maximised. Thus, when the host computers are controlled to run the virtual operating system instances, the system can be configured to target the target minimum heat dissipation rate to be achieved by the host computers. This embodiment is advantageous, for example, when the subsystem for utilising heat generated by the at least two host computers has an improved utilisation efficiency that depends on host computers generating heat above a defined amount (e.g. heat sufficient to cause evaporation of a coolant fluid). Thus, when computing the arrangement of the at least one virtual operating system instance on the at least two host computers, the system can optimise the computation based on the heat utilisation capacity of the subsystem. Thus, over time, a large amount of heat can be utilised, the utilised heat being available for external facilities.
[0012] Optionally, the step of estimating the heat dissipation rate comprises estimating the heat dissipation rate based on any of: - previous data regarding heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; - real-time data regarding heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; - data generated by a mathematical model, the mathematical model being a mathematical model of heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; and / or - data generated by a computer-implemented simulation, the computer-implemented simulation being a computer-implemented simulation of heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer.
[0013] Optionally, the step of controlling the at least two host computers to run the at least one virtual operating system instance comprises the step of migrating the virtual operating system instance from one host computer to a different host computer.
[0014] Optionally, the controller is further configured to perform any of: - shutting down a host computer that is not controlled to run the at least one virtual operating system instance; or - controlling a host computer that is not controlled to run the at least one virtual operating system instance to run in an energy saving mode. Thus, the number of resources that each host computer is idling is minimised.
[0015] Optionally, the at least one virtual operating system instance comprises any of: a virtual machine instance; and / or a container instance.
[0016] According to a second aspect of the present application, there is provided a method of utilizing heat generated by running at least one virtual operating system instance. The method comprises the steps of: - providing at least two host computers for running at least one virtual operating system instance, the at least two host computers being networked; - providing a subsystem for utilizing heat generated by the at least two host computers; - configuring one or more of the at least two host computers to run as a controller of the at least two host computers; - for each virtual operating system instance, estimating a heat dissipation rate that would be generated by running the instance as a guest of a host computer; - calculating an arrangement of the at least one virtual operating system instance on the at least two host computers such that the heat dissipation rate achieved by the at least two host computers is maximized; and - controlling the at least two host computers to run the at least one virtual operating system instance as defined in the calculated arrangement.
[0017] Optionally, the method comprises the steps of: - for each host computer, configuring the controller with a target minimum heat dissipation rate to be achieved by the host computer, and wherein the step of calculating an arrangement comprises the step of: - calculating an arrangement of the at least one virtual operating system instance on the at least two host computers such that the number of host computers achieving the respective target minimum heat dissipation rate is maximized.
[0018] Optionally, the step of estimating a heat dissipation rate comprises estimating the heat dissipation rate based on any of: - previous data regarding heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; - real-time data regarding heat dissipation rates generated by running virtual operating system instances as guests of host computers; - data generated by a mathematical model, the mathematical model being a mathematical model of heat dissipation rates generated by running virtual operating system instances as guests of host computers; and / or - data generated by a computer-implemented simulation, the computer-implemented simulation being a computer-implemented simulation of heat dissipation rates generated by running virtual operating system instances as guests of host computers.
[0019] Optionally, the step of controlling the at least two host computers to run the at least one virtual operating system instance comprises the step of migrating a virtual operating system instance from one host computer to a different host computer.
[0020] Optionally, the controller is further configured to perform any of: - shutting down a host computer that is not controlled to run at least one virtual operating system instance; or - controlling a host computer that is not controlled to run at least one virtual operating system instance to run in energy saving mode.
[0021] Optionally, the at least one virtual operating system instance comprises any of: a virtual machine instance; and / or a container instance.
[0022] According to a third aspect of the application, there is provided a system for utilizing heat produced by running at least one virtual operating system instance, the system comprising: - at least two host computers for running at least one virtual operating system instance, the at least two host computers being networked; - a subsystem for utilizing heat produced by the at least two host computers; and - a controller computer for controlling the at least two host computers. The controller computer is configured to perform the steps of: - for each virtual operating system instance, estimating a heat dissipation rate that would be produced by running the instance as a guest of a host computer; - calculating an arrangement of the at least one virtual operating system instance on the at least two host computers such that the heat dissipation rate achieved by the at least two host computers is maximized; and - controlling the at least two host computers to run the at least one virtual operating system instance in accordance with the arrangement defined in the calculation.
[0023] Optionally, for each host computer, the controller computer is configured with a target minimum heat dissipation rate to be achieved by the host computer, and the step of calculating an arrangement comprises the step of: - calculating an arrangement of the at least one virtual operating system instance on the at least two host computers such that the number of host computers achieving the respective target minimum heat dissipation rate is maximized.
[0024] Optionally, the step of estimating a heat dissipation rate comprises estimating a heat dissipation rate based on any of: - previous data regarding heat dissipation rates produced by running the virtual operating system instance as a guest of a host computer; - real-time data regarding heat dissipation rates produced by running virtual operating system instances as guests of host computers; - data generated by a mathematical model that is a mathematical model of heat dissipation rates produced by running virtual operating system instances as guests of host computers; and / or - data generated by a computer implemented simulation of the heat dissipation that would result from running a virtual operating system instance as a guest of a host computer.
[0025] Optionally, the step of controlling at least two host computers to run at least one virtual operating system instance comprises the step of migrating a virtual operating system instance from one host computer to a different host computer.
[0026] Optionally, the controller computer is further configured to perform any of the following steps: - shutting down a host computer that is not controlled to run at least one virtual operating system instance; or - controlling a host computer that is not controlled to run at least one virtual operating system instance to run in a power saving mode.
[0027] Optionally, the at least one virtual operating system instance comprises any of the following: a virtual machine instance; and / or a container instance.
[0028] According to a fourth aspect of the application, there is provided a method of utilising heat generated by running at least one virtual operating system instance, the method comprising the steps of: - providing at least two host computers for running the at least one virtual operating system instance, the at least two host computers being networked; - providing a subsystem for utilising heat generated by the at least two host computers; - providing a controller computer for controlling the at least two host computers; - for each virtual operating system instance, estimating, by the controller computer, a heat dissipation that would result from running that instance as a guest of a host computer; - calculating, by the controller computer, an arrangement of the at least one virtual operating system instance across the at least two host computers that maximises the heat dissipation achieved by the at least two host computers; and - controlling, by the controller computer, the at least two host computers to run the at least one virtual operating system instance in accordance with the arrangement defined in the calculated arrangement.
[0029] Optionally, the method comprises the step of: - for each host computer, configuring the controller computer with a target minimum heat dissipation that the host computer is to achieve, and wherein the step of calculating an arrangement comprises the step of: - calculating an arrangement of the at least one virtual operating system instance across the at least two host computers that maximises the number of host computers that achieve their respective target minimum heat dissipation.
[0030] Optionally, the step of estimating the heat rate comprises estimating the heat rate based on any of: - previous data relating to heat rates generated by running the virtual operating system instance as a guest of the host computer; - real-time data relating to heat rates generated by running the virtual operating system instance as a guest of the host computer; - data generated by a mathematical model, the mathematical model being a mathematical model of heat rates generated by running the virtual operating system instance as a guest of the host computer; and / or - data generated by a computer-implemented simulation, the computer-implemented simulation being a computer-implemented simulation of heat rates generated by running the virtual operating system instance as a guest of the host computer.
[0031] Optionally, the step of controlling at least two host computers to run at least one virtual operating system instance comprises the step of migrating a virtual operating system instance from one host computer to a different host computer.
[0032] Optionally, the method further comprises the steps of: - shutting down, by the controller computer, a host computer that is not controlled to run at least one virtual operating system instance; or - controlling, by the controller computer, a host computer that is not controlled to run at least one virtual operating system instance, to cause the host computer to run in a power saving mode.
[0033] Optionally, the at least one virtual operating system instance comprises any of: a virtual machine instance; and / or a container instance.
[0034] Accordingly, a system according to the first aspect of the application comprises one or more of the at least two host computers configured to run as a controller of the at least two host computers, while a system according to the third aspect of the application comprises a controller computer for controlling the at least two host computers. Similarly, a method according to the second aspect of the application comprises the step of configuring one or more of the at least two host computers to run as a controller of the at least two host computers, while a method according to the fourth aspect of the application comprises the step of providing a controller computer for controlling the at least two host computers. BRIEF DESCRIPTION OF DRAWINGS
[0035] In the drawings: Figure 1 A schematic diagram of a system comprising four host computers running four virtual operating system instances is shown; Figure 2A diagrammatic illustration is shown of a distribution of computing load on four host computers running virtual operating system instances defined by one arrangement; Figure 3 Another diagrammatic illustration is shown of a distribution of computing load on four host computers running virtual operating system instances defined by another arrangement. DETAILED DESCRIPTION
[0036] In the drawings, like or corresponding elements are denoted by like reference numerals. For the sake of clarity, in some of the drawings some elements can be without reference numerals. Those skilled in the art will understand that the drawings are merely schematic representations of major drawings. The relative scale of the individual elements can also be distorted.
[0037] Turning now to Figure 1 which shows a system embodiment 100 comprising four host computers 110a-110d for running four virtual operating system instances 200a-200d. The system 100 also comprises a subsystem 120 for utilizing heat generated by the four host computers 110a-110d. Those skilled in the art will understand that the system 100 shown in FIG. 1 is illustrated in a simplified manner for the purpose of illustrating embodiments of the present invention. In other embodiments, the number, structure, form, and / or organization of at least two host computers and heat utilization subsystems can vary significantly and within the range known in data center practice. Figure 1
[0038] The host computers 110a-110d can be embodied in a manner known from data center practice. For example, a data center can be provided in which a plurality of host computers are installed in various racks and connected to known systems available in a typical data center, such as power systems, networks, air cooling systems, etc. The host computers can be configured to operate software suitable for running virtual operating system instances, for example hypervisor software (e.g. KVM) or container engine software (e.g. Kubernetes).
[0039] The subsystem 120 for utilizing heat from the host computers 110a-110d can be provided in many known ways. For example, known subsystems are based on the use of two-phase coolants (e.g. 3M Novec 7100) in a closed loop system. The heat generated by the host computers 110a-110d is transferred to the two-phase coolant, which is then evaporated and condensed in a heat exchanger. The condensed coolant is then pumped back to the host computers 110a-110d. The heat exchanger can be cooled by air or water. TM Novec TM 7000 engineered fluid) having a predetermined boiling temperature. The utilization of heat is achieved by passing the liquid phase coolant into direct contact with one or more chips of the host computer 110a-110d. When this contact occurs and the chips generate sufficient heat to cause the coolant to boil, the coolant evaporates and flows in gaseous form to the sub-system 120. At this stage, the gaseous hot coolant is used as a source of heat to be utilized. The heat is utilized and transferred to an external facility for re-utilization. The utilization of heat from the coolant reduces the temperature of the coolant, which in turn causes the gaseous coolant to revert back to a liquid phase. The utilization of heat from the hot gaseous coolant can be achieved in many known ways, such as by a heat exchanger device in which the heat from the coolant is transferred to a separate circuit comprising a fluid, such as water, for transporting the utilized heat to an external facility. It should be appreciated that other sub-systems 120 for utilizing heat are known and can be used to utilize heat from one or more host computers.
[0040] In the system 100, Figure 1 The host computer 110a shown in the upper left corner is additionally configured to operate as a controller for controlling the operation of the four virtual operating system instances 200a-200d on the four host computers 110a-110d. It should be appreciated that the controller can be implemented in other known ways, including having the controller operate in a cooperative or decentralized manner by two or more host computers, as well as other known approaches. It should also be appreciated that the host computers 110a-110d can be configured to operate both as a controller and to operate one or more virtual operating system instances. Further, it should be appreciated that in other embodiments, the system 100 can include a controller computer (not shown) for controlling the host computers 110a-110d, rather than having one or more of the host computers 110a-110d configured to operate as a controller.
[0041] In Figure 1In the illustrated embodiment, the controller has controlled the four host computers 110a-110d to run four virtual operating system instances 200a-200d as shown: the host computer 110a in the upper left corner is controlled to run two instances 200a, 200b; the host computer 110b in the lower left corner is controlled to run one instance 200c; and the host computer 110c in the upper right corner is controlled to run one instance 200d. The arrangement of the four virtual operating system instances 200a-200d on the four host computers 110a-110d is such that the heat dissipation achieved by the four host computers 110a-110d is maximized. In calculating the arrangement, the controller performs the following steps: for each virtual operating system instance 200a-200d, it estimates the heat dissipation that would result from running that instance as a guest of one of the host computers 110a-110d. This estimation of heat dissipation can be implemented in many known ways. For example, the estimation can involve estimating the percentage of time that the temperature of a chip (such as a CPU) in the host computer is above some minimum temperature. Or, for example, the estimation can be implemented by estimating the heat dissipation in kWh. It will be appreciated that other known ways of estimating heat dissipation can be used.
[0042] In some embodiments, the heat dissipation estimation is based on any of: - prior data about heat dissipation resulting from running the virtual operating system instance as a guest of a host computer; - real-time data about heat dissipation resulting from running the virtual operating system instance as a guest of a host computer; - data generated by a mathematical model that is a mathematical model of heat dissipation resulting from running the virtual operating system instance as a guest of a host computer; and / or - data generated by a computer-implemented simulation that is a computer-implemented simulation of heat dissipation resulting from running the virtual operating system instance as a guest of a host computer.
[0043] These data options can be selected in a number of ways, either exclusively or in combination. Furthermore, these data options can be combined with other known data options.
[0044] After the estimation step, the controller performs the step of calculating the arrangement. Those skilled in the art will know a number of search algorithms for efficiently finding an arrangement in which the heat dissipation achieved by the four host computers 110a-110d is maximized. Furthermore, it will be appreciated that those skilled in the art will know how a number of heat dissipation estimates can be accumulated to represent a number of respective instances running as guests of the same host computer.
[0045] For example, in the illustrated embodiment, the controller has controlled the host computer 110a to run two instances 200a, 200b, and the host computer 110b to run one instance 200c. The controller has also controlled the host computer 110c to run one instance 200d. The arrangement of the four virtual operating system instances 200a-200d on the four host computers 110a-110d is such that the heat dissipation achieved by the four host computers 110a-110d is maximized. Figure 1In this particular embodiment, the host computers 110b, 110c in the lower left and upper right corners are controlled to each run only one instance 200c, 200d of the virtual operating system. This distribution of this part of the arrangement is created by estimating that each of the instances 200c, 200d will cause the host computer to generate a sufficiently high heat rate to justify running only one instance in each of the two host computers 110b, 110c.
[0046] For the system embodiment, Figure 2 A diagram is shown illustrating the distribution of the computational load 110a'-110d' over the four host computers running the instances of the virtual operating system computed in one arrangement. As in the previous diagram, Figure 2 The two host computers associated with the computational load 110a', 110b' on the left-hand side of the diagram have been allocated all the instances to run, while the two host computers associated with the computational load 110c', 110d' on the right-hand side of the diagram have not been allocated to run any instances. Furthermore, the latter host computers, i.e. the host computers associated with the computational load 110c', 110d' on the right-hand side, have been turned off in order to reduce power consumption. Thus, the utilization of the heat generated by the four host computers is maximized. Figure 2
[0047] For the system embodiment, Figure 3 Another diagram is shown illustrating the distribution of the computational load 110a"-110d" over the four host computers running the instances of the virtual operating system computed in another arrangement.
[0048] In Figure 3 For each of the four host computers, the controller of the system embodiment is configured with a target minimum heat rate to be reached by the host computer. These four target minimum heat rates are illustrated in Figure 3 In this particular embodiment, the controller is configured with the same target minimum heat rate for each of the four host computers, as indicated by the horizontal dashed lines intersecting the computational loads 110a"-110d". It should be understood that different target minimum heat rates can be configured for each host computer, and that such configuration can take into account various parameters related to the heat generating capacity of the host computers.
[0049] When computing the arrangement of at least one virtual operating system instance over the four host computers, the controller is configured to maximize the number of host computers reaching the respective target minimum heat rate. As Figure 3 indicated, all host computers running instances are reaching a computational load 110a"-110c" higher than the target minimum heat rate, even though some of them can still have the capacity to run further instances.
[0050] It should be noted that the above-mentioned embodiments illustrate rather than limit the application, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word 'comprising' does not exclude the presence of elements or steps other than those listed in a claim. The word 'a' or 'an' preceding an element does not exclude the presence of a plurality of such elements.
[0051] The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.
[0052] The application can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the system claims enumerating several means, several of these means can be embodied by one and the same item of hardware.
Claims
1. A system for utilizing heat generated by running at least one virtual operating system instance, the system comprising: - at least two host computers for running the at least one virtual operating system instance, the at least two host computers being networked; and - a subsystem for utilizing heat generated by the at least two host computers, wherein one or more of the at least two host computers are configured to run as a controller of the at least two host computers, and wherein the controller is configured to perform the following steps: - for each virtual operating system instance, estimate a heat dissipation rate that would be generated by running the instance as a guest of a host computer; - compute an arrangement of the at least one virtual operating system instance on the at least two host computers such that a heat dissipation rate achieved by the at least two host computers is maximized; and - control the at least two host computers to run the at least one virtual operating system instance in accordance with the arrangement as defined by the computation.
2. The system of claim 1, wherein, For each host computer, the controller is configured with a target minimum heat dissipation rate to be achieved by the host computer, and wherein the step of computing an arrangement comprises the step of: - computing an arrangement of the at least one virtual operating system instance on the at least two host computers such that a number of host computers achieving a respective target minimum heat dissipation rate is maximized.
3. The system of any of the preceding claims, wherein, The step of estimating a heat dissipation rate comprises estimating the heat dissipation rate based on any of: - previous data about heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; - real-time data about heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; - data generated by a mathematical model, the mathematical model being a mathematical model of heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; and / or - data generated by a computer-implemented simulation, the computer-implemented simulation being a computer-implemented simulation of heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer.
4. The system according to any of the preceding claims, wherein the step of controlling the at least two host computers to run the at least one virtual operating system instance comprises the step of migrating a virtual operating system instance from one host computer to a different host computer.
5. The system according to any of the preceding claims, wherein the controller is further configured to perform any of the following steps: - shutting down a host computer that is not controlled to run at least one virtual operating system instance; or - controlling a host computer that is not controlled to run at least one virtual operating system instance to run in an energy saving mode.
6. The system according to any of the preceding claims, wherein the at least one virtual operating system instance comprises any of: a virtual machine instance; and / or a container instance.
7. A method of utilizing heat generated by running at least one virtual operating system instance, the method comprising the steps of: - providing at least two host computers for running said at least one virtual operating system instance, said at least two host computers being networked; - providing a subsystem for utilizing heat generated by said at least two host computers; - configuring one or more of said at least two host computers to run as a controller of said at least two host computers; - for each virtual operating system instance, estimating a heat dissipation rate that would be generated by running said instance as a guest of a host computer; - computing an arrangement of said at least one virtual operating system instance on said at least two host computers such that a heat dissipation rate achieved by said at least two host computers is maximized; and - controlling said at least two host computers to run said at least one virtual operating system instance in accordance with the arrangement as defined by the computation.
8. The method according to claim 7, comprising the steps of: - for each host computer, configuring said controller with a target minimum heat dissipation rate to be achieved by said host computer, and wherein the step of computing an arrangement comprises the step of: - computing an arrangement of said at least one virtual operating system instance on said at least two host computers such that a number of host computers achieving a respective target minimum heat dissipation rate is maximized.
9. The method according to any one of claims 7 to 8, wherein the step of estimating a heat dissipation rate comprises estimating said heat dissipation rate based on any one of: - previous data on heat dissipation rates generated by running said virtual operating system instance as a guest of said host computer, - real-time data on heat dissipation rates generated by running said virtual operating system instance as a guest of a host computer; - data generated by a mathematical model, said mathematical model being a mathematical model of heat dissipation rates generated by running said virtual operating system instance as a guest of a host computer; and / or - data generated by a computer-implemented simulation, said computer-implemented simulation being a computer-implemented simulation of heat dissipation rates generated by running said virtual operating system instance as a guest of a host computer.
10. The method according to any one of claims 7 to 9, wherein the step of controlling said at least two host computers to run said at least one virtual operating system instance comprises the step of migrating a virtual operating system instance from one host computer to a different host computer.
11. The method according to any one of claims 7 to 10, wherein said controller is further configured to perform any one of the following steps: - shutting down a host computer that is not controlled to run at least one virtual operating system instance; or - controlling a host computer that is not controlled to run at least one virtual operating system instance to run in a power saving mode.
12. The method according to any one of claims 7 to 11, wherein said at least one virtual operating system instance comprises any one of: a virtual machine instance; and / or a container instance.
13. A system for utilizing heat generated by running at least one virtual operating system instance, the system comprising: - at least two host computers for running said at least one virtual operating system instance, said at least two host computers being networked; - a subsystem for utilizing heat generated by said at least two host computers; and - a controller computer for controlling said at least two host computers, and wherein said controller computer is configured to perform the following steps: - for each virtual operating system instance, estimating a heat dissipation rate that would be achieved by running said instance as a guest of a host computer; - computing an arrangement of said at least one virtual operating system instance on said at least two host computers such that the heat dissipation rate achieved by said at least two host computers is maximized; and - controlling said at least two host computers to run said at least one virtual operating system instance in accordance with the arrangement as defined by the computation.
14. The system of claim 13, wherein, For each host computer, said controller computer is configured with a target minimum heat dissipation rate to be achieved by said host computer, and wherein the step of computing an arrangement comprises the step of: - computing an arrangement of said at least one virtual operating system instance on said at least two host computers such that the number of host computers achieving a respective target minimum heat dissipation rate is maximized.
15. The system of any one of claims 13-14, wherein, The step of estimating a heat dissipation rate comprises estimating said heat dissipation rate based on any of: - previous data regarding heat dissipation rates achieved by running said virtual operating system instance as a guest of a host computer; - real-time data regarding heat dissipation rates achieved by running said virtual operating system instance as a guest of a host computer; - data generated by a mathematical model, said mathematical model being a mathematical model of heat dissipation rates achieved by running said virtual operating system instance as a guest of a host computer; and / or - data generated by a computer-implemented simulation, said computer-implemented simulation being a computer-implemented simulation of heat dissipation rates achieved by running said virtual operating system instance as a guest of a host computer.
16. The system of any one of claims 13 to 15, wherein the step of controlling said at least two host computers to run said at least one virtual operating system instance comprises the step of migrating a virtual operating system instance from one host computer to a different host computer.
17. The system of any one of claims 13 to 16, wherein said controller computer is further configured to perform any one of the following steps: - shutting down a host computer that is not controlled to run at least one virtual operating system instance; or - controlling a host computer that is not controlled to run at least one virtual operating system instance to run in a power saving mode.
18. The system of any one of claims 13 to 17, wherein said at least one virtual operating system instance comprises any one of: a virtual machine instance; and / or a container instance.
19. A method of utilizing heat generated by running at least one virtual operating system instance, said method comprising the steps of: - providing at least two host computers for running said at least one virtual operating system instance, said at least two host computers being networked; - providing a subsystem for utilizing heat generated by the at least two host computers; - providing a controller computer for controlling the at least two host computers; - for each virtual operating system instance, estimating, by the controller computer, a heat dissipation rate that would be generated by running the instance as a guest of a host computer; - computing, by the controller computer, an arrangement of the at least one virtual operating system instance on the at least two host computers such that the heat dissipation rate achieved by the at least two host computers is maximized; and - controlling, by the controller computer, the at least two host computers to run the at least one virtual operating system instance as defined by the computed arrangement.
20. The method of claim 19, comprising the steps of: - for each host computer, configuring the controller computer with a target minimum heat dissipation rate that the host computer is to achieve, and wherein the step of computing an arrangement comprises the step of: - computing an arrangement of the at least one virtual operating system instance on the at least two host computers such that the number of host computers that achieve the respective target minimum heat dissipation rate is maximized.
21. The method of any one of claims 19 to 20, wherein the step of estimating a heat dissipation rate comprises estimating the heat dissipation rate based on any one of: - previous data regarding heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; - real-time data regarding heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; - data generated by a mathematical model that is a mathematical model of heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer; and / or - data generated by a computer-implemented simulation that is a computer-implemented simulation of heat dissipation rates generated by running the virtual operating system instance as a guest of a host computer.
22. The method of any one of claims 19 to 21, wherein the step of controlling the at least two host computers to run the at least one virtual operating system instance comprises the step of migrating a virtual operating system instance from one host computer to a different host computer.
23. The method of any one of claims 19 to 22, wherein the method further comprises the steps of: - shutting down, by the controller computer, a host computer that is not controlled to run at least one virtual operating system instance; or - controlling, by the controller computer, a host computer that is not controlled to run at least one virtual operating system instance to cause the host computer to run in a power saving mode.
24. The method of any one of claims 19 to 23, wherein the at least one virtual operating system instance comprises any one of: a virtual machine instance; and / or a container instance.