Virtual machine migration method, device, equipment, medium and product

Abstract

The application provides a virtual machine migration method, device, equipment, medium and product, and relates to the technical field of computers, to solve the defects that the effect of virtual machine migration in the prior art is poor and the migration success rate is low, improve the effect of virtual machine migration, and improve the migration success rate. It comprises the following steps: obtaining the running parameters of each virtual machine running on each physical server in a plurality of physical servers in a prediction time period based on a pre-constructed average running parameter model in a current time period; determining the resource occupancy rate of each physical server based on the running parameters of at least one virtual machine running on each physical server; determining a source server from the plurality of physical servers based on the resource occupancy rate of each physical server, and the resource occupancy rate of the source server is greater than a preset resource occupancy rate; migrating a target virtual machine running on the source server to a target server before the end of the current time period.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a virtual machine migration method, apparatus, device, medium, and product. Background Technology

[0002] In server energy-saving technologies, virtualization is a common approach, including virtualization consolidation. Virtualization consolidation uses virtualization technology to run multiple virtual servers (virtual machines) on a single physical server, reducing the number of physical servers and thus lowering energy consumption. Alternatively, dynamic migration can be used to dynamically migrate virtual machines based on the load of physical servers, merging multiple low-load virtual machines onto a few physical servers while allowing other physical servers to enter power-saving or hibernation states.

[0003] However, virtual machine migration often only begins after the virtual machine's usage frequency has increased to the point that the physical server's Central Processing Unit (CPU) and memory usage have exceeded thresholds. This leads to sluggish performance on the physical server once CPU and memory usage reach their limits. Furthermore, when the physical server is sluggish, the migration speed decreases, increasing the likelihood of migration failures and negatively impacting the user experience. Therefore, current virtual machine migration methods are ineffective and have a low success rate. Summary of the Invention

[0004] This invention provides a virtual machine migration method, apparatus, device, medium, and product to address the shortcomings of existing technologies in migrating virtual machines, such as poor performance and low success rate. The invention aims to improve the effectiveness of virtual machine migration, increase the success rate, and enhance the user experience.

[0005] This invention provides a virtual machine migration method, comprising the following steps.

[0006] Based on a pre-built average operating parameter model, obtain the operating parameters of each virtual machine running on each of multiple physical servers in the current time period. Each physical server runs at least one virtual machine, and the predicted time period is the next time period after the current time period. Based on the operating parameters of at least one virtual machine running on each physical server, determine the resource utilization of each physical server; Based on the resource utilization rate of each physical server, the source server is determined from multiple physical servers, and the resource utilization rate of the source server is greater than the preset resource utilization rate. Before the end of the current time period, migrate the target virtual machine running on the source server to the target server. After the target virtual machine is migrated, the resource utilization of both the source server and the target server is less than or equal to the preset resource utilization.

[0007] According to a virtual machine migration method provided by the present invention, a pre-built average operating parameter model is constructed in the following manner: Divide the total daily duration into multiple equal-length time periods; Retrieve the historical running parameters of each virtual machine running on each physical server for each time period, spanning multiple days prior to the current time period; For each time period of the day, determine the average operating parameters of each virtual machine over multiple days; An average operating parameter model is constructed based on the average operating parameters of each virtual machine in each time period.

[0008] According to a virtual machine migration method provided by the present invention, the operating parameters include at least one of the following: CPU utilization rate and memory utilization rate.

[0009] According to a virtual machine migration method provided by the present invention, a source server is determined from multiple physical servers based on the resource utilization rate of each physical server, including: Based on the resource utilization of each physical server, multiple physical servers are divided into multiple server groups. The multiple server groups include: Level 1 performance server group, Level 2 performance server group and Level 3 performance server group. The resource utilization of physical servers in the Level 1 performance server group is greater than that of physical servers in the Level 2 performance server group, and the resource utilization of physical servers in the Level 2 performance server group is greater than that of physical servers in the Level 3 performance server group. For each of the multiple server groups, the physical server with a resource utilization rate greater than the preset resource utilization rate is identified as the source server.

[0010] According to a virtual machine migration method provided by the present invention, before the end of the current time period, a target virtual machine running on a source server is migrated to a target server, comprising: For the source server in the first-level performance server group, the virtual machine with the lowest running parameters among the virtual machines running on the source server is identified as the target virtual machine; The system iterates through multiple physical servers in descending order of resource utilization, and identifies the first physical server that meets the target conditions as the target server. The target conditions include that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0011] According to a virtual machine migration method provided by the present invention, before the end of the current time period, a target virtual machine running on a source server is migrated to a target server, comprising: For the source server in the secondary performance server group, all virtual machines running on the source server are identified as target virtual machines; The physical servers in the Level 2 and Level 3 performance server groups are traversed sequentially from highest to lowest resource utilization. The first physical server that meets the target condition is identified as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0012] According to a virtual machine migration method provided by the present invention, before the end of the current time period, a target virtual machine running on a source server is migrated to a target server, comprising: For the source server in the Tier 3 performance server group, all virtual machines running on the source server are identified as target virtual machines; The physical servers in the three-level performance server group are traversed in descending order of resource utilization. The first physical server that meets the target condition is identified as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0013] The present invention also provides a virtual machine migration device, comprising the following modules: an acquisition module, a processing module, and a migration module; The acquisition module is used to acquire the operating parameters of each virtual machine running on each of multiple physical servers in the predicted time period based on a pre-built average operating parameter model. Each physical server has at least one virtual machine running, and the predicted time period is the next time period after the current time period. The processing module is used to determine the resource utilization of each physical server based on the running parameters of at least one virtual machine running on each physical server. The processing module is also used to determine the source server from multiple physical servers based on the resource utilization rate of each physical server, wherein the resource utilization rate of the source server is greater than the preset resource utilization rate. The migration module is used to migrate the target virtual machine running on the source server to the target server before the end of the current time period. After the target virtual machine is migrated, the resource utilization of both the source server and the target server is less than or equal to the preset resource utilization.

[0014] According to the virtual machine migration apparatus provided by the present invention, the processing module is further configured to divide the total duration of each day into multiple time periods of equal length. The acquisition module is also used to acquire the historical running parameters of each virtual machine running on each physical server in each time period over several days prior to the current time period; The processing module is also used to determine the average operating parameters of each virtual machine over multiple days for each time period of the day; The processing module is also used to build an average operating parameter model based on the average operating parameters of each virtual machine in each time period.

[0015] According to the present invention, a virtual machine migration device has operating parameters including at least one of the following: CPU utilization rate and memory utilization rate.

[0016] According to the present invention, a virtual machine migration apparatus, including a processing module, is specifically used for: Based on the resource utilization of each physical server, multiple physical servers are divided into multiple server groups. The multiple server groups include: Level 1 performance server group, Level 2 performance server group and Level 3 performance server group. The resource utilization of physical servers in the Level 1 performance server group is greater than that of physical servers in the Level 2 performance server group, and the resource utilization of physical servers in the Level 2 performance server group is greater than that of physical servers in the Level 3 performance server group. For each of the multiple server groups, the physical server with a resource utilization rate greater than the preset resource utilization rate is identified as the source server.

[0017] According to the present invention, a virtual machine migration device, including a migration module, is specifically used for: For the source server in the first-level performance server group, the virtual machine with the lowest running parameters among the virtual machines running on the source server is identified as the target virtual machine; The system iterates through multiple physical servers in descending order of resource utilization, and identifies the first physical server that meets the target conditions as the target server. The target conditions include that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0018] According to the present invention, a virtual machine migration device, including a migration module, is specifically used for: For the source server in the secondary performance server group, all virtual machines running on the source server are identified as target virtual machines; The physical servers in the Level 2 and Level 3 performance server groups are traversed sequentially from highest to lowest resource utilization. The first physical server that meets the target condition is identified as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0019] According to the present invention, a virtual machine migration device, including a migration module, is specifically used for: For the source server in the Tier 3 performance server group, all virtual machines running on the source server are identified as target virtual machines; The physical servers in the three-level performance server group are traversed in descending order of resource utilization. The first physical server that meets the target condition is identified as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0020] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the virtual machine migration methods described above.

[0021] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the virtual machine migration methods described above.

[0022] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the virtual machine migration methods described above.

[0023] This invention provides a virtual machine migration method, apparatus, device, medium, and product. By obtaining the operating parameters of each virtual machine running on each of multiple physical servers within a predicted time period based on a pre-built average operating parameter model, the resource utilization rate of each physical server can be determined based on the operating parameters of at least one virtual machine running on each physical server. Thus, based on the resource utilization rate of each physical server, a source server with a resource utilization rate greater than a preset resource utilization rate is identified from the multiple physical servers. Before the end of the current time period, the target virtual machine running on the source server is migrated to the target server. This ensures that after the target virtual machine is migrated, the resource utilization rates of both the source server and the target server are less than or equal to the preset resource utilization rate. Therefore, by determining the resource utilization rate of physical servers in advance and migrating virtual machines ahead of time, the migration of virtual machines can be avoided after the resource utilization rate of physical servers becomes too high, improving the efficiency and success rate of virtual machine migration. Furthermore, by migrating all virtual machines running on a physical server to other physical servers and shutting down idle physical servers, energy saving is improved. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1 This is one of the flowcharts illustrating the virtual machine migration method provided by the present invention.

[0026] Figure 2 This is the second flowchart of the virtual machine migration method provided by the present invention.

[0027] Figure 3 This is the third flowchart of the virtual machine migration method provided by the present invention.

[0028] Figure 4 This is the fourth flowchart of the virtual machine migration method provided by the present invention.

[0029] Figure 5 This is the fifth flowchart of the virtual machine migration method provided by the present invention.

[0030] Figure 6 This is the sixth flowchart of the virtual machine migration method provided by the present invention.

[0031] Figure 7This is a schematic diagram of the virtual machine migration system provided by the present invention.

[0032] Figure 8 This is a schematic diagram of the virtual machine migration device provided by the present invention.

[0033] Figure 9 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0035] Currently, there are limitations to using dynamic migration technology to achieve energy savings on physical servers. Existing technology cannot predict virtual machine migration in advance, and virtual machine memory and CPU usage often vary throughout the day due to human activity patterns. During peak human activity periods, virtual machines are used more frequently and by more users, resulting in higher usage rates; conversely, during off-peak periods, both frequency and user activity are lower, leading to lower usage rates. Current technology cannot predict and plan migrations based on virtual machine usage patterns; instead, it performs migrations in real-time. Often, virtual machine usage increases until the physical server's memory and CPU usage exceed thresholds before migration begins. Common sense tells us that once CPU and memory usage reach their upper limits, performance stagnates, reducing the dynamic migration rate and increasing the likelihood of migration failures, significantly degrading the user experience.

[0036] To address this, the present invention proposes a virtual machine migration system, comprising an energy-saving management module, a virtual machine scheduling module, and multiple physical servers. The energy-saving management module is used to obtain the operating parameters of the virtual machines running on the physical servers, and predicts the dynamic migration strategy of the virtual machines based on the operating parameters. The virtual machine scheduling module is used to migrate the virtual machines.

[0037] The following is combined with Figures 1 to 9 This invention describes the virtual machine migration method, apparatus, device, medium, and product provided in embodiments of the present invention.

[0038] Figure 1 This is one of the flowcharts illustrating the virtual machine migration method provided by the present invention, such as... Figure 1 As shown, the method includes the following: Step 101: Based on the pre-built average operating parameter model, obtain the operating parameters of each virtual machine running on each of the multiple physical servers during the predicted time period.

[0039] In this case, at least one virtual machine is running on a physical server, and the predicted time period is the next time period after the current time period.

[0040] Step 102: Determine the resource utilization rate of each physical server based on the running parameters of at least one virtual machine running on each physical server.

[0041] Step 103: Determine the source server from multiple physical servers based on the resource utilization rate of each physical server.

[0042] Among them, the resource utilization rate of the source server is greater than the preset resource utilization rate.

[0043] Step 104: Before the end of the current time period, migrate the target virtual machine running on the source server to the target server.

[0044] Among them, after the target virtual machine is migrated, the resource utilization rate of both the source server and the target server is less than or equal to the preset resource utilization rate.

[0045] In one possible implementation, the energy-saving management module can predict the dynamic migration strategy of virtual machines based on the operating parameters of all virtual machines (i.e., each virtual machine running on each physical server in multiple physical servers), so that the virtual machine scheduling module can use the dynamic migration strategy to perform dynamic migration of virtual machines.

[0046] In one possible implementation, the operating parameters include at least one of the following: CPU utilization and memory utilization.

[0047] In one possible implementation, it is first necessary to determine the time period in which the current time is located based on the current time, and denot it as the current time period. The next time period after the current time period is then denoted as the predicted time period.

[0048] For example, assuming the current time is 13:45, based on the time period comparison table shown in Table 1, the current time period can be determined to be time period 5 (i.e., 12:00-15:00), and the predicted time period is time period 6 (i.e., 15:00-18:00).

[0049] Furthermore, based on a pre-built average operating parameter model, the operating parameters of all virtual machines running on each physical server within the current time period can be retrieved from all virtual machines on multiple physical servers within the current time period, corresponding to the predicted time period.

[0050] In one possible implementation, the virtual machine's operating parameters may include: average CPU utilization and average memory utilization. The average CPU utilization represents the average of the Q CPU utilization values ​​corresponding to each time period within a certain duration (e.g., Q days); the average memory utilization represents the average of the Q memory utilization values ​​corresponding to each time period within a certain duration (e.g., Q days).

[0051] In one possible implementation, the Q-day corresponding to a certain duration can be 5, 8, 10, or 15 days. The range of values ​​for Q is not specifically limited and can be determined based on the actual solution.

[0052] Specifically, for each physical server, the average CPU utilization during the predicted time period m can be retrieved from the runtime parameters of each virtual machine running on that physical server. The CPU utilization (CPUS) of the physical server during the predicted time period m is obtained by summing these values. Additionally, the average memory usage during the predicted time period m is calculated from the runtime parameters of each virtual machine running on that physical server. Add them together to get the memory utilization rate (NCS) of the physical server during the predicted time period m.

[0053] That is, the resource utilization of each physical server includes: the CPU utilization (CPUS) of the physical server in each time period and the memory utilization (NCS) of the physical server in each time period.

[0054] In this way, based on the determined resource utilization of each physical server, a dynamic migration strategy for virtual machines running on multiple physical servers can be planned for the predicted time period. Furthermore, before the end of the current time period, the virtual machine migration and physical server power-off operations are performed according to the planned dynamic migration strategy.

[0055] It should be noted that the source server is the physical server on which the target virtual machine is located before migration, and the target server is the physical server on which the target virtual machine is located after migration. If the target virtual machine is running on the source server during the prediction period, the resource utilization of the source server will be greater than the preset resource utilization. Conversely, if the target virtual machine is running on the target server during the prediction period, the resource utilization of both the source and target servers will not exceed the preset resource utilization.

[0056] In one possible implementation, if all virtual machines running on the source server are used as target virtual machines, that is, if all virtual machines running on the source server are migrated to the target server, then the source server becomes an idle physical server and can be shut down by power-off to reduce energy consumption.

[0057] In this embodiment, by obtaining the operating parameters of each virtual machine running on each of multiple physical servers during the predicted time period based on a pre-built average operating parameter model, the resource utilization rate of each physical server can be determined based on the operating parameters of at least one virtual machine running on each physical server. Thus, based on the resource utilization rate of each physical server, a source server with a resource utilization rate greater than a preset resource utilization rate is identified from the multiple physical servers. Before the end of the current time period, the target virtual machine running on the source server is migrated to the target server. This ensures that after the target virtual machine is migrated, the resource utilization rates of both the source server and the target server are less than or equal to the preset resource utilization rate. Therefore, by determining the resource utilization rate of physical servers in advance and migrating virtual machines ahead of time, the migration of virtual machines can be avoided after the resource utilization rate of physical servers becomes too high, improving the efficiency and success rate of virtual machine migration. Furthermore, by migrating all virtual machines running on a physical server to other physical servers and shutting down idle physical servers, energy saving is improved.

[0058] Figure 2 This is the second flowchart illustrating the virtual machine migration method provided by this invention, as shown below. Figure 2 As shown, the pre-built average operating parameter model is constructed in the following way: Step 201: Divide the total duration of each day into multiple equal time periods.

[0059] Step 202: Obtain the historical running parameters of each virtual machine running on each physical server for each time period within the previous several days.

[0060] Step 203: For each time period of the day, determine the average operating parameters of each virtual machine over multiple days.

[0061] Step 204: Construct an average operating parameter model based on the average operating parameters of each virtual machine in each time period.

[0062] In one possible implementation, the energy-saving management module can collect the operating parameters of each virtual machine running on each physical server in real time to formulate and store an average operating parameter model for the virtual machines. Operating parameters include the CPU utilization and memory utilization of the virtual machines during runtime.

[0063] In one possible implementation, the day can be divided into several equal-length time periods according to settings. The specific division can be determined based on actual settings.

[0064] It should be noted that a day is 24 hours long and can be divided into 8 equal time periods. The 8 time periods correspond to the daily time as shown in Table 1. Table 1 is a time period comparison table.

[0065] Table 1

[0066] Alternatively, the day can be divided into 6 equal time periods, and the corresponding time periods for each day are shown in Table 2.

[0067] Table 2

[0068] In one possible implementation, the running parameters of the virtual machines are collected, and the historical running parameters of each virtual machine running on each physical server in each time period are obtained over several days prior to the current time period. The average running parameter model of each virtual machine is then calculated, and the expression of the average running parameter model is shown in Formula 1.

[0069] Formula 1 in, This represents the average operating parameter model of the nth virtual machine. This represents the running parameters of the nth virtual machine during time period m. This represents the average CPU utilization of the nth virtual machine within the time period m. This represents the average memory usage of the nth virtual machine within the time period m.

[0070] In one possible implementation, average CPU utilization The calculation process includes: obtaining the CPU utilization rate of virtual machine n within the time period m over Q consecutive days, and calculating the average CPU utilization rate based on the CPU utilization rate of virtual machine n within the time period m over Q days. The calculation formula is shown in Formula 2.

[0071] Formula 2 in, This represents the CPU utilization of virtual machine n on day w within a time period m in a continuous Q-day cycle.

[0072] For example, Q is set to 5, and the CPU utilization of virtual machine 1 in time period 1 is obtained for 5 consecutive days. The obtained data is shown in Table 3.

[0073] Table 3

[0074] Based on average CPU utilization Formula 2 can be used to calculate Formula 3.

[0075] Formula 3 This indicates that virtual machine 1's average CPU utilization during time period 1. It is 5.6.

[0076] In one possible implementation, average memory usage The calculation process includes: obtaining the memory usage rate of virtual machine n within a time period m over Q consecutive days, and calculating the average memory usage rate based on the memory usage rate of virtual machine n within a time period m over Q days. The calculation formula is shown in Formula 4.

[0077] Formula 4 in, This represents the memory usage of virtual machine n on day w within a time period m in a continuous Q-day cycle.

[0078] For example, Q is set to 5, and the memory usage of virtual machine 1 in time period 1 is obtained for 5 consecutive days. The obtained data is shown in Table 4.

[0079] Table 4

[0080] Based on average memory usage Formula four can be used to calculate formula five.

[0081] Formula 5 This indicates that virtual machine 1's average memory usage during time period 1. It is 14.8.

[0082] In this embodiment, the total duration of each day is divided into multiple equal-length time periods. Then, the historical operating parameters of each virtual machine running on each physical server for each time period are obtained from the previous several days. Based on the obtained historical operating data, the average operating parameters of each virtual machine over the several days are determined. In this way, an accurate average operating parameter model can be constructed based on the average operating parameters of each virtual machine in each time period.

[0083] Figure 3 This is the third flowchart of the virtual machine migration method provided by the present invention, as shown below. Figure 3 As shown, the method includes the following: Step 301: Based on the pre-built average operating parameter model, obtain the operating parameters of each virtual machine running on each of the multiple physical servers during the predicted time period.

[0084] Step 302: Determine the resource utilization rate of each physical server based on the running parameters of at least one virtual machine running on each physical server.

[0085] Step 303: Based on the resource utilization rate of each physical server, divide the multiple physical servers into multiple server groups.

[0086] The server groups include: a Tier 1 performance server group, a Tier 2 performance server group, and a Tier 3 performance server group. The resource utilization rate of the physical servers in the Tier 1 performance server group is greater than that of the physical servers in the Tier 2 performance server group, and the resource utilization rate of the physical servers in the Tier 2 performance server group is greater than that of the physical servers in the Tier 3 performance server group.

[0087] Step 304: For each server group in the multiple server groups, identify the physical server with a resource utilization rate greater than the preset resource utilization rate as the source server.

[0088] Step 305: Before the end of the current time period, migrate the target virtual machine running on the source server to the target server.

[0089] In one possible implementation, all physical servers can be divided into high-level physical servers (i.e., level 1 performance server group), medium-level physical servers (i.e., level 2 performance server group), and low-level physical servers (i.e., level 3 performance server group) according to pre-defined partitioning rules.

[0090] Specifically, the division rules can be set based on actual experience, such as dividing based on the CPU utilization (CPUS) value, as shown in Table 5.

[0091] Table 5

[0092] Alternatively, the partitioning rules can also be based on the memory usage rate (NCS) value, as shown in Table 6.

[0093] Table 6

[0094] Furthermore, preset resource utilization rates can be set for each of the Tier 1, Tier 2, and Tier 3 performance server groups. For example, the preset resource utilization rates for the Tier 1 performance server group include: first preset CPU utilization rate T1 and first preset memory utilization rate U1; the preset resource utilization rates for the Tier 2 performance server group include: second preset CPU utilization rate T2 and second preset memory utilization rate U2; and the preset resource utilization rates for the Tier 3 performance server group include: third preset CPU utilization rate T3 and third preset memory utilization rate U3.

[0095] Therefore, this application divides multiple physical servers into multiple server groups based on the resource utilization rate of each physical server. This allows for the accurate identification of the physical server with a resource utilization rate exceeding a preset rate as the source server within each server group. This ensures accurate identification of the physical server requiring virtual machine migration, avoiding incorrect virtual machine migrations that could lead to physical server malfunctions.

[0096] Figure 4 This is the fourth flowchart of the virtual machine migration method provided by the present invention, as shown below. Figure 4 As shown, the above "Step 305: Before the end of the current time period, migrate the target virtual machine running on the source server to the target server" specifically includes the following: Step 401: For the source server in the first-level performance server group, identify the virtual machine with the lowest running parameters among the virtual machines running on the source server as the target virtual machine.

[0097] Step 402: Traverse multiple physical servers in descending order of resource utilization, and determine the first physical server that meets the target condition as the target server.

[0098] The target conditions include the sum of the physical server's resource utilization rate and the target virtual machine's operating parameters being less than or equal to the preset resource utilization rate.

[0099] Step 403: Migrate the target virtual machine running on the source server to the target server.

[0100] In one possible implementation, for the physical servers included in the Tier 1 performance server group, it can be determined whether there are any physical servers with a CPU utilization (CPUS) greater than a set upper limit (i.e., a first preset CPU utilization (T1)). If so, all virtual machines running on that physical server are evaluated based on their average CPU utilization over the predicted time period. The values ​​are sorted in descending order, and the migration path of the last virtual machine in the sort (i.e., the target virtual machine) is determined.

[0101] Specifically, to determine the migration path of the target virtual machine, it is possible to traverse all physical servers (i.e., multiple physical servers) included in the first-level performance server group, the second-level performance server group, and the third-level performance server group. This process continues until a physical server is identified in which, after the virtual machine migration, the CPU utilization (CPUS) is less than or equal to the upper limit (first preset CPU utilization (T1)) and the memory utilization (NCS) is less than or equal to the upper limit (first preset memory utilization (U1)). At this point, the traversal stops, and the identified physical server is designated as the target server.

[0102] In one possible implementation, the process of determining the source server (target virtual machine) and the target server can be repeated until there are no physical servers with CPU utilization (CPUS) greater than the set upper limit, at which point the process ends.

[0103] In one possible implementation, for the physical servers included in the Tier 1 performance server group, it can be further determined whether there are any physical servers with a memory utilization rate (NCS) greater than a set upper limit (first preset memory utilization rate U1). If so, all virtual machines running on that physical server are evaluated based on their average memory utilization rate over the predicted time period. The values ​​are sorted in descending order, and the migration path of the last virtual machine in the sort (i.e., the target virtual machine) is determined.

[0104] Specifically, to determine the migration path of the target virtual machine, it is possible to traverse all physical servers (i.e., multiple physical servers) included in the first-level performance server group, the second-level performance server group, and the third-level performance server group. This process continues until a physical server is identified whose CPU utilization (CPUS) is less than or equal to the upper limit (first preset CPU utilization T1) and whose memory utilization (NCS) is less than or equal to the upper limit (first preset memory utilization U1) after the virtual machine migration. At this point, the traversal stops, and the identified physical server is designated as the target server.

[0105] In one possible implementation, the values ​​of the preset CPU utilization T and preset memory utilization U can be determined according to the actual situation. Specifically, the upper limit of CPU utilization CPUS that ensures the smooth and stable operation of the physical server is taken as the value of T, and the upper limit of memory utilization NCS that ensures the smooth and stable operation of the physical server is taken as the value of U.

[0106] Therefore, this application targets the source server in a Tier 1 performance server group by identifying the virtual machine with the lowest running parameters among the virtual machines running on the source server as the target virtual machine. Multiple physical servers are then traversed sequentially according to resource utilization from high to low to determine the physical server that meets the target criteria as the target server. The target virtual machine running on the source server is then migrated to the target server. This ensures that after the target virtual machine is migrated, the resource utilization of both the source and target servers will not exceed a preset resource utilization rate. This allows for accurate virtual machine migration, improving the efficiency and success rate of the migration process.

[0107] Figure 5 This is the fifth flowchart illustrating the virtual machine migration method provided by this invention, as shown below. Figure 5 As shown, the above "Step 305, before the end of the current time period, migrate the target virtual machine running on the source server to the target server" may further include the following: Step 501: For the source server in the secondary performance server group, identify all virtual machines running on the source server as target virtual machines.

[0108] Step 502: In order of resource utilization from high to low, traverse the physical servers included in the secondary performance server group and the tertiary performance server group, and determine the first physical server that meets the target condition as the target server.

[0109] The target conditions include the sum of the physical server's resource utilization rate and the target virtual machine's operating parameters being less than or equal to the preset resource utilization rate.

[0110] Step 503: Migrate the target virtual machine running on the source server to the target server.

[0111] In one possible implementation, for the physical servers included in the secondary performance server group, all virtual machines running on physical servers with a CPU utilization rate (CPUS) greater than a set upper limit (i.e., the second preset CPU utilization rate (T2)) can be identified as target virtual machines.

[0112] Alternatively, all virtual machines running on physical servers with a memory utilization rate (NCS) greater than the set upper limit (i.e., the second preset memory utilization rate (U2)) can be identified as target virtual machines.

[0113] Furthermore, when determining the migration path of the target virtual machine, all physical servers included in the secondary performance server group and the tertiary performance server group can be traversed. The traversal continues until a physical server is determined whose CPU utilization (CPUS) is less than or equal to the upper limit (second preset CPU utilization T2) and memory utilization (NCS) is less than or equal to the upper limit (second preset memory utilization U2) after the virtual machine migration. The traversal then stops, and the determined physical server is identified as the target server.

[0114] In this way, by migrating all virtual machines running on the source server in the secondary performance server group to the target server, the source server can be left idle, and the idle source server can be powered off and shut down, thereby reducing energy consumption.

[0115] In one possible implementation, for the physical servers included in the secondary performance server group, the source server, target virtual machine, and target server can be determined in the same way as the primary performance server group, thereby migrating the target virtual machine.

[0116] Therefore, this application targets the source server in a level-two performance server group by identifying all virtual machines running on the source server as target virtual machines. It then sequentially traverses the physical servers within the level-two and level-three performance server groups according to their resource utilization from highest to lowest, identifying the physical servers that meet the target criteria as target servers. This allows the target virtual machines running on the source server to be migrated to the target server. This ensures that after the target virtual machines are migrated, the resource utilization of both the source and target servers will not exceed a preset resource utilization rate. This results in accurate virtual machine migration, improving the efficiency and success rate of the migration process.

[0117] Figure 6 This is the sixth flowchart of the virtual machine migration method provided by the present invention, as shown below. Figure 6 As shown, the above "Step 305, before the end of the current time period, migrate the target virtual machine running on the source server to the target server" may further include the following: Step 601: For the source server in the Tier 3 performance server group, identify all virtual machines running on the source server as target virtual machines.

[0118] Step 602: Traverse the physical servers included in the Level 3 performance server group in descending order of resource utilization, and determine the first physical server that meets the target condition as the target server.

[0119] The target conditions include the sum of the physical server's resource utilization rate and the target virtual machine's operating parameters being less than or equal to the preset resource utilization rate.

[0120] Step 603: Migrate the target virtual machine running on the source server to the target server.

[0121] In one possible implementation, for the physical servers included in the Tier 3 performance server group, all virtual machines running on physical servers with a CPU utilization rate (CPUS) greater than a set upper limit (i.e., the third preset CPU utilization rate (T3)) can be identified as target virtual machines.

[0122] Alternatively, all virtual machines running on physical servers with a memory utilization rate (NCS) greater than the set upper limit (i.e., the third preset memory utilization rate (U3)) can be identified as target virtual machines.

[0123] Furthermore, when determining the migration path of the target virtual machine, all physical servers included in the three-level performance server group can be traversed until a physical server is identified in which the CPU utilization CPUS is less than or equal to the upper limit (third preset CPU utilization T3) and the memory utilization NCS is less than or equal to the upper limit (third preset memory utilization U3) after the virtual machine migration. Then, the traversal stops and the identified physical server is determined as the target server.

[0124] In this way, by migrating all virtual machines running on the source server in the Tier 3 performance server group to the target server, the source server can be left idle, and the idle source server can be powered off and shut down, thereby reducing energy consumption.

[0125] In one possible implementation, for the physical servers included in a Tier 3 performance server group, the source server, target virtual machine, and target server can be determined in the same way as a Tier 1 performance server group, thereby migrating the target virtual machine.

[0126] Therefore, this application targets the source server in a Tier 3 performance server group by identifying all virtual machines running on the source server as target virtual machines. It then sequentially traverses the physical servers within the Tier 3 performance server group according to their resource utilization from highest to lowest, identifying the physical servers that meet the target criteria as target servers. The target virtual machines running on the source server are then migrated to the target server. This ensures that after the target virtual machine migration, the resource utilization of both the source and target servers will not exceed a preset resource utilization rate. This allows for accurate virtual machine migration, improving the efficiency and success rate of the migration process.

[0127] In this embodiment, an average operating parameter model is established by collecting virtual machine operating parameters. Using this model, the migration of virtual machines within a predicted time period is predicted, a dynamic migration strategy is developed, and migration is performed at the end of the current time period. This effectively avoids the problems of real-time migration in traditional technologies, where the CPU and memory utilization rates of virtual machines vary at different times. Migration can occur when the CPU utilization (CPUS) and memory utilization (NCS) of the physical server suddenly increase, leading to physical server overload, decreased migration speed, and migration failure. This effectively improves the user experience.

[0128] In terms of technological innovation, advanced sensors can be integrated for precise monitoring and control, new heat dissipation technologies can be explored and combined with intelligent temperature control, waste heat recovery and utilization can be studied, and artificial intelligence and machine learning can be used to predict and optimize energy consumption. In terms of management strategies, dynamic resource allocation can be implemented to adjust according to business needs, load balancing can be optimized to ensure balanced load distribution, energy-saving factors can be considered in green data center design, and personalized energy-saving suggestions can be provided by analyzing user behavior. From the perspective of cooperation and sharing, collaboration with industry partners can promote energy-saving management, explore resource-sharing models to improve utilization, and strengthen industry-academia-research cooperation to innovate energy-saving technologies.

[0129] Regarding technical risks, thorough reliability testing was conducted before deployment, a comprehensive backup and recovery mechanism was established, and compatibility testing was performed. Regarding risk management, compliance with policies and regulations was maintained, security management was strengthened, and training and education were provided.

[0130] Based on this, this application can significantly reduce operating costs. On the one hand, it saves energy costs, significantly reducing electricity expenses in data center operations; on the other hand, it reduces equipment maintenance costs, decreasing equipment wear, failures, and the frequency of replacement. Secondly, it can enhance competitiveness, establish an environmentally friendly image, attract customers and partners; ensure service quality, improve customer satisfaction, and strengthen market competitiveness. Furthermore, it can adapt to industry development trends, meet sustainable development requirements, and easily obtain policy support; it can also address the challenges of data center scale growth by providing sustainable development solutions.

[0131] In one possible implementation, Figure 7 This is a schematic diagram of the virtual machine migration system provided by the present invention, as shown below. Figure 7 As shown, the virtual machine migration system includes: an energy-saving management module, a virtual machine scheduling module, and multiple physical servers. Each of the multiple physical servers runs multiple virtual machines. The energy-saving management module, the virtual machine scheduling module, and the multiple physical servers are connected via a network.

[0132] The virtual machine migration apparatus provided by the present invention is described below. The virtual machine migration apparatus described below and the virtual machine migration method described above can be referred to in correspondence.

[0133] Figure 8 This is a schematic diagram of the virtual machine migration device provided by the present invention, as shown below. Figure 8 As shown, the virtual machine migration device includes the following modules: an acquisition module 801, a processing module 802, and a migration module 803; The acquisition module 801 is used to acquire the operating parameters of each virtual machine running on each of multiple physical servers in the predicted time period based on a pre-built average operating parameter model in the current time period. At least one virtual machine is running on each physical server, and the predicted time period is the next time period after the current time period. Processing module 802 is used to determine the resource utilization of each physical server based on the operating parameters of at least one virtual machine running on each physical server. The processing module 802 is also used to determine the source server from multiple physical servers based on the resource utilization rate of each physical server, wherein the resource utilization rate of the source server is greater than the preset resource utilization rate. Migration module 803 is used to migrate the target virtual machine running on the source server to the target server before the end of the current time period. After the target virtual machine is migrated, the resource utilization rate of both the source server and the target server is less than or equal to the preset resource utilization rate.

[0134] According to a virtual machine migration device provided by the present invention, the processing module 802 is further configured to divide the total duration of each day into multiple equal time periods. The acquisition module 801 is also used to acquire the historical running parameters of each virtual machine running on each physical server in each time period over several days prior to the current time period; The processing module 802 is also used to determine the average operating parameters of each virtual machine over multiple days for each time period of the day; The processing module 802 is also used to build an average operating parameter model based on the average operating parameters of each virtual machine in each time period.

[0135] According to the present invention, a virtual machine migration device has operating parameters including at least one of the following: CPU utilization rate and memory utilization rate.

[0136] According to the present invention, a virtual machine migration apparatus, wherein the processing module 802 is specifically used for: Based on the resource utilization of each physical server, multiple physical servers are divided into multiple server groups. The multiple server groups include: Level 1 performance server group, Level 2 performance server group and Level 3 performance server group. The resource utilization of physical servers in the Level 1 performance server group is greater than that of physical servers in the Level 2 performance server group, and the resource utilization of physical servers in the Level 2 performance server group is greater than that of physical servers in the Level 3 performance server group. For each of the multiple server groups, the physical server with a resource utilization rate greater than the preset resource utilization rate is identified as the source server.

[0137] According to the present invention, a virtual machine migration apparatus, a migration module 803, is specifically used for: For the source server in the first-level performance server group, the virtual machine with the lowest running parameters among the virtual machines running on the source server is identified as the target virtual machine; The system iterates through multiple physical servers in descending order of resource utilization, and identifies the first physical server that meets the target conditions as the target server. The target conditions include that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0138] According to the present invention, a virtual machine migration apparatus, a migration module 803, is specifically used for: For the source server in the secondary performance server group, all virtual machines running on the source server are identified as target virtual machines; The physical servers in the Level 2 and Level 3 performance server groups are traversed sequentially from highest to lowest resource utilization. The first physical server that meets the target condition is identified as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0139] According to the present invention, a virtual machine migration apparatus, a migration module 803, is specifically used for: For the source server in the Tier 3 performance server group, all virtual machines running on the source server are identified as target virtual machines; The physical servers in the three-level performance server group are traversed in descending order of resource utilization. The first physical server that meets the target condition is identified as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

[0140] Figure 9 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 9 As shown, the electronic device may include: a processor 910, a communications interface 920, a memory 930, and a communication bus 940, wherein the processor 910, the communications interface 920, and the memory 930 communicate with each other through the communication bus 940. The processor 910 can call logical instructions in the memory 930 to execute a virtual machine migration method, which includes: obtaining the operating parameters of each virtual machine running on each of a plurality of physical servers in a predicted time period based on a pre-built average operating parameter model in the current time period, wherein each physical server runs at least one virtual machine, and the predicted time period is the next time period after the current time period; determining the resource utilization rate of each physical server based on the operating parameters of the at least one virtual machine running on each physical server; determining a source server from the plurality of physical servers based on the resource utilization rate of each physical server, wherein the resource utilization rate of the source server is greater than a preset resource utilization rate; and migrating the target virtual machine running on the source server to the target server before the end of the current time period, wherein after the target virtual machine is migrated, the resource utilization rate of both the source server and the target server is less than or equal to the preset resource utilization rate.

[0141] Furthermore, the logical instructions in the aforementioned memory 930 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0142] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the virtual machine migration method provided by the above methods. The method includes: obtaining the operating parameters of each virtual machine running on each of a plurality of physical servers in a predicted time period based on a pre-built average operating parameter model in the current time period, wherein at least one virtual machine runs on each physical server, and the predicted time period is the next time period after the current time period; determining the resource utilization rate of each physical server based on the operating parameters of at least one virtual machine running on each physical server; determining a source server from the plurality of physical servers based on the resource utilization rate of each physical server, wherein the resource utilization rate of the source server is greater than a preset resource utilization rate; and migrating the target virtual machine running on the source server to the target server before the end of the current time period, wherein after the target virtual machine is migrated, the resource utilization rate of both the source server and the target server is less than or equal to the preset resource utilization rate.

[0143] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is implemented to perform the virtual machine migration method provided by the above methods. The method includes: obtaining, in the current time period, the operating parameters of each virtual machine running on each of a plurality of physical servers in a predicted time period based on a pre-built average operating parameter model, wherein at least one virtual machine runs on each physical server, and the predicted time period is the next time period after the current time period; determining the resource utilization rate of each physical server based on the operating parameters of the at least one virtual machine running on each physical server; determining a source server from the plurality of physical servers based on the resource utilization rate of each physical server, wherein the resource utilization rate of the source server is greater than a preset resource utilization rate; and migrating the target virtual machine running on the source server to the target server before the end of the current time period, wherein after the target virtual machine is migrated, the resource utilization rate of both the source server and the target server is less than or equal to the preset resource utilization rate.

[0144] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0145] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A virtual machine migration method, characterized in that, include: Based on a pre-built average operating parameter model, the operating parameters of each virtual machine running on each of multiple physical servers in the current time period are obtained in the predicted time period. At least one virtual machine runs on a physical server, and the predicted time period is the next time period after the current time period. Based on the operating parameters of at least one virtual machine running on each physical server, determine the resource utilization of each physical server; Based on the resource utilization rate of each physical server, a source server is determined from the plurality of physical servers, wherein the resource utilization rate of the source server is greater than a preset resource utilization rate; Before the end of the current time period, the target virtual machine running on the source server is migrated to the target server. After the target virtual machine is migrated, the resource utilization rate of both the source server and the target server is less than or equal to the preset resource utilization rate.

2. The virtual machine migration method according to claim 1, characterized in that, The pre-built average operating parameter model is constructed in the following way: Divide the total daily duration into multiple equal-length time periods; Retrieve the historical running parameters of each virtual machine running on each physical server for each time period, spanning multiple days prior to the current time period; For each time period of the day, determine the average operating parameters of each virtual machine over multiple days; The average operating parameter model is constructed based on the average operating parameters of each virtual machine in each time period.

3. The virtual machine migration method according to claim 2, characterized in that, The operating parameters include at least one of the following: CPU utilization rate and memory utilization rate.

4. The virtual machine migration method according to any one of claims 1-3, characterized in that, The process of determining the source server from the plurality of physical servers based on the resource utilization rate of each physical server includes: Based on the resource utilization rate of each physical server, the multiple physical servers are divided into multiple server groups, including: a first-level performance server group, a second-level performance server group, and a third-level performance server group. The resource utilization rate of the physical servers in the first-level performance server group is greater than that of the physical servers in the second-level performance server group, and the resource utilization rate of the physical servers in the second-level performance server group is greater than that of the physical servers in the third-level performance server group. For each of the multiple server groups, the physical server with a resource utilization rate greater than the preset resource utilization rate is identified as the source server.

5. The virtual machine migration method according to claim 4, characterized in that, The step of migrating the target virtual machine running on the source server to the target server before the end of the current time period includes: For the source server in the first-level performance server group, the virtual machine with the lowest running parameters among the virtual machines running on the source server is determined as the target virtual machine; The multiple physical servers are traversed sequentially in descending order of resource utilization. The first physical server that meets the target condition is determined as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

6. The virtual machine migration method according to claim 4, characterized in that, The step of migrating the target virtual machine running on the source server to the target server before the end of the current time period includes: For the source server in the secondary performance server group, all virtual machines running on the source server are identified as the target virtual machines; The physical servers included in the secondary performance server group and the tertiary performance server group are traversed sequentially in descending order of resource utilization. The first physical server that meets the target condition is determined as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

7. The virtual machine migration method according to claim 4, characterized in that, The step of migrating the target virtual machine running on the source server to the target server before the end of the current time period includes: For the source server in the three-level performance server group, all virtual machines running on the source server are identified as the target virtual machines; The physical servers in the three-level performance server group are traversed sequentially in descending order of resource utilization. The first physical server that meets the target condition is determined as the target server. The target condition includes that the sum of the physical server's resource utilization and the target virtual machine's running parameters is less than or equal to the preset resource utilization. Migrate the target virtual machine running on the source server to the target server.

8. A virtual machine migration device, characterized in that, include: Acquisition module, processing module, and migration module; The acquisition module is used to acquire the operating parameters of each virtual machine running on each of multiple physical servers in the predicted time period based on a pre-built average operating parameter model in the current time period. At least one virtual machine is running on a physical server, and the predicted time period is the next time period after the current time period. The processing module is used to determine the resource utilization rate of each physical server based on the operating parameters of at least one virtual machine running on each physical server. The processing module is further configured to determine a source server from the plurality of physical servers based on the resource utilization rate of each physical server, wherein the resource utilization rate of the source server is greater than a preset resource utilization rate. The migration module is used to migrate the target virtual machine running on the source server to the target server before the end of the current time period. After the target virtual machine is migrated, the resource utilization rate of both the source server and the target server is less than or equal to the preset resource utilization rate.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the virtual machine migration method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the virtual machine migration method as described in any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the virtual machine migration method as described in any one of claims 1 to 7.

Images