Cloud host migration method, device, electronic device and readable storage medium

Abstract

The embodiment of the present invention provides a migration method, device, electronic device and readable storage medium for a cloud host. The method includes: obtaining a host set, adding a host that meets a first condition in the host set to a candidate host set, obtaining the air outlet temperature of a host that does not meet the first condition in the host set, determining a host whose air outlet temperature meets a second condition as a host to be migrated, and determining a cloud host to be migrated from cloud hosts running on the host to be migrated, adding a host whose air outlet temperature does not meet the second condition to a candidate host set, obtaining the resource load value of the candidate host in the candidate host set, calculating the score of the candidate host according to the resource load value of the candidate host, taking the candidate host with the lowest score as the target host, migrating the cloud host to be migrated in the host to be migrated to the target host, and reducing the air outlet temperature of the host to be migrated.

Description

Technical Field

[0001] The present invention belongs to the technical field of cloud computing, and in particular relates to a cloud host migration method, device, electronic device and readable storage medium. Background Art

[0002] In the current cloud platform construction and application, especially the public cloud platform construction, a resource pool will deploy tens of thousands of hosts. If a host runs high-energy-consuming tasks for a long time, or the host is overloaded, the temperature of the host will be too high, thus creating risks.

[0003] The host machine is overloaded for a long time and the air outlet temperature is too high, which will affect the performance of the host machine components. Summary of the invention

[0004] The present invention provides a cloud host migration method, device, electronic device and readable storage medium to solve technical problems in related technologies.

[0005] In a first aspect, the present invention provides a cloud host migration method, the method comprising:

[0006] Obtain a host set, and add a host in the host set that meets a first condition to a candidate host set; the first condition includes: the host is in an enabled state and no cloud host is running;

[0007] Obtaining the outlet temperature of the host machine in the host machine set that does not meet the first condition, determining the host machine whose outlet temperature meets the second condition as the host machine to be migrated, and determining the cloud host to be migrated from the cloud hosts running on the host machine to be migrated;

[0008] Adding the host machine whose air outlet temperature does not meet the second condition to the candidate host machine set;

[0009] Obtain resource load values ​​of candidate host machines in the candidate host machine set, calculate scores of the candidate host machines according to the resource load values ​​of the candidate host machines, and use the candidate host machine with the lowest score as the target host machine; the scores of the candidate host machines are positively correlated with the resource load values ​​of the candidate host machines;

[0010] Migrate the cloud host to be migrated in the host to be migrated to the target host.

[0011] In a second aspect, the present invention provides a cloud host migration device, the device comprising:

[0012] A first adding module is used to obtain a host set, and add a host in the host set that meets a first condition to a candidate host set; the first condition includes: the host is in an enabled state and no cloud host is running;

[0013] A first determination module is used to obtain the air outlet temperature of the host machine in the host machine set that does not meet the first condition, determine the host machine whose air outlet temperature meets the second condition as the host machine to be migrated, and determine the cloud host to be migrated from the cloud hosts running on the host machine to be migrated;

[0014] A second adding module, configured to add the host machine whose air outlet temperature does not satisfy the second condition into the candidate host machine set;

[0015] A calculation module, used to obtain the resource load value of the candidate host in the candidate host set, calculate the score of the candidate host according to the resource load value of the candidate host, and use the candidate host with the lowest score as the target host; the score of the candidate host is positively correlated with the resource load value of the candidate host;

[0016] The migration module is used to migrate the cloud host to be migrated in the host to be migrated to the target host.

[0017] In a third aspect, the present invention provides an electronic device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned cloud host migration method when executing the program.

[0018] In a fourth aspect, the present invention provides a readable storage medium, which, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute the above-mentioned cloud host migration method.

[0019] In an embodiment of the present invention, a host machine set is obtained, and a host machine in the host machine set that meets a first condition is added to a candidate host machine set, the first condition comprising: the host machine is in an enabled state and a cloud host is not running, the outlet temperature of the host machine in the host machine set that does not meet the first condition is obtained, a host machine whose outlet temperature meets a second condition is determined as a host machine to be migrated, and a cloud host to be migrated is determined from the cloud hosts running on the host machine to be migrated, and a host machine whose outlet temperature does not meet the second condition is added to the candidate host machine set, and the resource load values ​​of the candidate host machines in the candidate host machine set are obtained. According to the resource load values ​​of the candidate host machines, the scores of the candidate host machines are calculated, and the candidate host machine with the lowest score is selected as the target host machine, and the cloud hosts to be migrated in the host machines to be migrated are migrated to the target host machine. The embodiment of the present invention divides the host machine to be migrated and the candidate host machine according to the outlet temperature of the host machine. When the temperature of the host machine to be migrated is too high, a suitable target host machine is selected for the cloud host to be migrated running on the host machine to be migrated, so as to migrate the cloud host to be migrated to the target host machine. The outlet temperature of the host machine is used as a key factor to adjust the cloud host resources to reduce the overload of the host machine to be migrated, thereby reducing the outlet temperature of the host machine to be migrated. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 It is a flowchart of the steps of a cloud host migration method provided by an embodiment of the present invention;

[0022] Figure 2 It is a flowchart of the steps of another cloud host migration method provided by an embodiment of the present invention;

[0023] Figure 3 is a flowchart of steps for screening a host machine to be migrated and a target host machine provided by an embodiment of the present invention;

[0024] Figure 4 This is a flowchart of steps for selecting a cloud host to be migrated provided by an embodiment of the present invention;

[0025] Figure 5 A flowchart of the steps of selecting a target host for a cloud host to be migrated provided by an embodiment of the present invention;

[0026] Figure 6 A cloud host migration device provided by an embodiment of the present invention;

[0027] Figure 7 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] In the current cloud platform construction and application, especially the public cloud platform construction, a resource pool will deploy tens of thousands of physical hosts, but a large number of hosts need to be maintained or removed from the shelves before their host life cycle. The main reason is that the host is not used under friendly protection. For example, if a host runs high-energy-consuming tasks for a long time or the host is overloaded, the host temperature will be too high. If the host is at too high a temperature for a long time, the following potential risks will arise: Performance degradation: Overheating will reduce the operating efficiency of the host and may cause performance degradation. High host machine temperature can also affect the performance of host machine components, such as the central processing unit (CPU), memory, and storage devices, resulting in slower response time and reduced throughput; System instability and failure: Overheating can make the host machine unstable and cause unexpected system crashes or shutdowns, which can directly lead to service interruptions, data loss, and possible damage to hardware components; Hardware damage: Prolonged exposure to high temperatures can cause permanent damage to the host machine's hardware components. Heat can degrade sensitive electronic circuits, shorten the service life of components, and increase the likelihood of hardware failure; Fire hazards: In extreme cases, overheating combined with other factors such as electrical failures or inadequate cooling systems may create fire hazards in the data center or host machine room, posing a significant threat to the safety of equipment and nearby personnel.

[0030] The long-term overload of the host machine causes the temperature of the host machine outlet to be too high, which will increase the probability of damage to the surrounding components, thereby affecting the performance of the host machine. In an embodiment of the present invention, a host machine set is obtained, and the host machines in the host machine set that meet the first condition are added to the candidate host machine set. The first condition includes: the host machine is in an enabled state and the cloud host is not running. The outlet temperature of the host machine in the host machine set that does not meet the first condition is obtained, and the host machine whose outlet temperature meets the second condition is determined as the host machine to be migrated, and the cloud host to be migrated is determined from the cloud hosts running on the host machine to be migrated, and the host machine whose outlet temperature does not meet the second condition is added to the candidate host machine set, and the resource load value of the candidate host machine in the candidate host machine set is obtained. According to the resource load value of the candidate host machine, the score of the candidate host machine is calculated, and the candidate host machine with the lowest score is used as the target host machine, and the cloud host to be migrated in the host machine to be migrated is migrated to the target host machine. The embodiment of the present invention divides the host machine to be migrated and the candidate host machine according to the outlet temperature of the host machine. When the temperature of the host machine to be migrated is too high, a suitable target host machine is selected for the cloud host to be migrated running on the host machine to be migrated, so as to migrate the cloud host to be migrated to the target host machine. The outlet temperature of the host machine is used as a key factor to adjust the cloud host resources to reduce the overload of the host machine to be migrated, thereby reducing the outlet temperature of the host machine to be migrated, reducing the probability of damage to components around the host machine, and improving the performance of the host machine.

[0031] Figure 1 is a flow chart of the steps of a cloud host migration method provided by an embodiment of the present invention, such as Figure 1 As shown, the method may include:

[0032] Step 101: Obtain a host set, and add a host in the host set that meets a first condition to a candidate host set; the first condition includes: the host is in an enabled state and a cloud host is not running.

[0033] For example, a host includes information of multiple hosts. A cloud host can only exist in one host at the same time, and a host can include multiple cloud hosts. A host can be a physical device, for example, a server. A host can also be called a computing node.

[0034] For example, the first condition may be that the host is in an enabled state and the cloud host is not running, the host's running information is obtained from the host set, the host in the host set that is in an enabled state and does not run the cloud host is determined as a candidate host, and is added to the candidate host set.

[0035] Step 102: Obtain the outlet temperature of the host machine in the host machine set that does not meet the first condition, determine the host machine whose outlet temperature meets the second condition as the host machine to be migrated, and determine the cloud host to be migrated from the cloud hosts running on the host machine to be migrated.

[0036] For example, if the host machine is in an enabled state and is running a cloud host, the historical monitoring data of the host machine, namely, the air outlet temperature, can be obtained from the cloud monitoring platform.

[0037] For example, the second condition may be that the air outlet temperature of the host machine is greater than or equal to a first preset value. The average air outlet temperature of the host machine is calculated based on the air outlet temperature of the host machine, and when the average air outlet temperature of the host machine is greater than or equal to the first preset value, the host machine is determined as the host machine to be migrated.

[0038] Step 103: Add the host machines whose air outlet temperatures do not meet the second condition into the candidate host machine set.

[0039] For example, when the average value of the outlet temperature of the host machine is less than a first preset value, the host machine is determined as a candidate host machine and added to the candidate host machine set.

[0040] Step 104, obtaining the resource load values ​​of the candidate host machines in the candidate host machine set, calculating the scores of the candidate host machines according to the resource load values ​​of the candidate host machines, and taking the candidate host machine with the lowest score as the target host machine; the scores of the candidate host machines are positively correlated with the resource load values ​​of the candidate host machines.

[0041] For example, the resource load value of the candidate host machine may be an average value of the outlet temperature of the candidate host machine, may be a memory utilization rate, or may be a CPU utilization rate.

[0042] For example, a low resource load value of the candidate host indicates that the number of cloud hosts running on the candidate host is small, or the cloud host runs few tasks. In this case, the migrated cloud host can be accepted.

[0043] For example, the higher the resource load value of the candidate host machine, the higher the score of the candidate host machine, and the lower the resource load value of the candidate host machine, the lower the score of the candidate host machine.

[0044] Step 105: Migrate the cloud host to be migrated in the host to be migrated to the target host.

[0045] For example, in order to prevent the cloud host from performing secondary or chained migration operations, before the cloud host is migrated, the migration task list is deduplicated or chained migration task elimination operations are performed to check whether there are multiple migration operations on the same cloud host. If there are multiple migrations, the target host of the cloud host migration is checked, and the final migration path is sorted out and the migration dependency is eliminated. If the host has both cloud hosts to be migrated out and cloud hosts to be migrated in, the priority of the cloud host to be migrated out will be increased, and the cloud host will be migrated to the specified target host.

[0046] In an embodiment of the present invention, a host machine set is obtained, and a host machine in the host machine set that meets a first condition is added to a candidate host machine set, the first condition comprising: the host machine is in an enabled state and a cloud host is not running, the outlet temperature of the host machine in the host machine set that does not meet the first condition is obtained, a host machine whose outlet temperature meets a second condition is determined as a host machine to be migrated, and a cloud host to be migrated is determined from the cloud hosts running on the host machine to be migrated, and a host machine whose outlet temperature does not meet the second condition is added to the candidate host machine set, and the resource load values ​​of the candidate host machines in the candidate host machine set are obtained. According to the resource load values ​​of the candidate host machines, the scores of the candidate host machines are calculated, and the candidate host machine with the lowest score is selected as the target host machine, and the cloud hosts to be migrated in the host machines to be migrated are migrated to the target host machine. The embodiment of the present invention divides the host machine to be migrated and the candidate host machine according to the outlet temperature of the host machine. When the temperature of the host machine to be migrated is too high, a suitable target host machine is selected for the cloud host to be migrated running on the host machine to be migrated, so as to migrate the cloud host to be migrated to the target host machine. The outlet temperature of the host machine is used as a key factor to adjust the cloud host resources to reduce the overload of the host machine to be migrated, thereby reducing the outlet temperature of the host machine to be migrated.

[0047] Figure 2 is a flowchart of another cloud host migration method provided by an embodiment of the present invention. Figure 2 As shown, the method may include:

[0048] Step 201: Obtain a host set, and add a host in the host set that meets a first condition to a candidate host set; the first condition includes: the host is in an enabled state and a cloud host is not running.

[0049] This step may specifically refer to the above step 101, which will not be described in detail here.

[0050] Step 202: Obtain the outlet temperature of the host machine in the host machine set that does not meet the first condition, and calculate the average value of the outlet temperature of the host machine within a preset time range.

[0051] For example, if the host machine is enabled and the cloud host is running, the host machine outlet temperature is obtained and the average outlet temperature is calculated.

[0052] For example, the calculation formula is as follows:

[0053]

[0054] Among them, H outlet_temp Indicates the outlet temperature of a host machine, PM outlet_temp Indicates the average temperature of the air outlet of a host machine in n time periods, where n is a positive integer.

[0055] Step 203: If the average value is greater than or equal to a first preset value, the host machine is determined as the host machine to be migrated, and a cloud host to be migrated is determined from the cloud hosts running on the host machine to be migrated.

[0056] For example, the first preset value may be W threshold , which indicates the limit threshold set for a centralized storage resource pool, and the value range is {0,100}.

[0057] For example, if the host's PM outlet_temp Greater than or equal to W threshold , it means that the outlet temperature of the host machine is too high and migration is required, and the host machine is determined as the host machine to be migrated.

[0058] Optionally, step 203 may specifically include:

[0059] Sub-step 2031, obtaining the L3 cache utilization of the cloud host running on the host to be migrated;

[0060] Sub-step 2032: determining a cloud host to be migrated in the host to be migrated according to the L3 cache utilization of the cloud host running on the host to be migrated.

[0061] For example, for sub-steps 2031-2032, the application of the third-level cache can further reduce memory latency and improve the performance of the processor when calculating large amounts of data. The third-level (L3, Level 3) cache utilization refers to the proportion of the L3 cache used during the operation of the CPU. The L3 cache is the third level cache in the CPU cache, which usually has a large capacity and is used to store data with low access frequency but still needs to be frequently accessed. The utilization of the L3 cache reflects the extent to which the CPU uses the L3 cache, which directly affects the overall performance and efficiency of the CPU.

[0062] For example, after obtaining the L3 cache utilization of the cloud host running on the host to be migrated, the cloud host to be migrated in the host to be migrated is determined according to the L3 cache utilization of the cloud host running on the host to be migrated. Since the L3 cache utilization reflects the CPU's demand for cache under high load conditions, when the CPU's demand for cache increases, the L3 cache utilization will increase. Therefore, according to the L3 cache utilization, the accuracy of determining the cloud host to be migrated in the host to be migrated can be improved.

[0063] Optionally, the cloud host to be migrated includes: a high-priority cloud host and a noisy cloud host, the high-priority cloud host is a cloud host whose level 3 cache utilization decrease value exceeds a second preset value; the noisy cloud host is a cloud host whose level 3 cache utilization increase value exceeds the second preset value; sub-step 2032 may specifically include:

[0064] Sub-step 20321, obtaining the L3 cache utilization rate of the cloud host in the current cycle and the L3 cache utilization rate of the previous cycle;

[0065] Sub-step 20322: Calculate the difference between the L3 cache utilization of the cloud host in the current cycle and the L3 cache utilization of the previous cycle;

[0066] Sub-step 20323: if the difference is less than the product of the second preset value and the L3 cache utilization rate of the previous cycle, then determining that the cloud host is a noisy cloud host;

[0067] Sub-step 20324: If the difference is less than the inverse value of the product of the second preset value and the third-level cache utilization of the previous cycle, determine that the cloud host is a high-priority cloud host.

[0068] For sub-steps 20321-20324, high-priority cloud hosts and noisy cloud hosts are cyclically screened out. For hosts with high-priority cloud hosts and noisy cloud hosts, they are marked as hosts that need to release noise, that is, hosts to be migrated. For candidate hosts that do not have high-priority and noisy cloud hosts, they are marked as candidate hosts that can accommodate noisy cloud hosts. Based on the L3 cache utilization of the cloud hosts sampled at multiple sampling time points, the L3 cache utilization of the cloud host in the current cycle is calculated, and the calculation formula is as follows:

[0069]

[0070] Among them, VM l3_cache(i) is the L3 cache utilization of the cloud host sampled at the i-th sampling time point, i = 1, 2, ..., n, n is the number of sampling time points in a cycle, VM curr l3 cache Indicates the L3 cache utilization of the cloud host in the current cycle.

[0071] After obtaining the L3 cache utilization of the cloud host in the current cycle, the L3 cache utilization of the cloud host in the previous cycle is calculated based on the L3 cache utilization of the cloud host in the current cycle and the L3 cache utilization of the cloud host sampled at multiple sampling time points. The calculation formula is as follows:

[0072]

[0073] Among them, VM curr l3 cache Indicates the L3 cache utilization of the cloud host in the current cycle, VM l3_cache(i) is the L3 cache utilization of the cloud host sampled at the i-th sampling time point, i = 1, 2, ..., n, n is the number of sampling time points in a cycle, VM pre_l3_cache Indicates the L3 cache utilization of the cloud host in the last cycle.

[0074] For example, if the current cycle L3 cache utilization of the cloud host satisfies the following formula, it means that the L3 cache utilization of the cloud host drops by more than the threshold threshold, and it is marked as a high-priority cloud host. Then, the L3 cache utilization of the cloud host is monitored in the reverse order of the set cloud host priority. The formula is:

[0075]

[0076] Among them, VM curr l3 cache Indicates the L3 cache utilization of the cloud host in the current cycle, VM pre_l3_cache Indicates the L3 cache utilization of the cloud host in the last cycle. The threshold for the change in L3 cache utilization.

[0077] For example, if the current cycle L3 cache utilization of the cloud host satisfies the following formula, it means that the L3 cache utilization of the cloud host has risen above the threshold, and it is marked as a noise cloud host. Then, the L3 cache utilization of the cloud host is monitored in the reverse order of the set cloud host priority. The formula is:

[0078]

[0079] Among them, VM curr l3 cache Indicates the L3 cache utilization of the cloud host in the current cycle, VM pre_l3_cache Indicates the L3 cache utilization of the cloud host in the last cycle. The threshold for the change in L3 cache utilization.

[0080] Step 204: If the average value is less than the first preset value, add the host machine to the candidate host machine set.

[0081] For example, if the host's RMoutlet_temp Less than w threshold , it means that the outlet temperature of the host machine is not high, and the host machine is determined as a candidate host machine.

[0082] Step 205, obtaining the resource load values ​​of the candidate host machines in the candidate host machine set, calculating the scores of the candidate host machines according to the resource load values ​​of the candidate host machines, and taking the candidate host machine with the lowest score as the target host machine; the scores of the candidate host machines are positively correlated with the resource load values ​​of the candidate host machines.

[0083] This step may be specifically referred to as the above step 104, and will not be described in detail here.

[0084] Optionally, step 205 may specifically include:

[0085] Sub-step 2051: obtaining a resource load value of the cloud host to be migrated, and adding the resource load value of the cloud host to be migrated and the resource load value of the candidate host to obtain a total resource load value of the candidate host;

[0086] Sub-step 2052: normalizing the total resource load value of the candidate host machine to obtain a normalized total resource load value;

[0087] Sub-step 2053: obtaining the weight of the candidate host machine, and calculating the score of the candidate host machine according to the weight of the candidate host machine and the normalized total resource load value.

[0088] For example, for sub-steps 2051 to 2053, the resource load value of the candidate host may be

[0089] PM′ metric(i) =PM metric(i) +VM metric(j)

[0090] The average value of the outlet temperature of the candidate host machine can be the memory utilization or the CPU utilization. The resource load value of the cloud host to be migrated is obtained. If the sum of the resource load value of the cloud host to be migrated and the resource load value of the candidate host machine does not exceed the resource load upper limit of the candidate host machine, the candidate host machine is determined as the target host machine corresponding to the cloud host to be migrated. The calculation formula is as follows:

[0091] Among them, VM metric(j) Indicates the resource load value of the cloud host to be migrated, PM metric(i) Represents the resource load value of the candidate host, PM′ metric(i) The final resource load value of the candidate host.

[0092] For example, obtain the cloud host to be migrated with the lowest priority on the host to be migrated and obtain the resource load value of the cloud host to be migrated. Obtain the resource load value and resource load upper limit value of each candidate host. When the sum of the resource load value of the cloud host to be migrated with the lowest priority and the resource load value of the candidate host does not exceed the resource load upper limit value of the candidate host, it is proved that the candidate host can bear the resource load of the cloud host to be migrated. In this way, according to the order of priority, for each cloud host to be migrated, the target host corresponding to the cloud host to be migrated can be screened out from the candidate hosts.

[0093] For example, when a cloud host to be migrated corresponds to multiple candidate hosts, the multiple candidate hosts can be sorted. The resource load values ​​of the multiple candidate hosts are normalized, the properties of the multiple candidate hosts are obtained, and the weights of the multiple candidate hosts are determined according to the properties of the multiple candidate hosts. In other words, the memory utilization, CPU utilization and air outlet temperature of the multiple candidate hosts are obtained and used as weight values ​​to calculate the weights of the multiple candidate hosts.

[0094] For example, the scores of the multiple candidate hosts are determined according to the weights of the multiple candidate hosts and the normalized resource load values, and the cloud host to be migrated is migrated to the candidate host with the lowest score. In order to facilitate measurement within the same interval, the resource load values ​​of the multiple candidate hosts are normalized. The weights corresponding to the attributes of the multiple candidate hosts can be pre-configured. The formula for calculating the scores of the candidate hosts is as follows:

[0095]

[0096] Among them, PM metric(i) represents the resource load value of the i-th candidate host, PM metric_avg(i) represents the average resource load value of the candidate host, m represents the number of candidate hosts, and W metric Represents the weight value, PM score(i) Represents the score of the i-th candidate host.

[0097] Step 206: Migrate the cloud host to be migrated in the host machine to be migrated to the target host machine.

[0098] This step may be specifically referred to as the above step 105, and will not be described in detail here.

[0099] Optionally, the method further includes:

[0100] Step 207: Add the host machines to be migrated to the first list in descending order according to the outlet temperatures of the host machines to be migrated;

[0101] Step 208: Add the candidate host machines to the second list in order from low to high according to the outlet temperatures of the candidate host machines;

[0102] Step 209: If the first list is empty and the second list is empty, a notification indicating that migration is impossible is sent;

[0103] Step 210: If the first list is not empty and the second list is empty, send an alarm notification.

[0104] For step 207-step 210, the host machines are classified according to the set thresholds, and the host machines to be migrated and the candidate host machines need to be migrated. If the average temperature of the host machine outlet within the preset period exceeds the set threshold, it is determined as a host machine to be migrated, added to the first list, that is, the source host machine list, and sorted from high to low according to the average temperature value. Those that do not exceed the set threshold are determined as candidate host machines, and are added to the second list, that is, the candidate host machine list, and sorted from low to high according to the average temperature value. If one of the first list or the second list is empty, it means that migration cannot be performed this time; if the first list is not empty and the second list is empty, an alarm notification needs to be sent to relevant personnel.

[0105] For example, the first list may also be referred to as a source computing node list, and the second list may also be referred to as a target computing node list.

[0106] Optionally, the method further includes:

[0107] Step 211: determine the host machine that does not include the cloud host to be migrated as the host machine to be filtered;

[0108] Step 212: Screen the host machines to be filtered according to the resource load of the cloud host running on the host machines to be filtered to obtain the candidate host machines.

[0109] For example, the host to be filtered is screened according to the cloud host resource load through the static scheduling algorithm to obtain the candidate host. Specifically, the available load of the host to be filtered is calculated according to the load of the host to be filtered, the over-provisioning ratio of the host to be filtered, and the resource load of the cloud host. The calculation formula is as follows:

[0110]

[0111] Among them, PM avail_flavor(i) Indicates the available load of the host to be filtered, PM flavor(i) Indicates the load of the host to be filtered, PMflavor_ratio Indicates the over-provisioning ratio of the host to be filtered, VM flavor(j) Indicates the resource load of the cloud host, and vms indicates the number of running cloud hosts.

[0112] For example, choosing a suitable target host is crucial to the dynamic resource scheduling algorithm and will directly affect the effect of dynamic resource scheduling. If the selected target host is not ideal, it will cause unnecessary migration, and even secondary migration or chain migration, increasing the workload of other hosts. The cloud host should be judged whether to migrate and adjust the host based on the L3 cache noise, provided that it does not affect the resource allocation and operation of the data center using the static scheduling algorithm. Therefore, in response to this problem, when selecting a suitable target host for the cloud host, it is necessary to consider the resource quota and resource load of each host on the cloud platform at the same time to ensure that the original order of the cloud platform will not be disrupted.

[0113] The test scenario of the embodiment of the present invention is as follows. First, prepare a cloud platform with at least 10 or more host machines. Secondly, create a batch of cloud hosts in the cloud platform and distribute them on each host machine to ensure that the resource allocation rate of the cloud platform reaches between 60% and 80%. The cloud host simulates the use of business scenarios to overload the computing resources used by the host machine, so that the temperature of the host machine is overloaded and exceeds the specified threshold temperature. Then, set the dynamic resource scheduling method, and ensure that the resources on the computing node are dynamically adjusted at intervals. The conditions for triggering dynamic resource scheduling include alarm triggering and regular triggering. Finally, the cloud host on the host machine will adjust the resources according to the set dynamic resource scheduling method. When the host machine uses an overloaded computing node and the temperature is too high, the cloud host running on it will be regularly migrated to other hosts with too low temperatures, thereby reducing the use of the host machine and the temperature gradually returning to the threshold range.

[0114] In an embodiment of the present invention, a host machine set is obtained, and a host machine in the host machine set that meets a first condition is added to a candidate host machine set, the first condition comprising: the host machine is in an enabled state and a cloud host is not running, the outlet temperature of the host machine in the host machine set that does not meet the first condition is obtained, a host machine whose outlet temperature meets a second condition is determined as a host machine to be migrated, and a cloud host to be migrated is determined from the cloud hosts running on the host machine to be migrated, and a host machine whose outlet temperature does not meet the second condition is added to the candidate host machine set, and the resource load values ​​of the candidate host machines in the candidate host machine set are obtained. According to the resource load values ​​of the candidate host machines, the scores of the candidate host machines are calculated, and the candidate host machine with the lowest score is selected as the target host machine, and the cloud hosts to be migrated in the host machines to be migrated are migrated to the target host machine. The embodiment of the present invention divides the host machine to be migrated and the candidate host machine according to the outlet temperature of the host machine. When the temperature of the host machine to be migrated is too high, a suitable target host machine is selected for the cloud host to be migrated running on the host machine to be migrated, so as to migrate the cloud host to be migrated to the target host machine. The outlet temperature of the host machine is used as a key factor to adjust the cloud host resources to reduce the overload of the host machine to be migrated, thereby reducing the outlet temperature of the host machine to be migrated.

[0115] Figure 3 is a flowchart of the steps of screening a host machine to be migrated and a target host machine provided by an embodiment of the present invention, see Figure 3 , the method may include:

[0116] Step S1: Filter computing nodes that need to be adjusted from all computing nodes in the data center.

[0117] Step S2: If the server enables a non-running cloud host, a target computing node for cloud host migration adjustment is obtained.

[0118] Step S3: If the server enables and runs the cloud host, a target computing node for cloud host migration adjustment is obtained.

[0119] Step S4: Obtain N cycles of air outlet temperature monitoring data of each computing node and calculate the average value.

[0120] Step S5: compare the average air outlet temperature of each computing node with the set air outlet temperature threshold.

[0121] Step S6: If the average temperature of the air outlet is greater than the temperature threshold, proceed to step S3.

[0122] Step S7: If the average temperature of the air outlet is less than or equal to the temperature threshold, a target computing node for cloud host migration adjustment is obtained.

[0123] Figure 4is a flowchart of the steps for selecting a cloud host to be migrated provided by an embodiment of the present invention, see Figure 4 , the method may include:

[0124] Step M1: Screen computing nodes and cloud hosts with L3 cache noise.

[0125] Step M2: Filter according to the special attributes of the cloud host.

[0126] Step M3: Obtain the resource historical monitoring data of all computing nodes and cloud hosts within N cycles and calculate the average value as the resource load value.

[0127] Step M4: Circularly select computing nodes and noisy cloud hosts that need to be adjusted from the source computing nodes.

[0128] Step M5: sort the cloud hosts on the computing nodes in ascending and descending order according to the set priorities.

[0129] Step M6: Find potential high-priority cloud hosts in order from the sequential list of cloud hosts.

[0130] Step M7: Obtain the current cycle L3 cache utilization and the previous cycle L3 cache utilization of the cloud host respectively.

[0131] Step M8: When the L3 cache utilization rate of the cloud host decreases by no more than the threshold, proceed to step M9.

[0132] Step M9: Select the next potential priority cloud host.

[0133] Step M10: When the L3 cache utilization of the cloud host decreases by more than a variable threshold, find potential noisy cloud hosts in the reverse order list of cloud hosts.

[0134] Step M11: Determine whether the potential noise cloud host and the potential high priority cloud host are the same.

[0135] Step M12: If the potential noise cloud host and the potential high-priority cloud host are the same, then the next computing node is determined.

[0136] Step M13: If the potential noisy cloud host is different from the potential high-priority cloud host, obtain the current cycle L3 cache utilization and the previous cycle L3 cache utilization of the potential noisy cloud host.

[0137] Step M14: determine whether the L3 cache increase exceeds the change threshold.

[0138] Step M15: When the L3 cache utilization increase value of the cloud host does not exceed the change threshold, determine from the reverse list of cloud hosts whether the next cloud host is a potential noise cloud host.

[0139] Step M16: When the L3 cache utilization of the cloud host decreases by more than the change threshold, a mapping relationship is established between the high-priority cloud host and the noisy cloud host and the computing node, and the computing node is classified as a computing node that needs to release noise.

[0140] Step M17: Mark the computing nodes that are not in the computing node list that needs to release noise as target computing nodes.

[0141] Step M18: Return the computing nodes that need to release noise and the target computing node list.

[0142] Figure 5 is a flowchart of the steps for selecting a target host for a cloud host to be migrated provided by an embodiment of the present invention, see Figure 5 , the method may include:

[0143] Step N1: Select a suitable target computing node for the migration cloud host from the target computing node list.

[0144] Step N2: Preliminarily select a list of suitable computing nodes based on the type and special attributes of the cloud host.

[0145] Step N3: Determine whether there is a computing node that satisfies the requirement.

[0146] Step N4: If there is no computing node that meets the conditions, select the next target computing node.

[0147] Step N5: If there are computing nodes that meet the conditions, filter out a list of suitable target computing nodes according to the type of cloud host.

[0148] Step N6: Obtain monitoring indicators of the cloud host resources within N cycles and calculate the average value.

[0149] Step N7: Integrate the actual resource utilization of the cloud host with the actual resource utilization of the computing node.

[0150] Step N8: Determine whether it is within the threshold range.

[0151] Step N9: If it is within the threshold range, select the next target computing node.

[0152] Step N10: If it is within the threshold range, add it to the new target computing node list.

[0153] Step N11: normalize each resource load value and outlet temperature of the target computing node and multiply them by their weight values ​​to calculate the node score, and then sort them.

[0154] Step N12: Select the computing node with the lowest score as the final target computing node, and establish a migration mapping relationship.

[0155] Figure 6 A cloud host migration device 30 provided in an embodiment of the present invention may include:

[0156] The first adding module 301 is used to obtain a host set, and add a host in the host set that meets a first condition to a candidate host set; the first condition includes: the host is in an enabled state and no cloud host is running;

[0157] A first determination module 302 is used to obtain the outlet temperature of the host machine in the host machine set that does not meet the first condition, determine the host machine whose outlet temperature meets the second condition as the host machine to be migrated, and determine the cloud host to be migrated from the cloud hosts running on the host machine to be migrated;

[0158] A second adding module 303, configured to add the host machine whose air outlet temperature does not satisfy the second condition into the candidate host machine set;

[0159] The calculation module 304 is used to obtain the resource load value of the candidate host in the candidate host set, calculate the score of the candidate host according to the resource load value of the candidate host, and use the candidate host with the lowest score as the target host; the score of the candidate host is positively correlated with the resource load value of the candidate host;

[0160] The migration module 305 is used to migrate the cloud host to be migrated in the host to be migrated to the target host.

[0161] Optionally, the determining module includes:

[0162] A first calculation submodule, used to calculate the average value of the air outlet temperature of the host machine within a preset time range;

[0163] A first determining submodule, configured to determine the host machine as the host machine to be migrated if the average value is greater than or equal to a first preset value;

[0164] The second adding module includes:

[0165] An adding submodule is used to add the host machine to the candidate host machine set if the average value is less than the first preset value.

[0166] Optionally, the determining module includes:

[0167] An acquisition submodule, used to acquire the utilization rate of the third-level cache of the cloud host running on the host to be migrated;

[0168] The second determination submodule is used to determine the cloud host to be migrated in the host machine to be migrated according to the third-level cache utilization rate of the cloud host running on the host machine to be migrated.

[0169] Optionally, the cloud host to be migrated includes: a high-priority cloud host and a noisy cloud host; the high-priority cloud host is a cloud host whose level 3 cache utilization decrease value exceeds a second preset value; the noisy cloud host is a cloud host whose level 3 cache utilization increase value exceeds the second preset value;

[0170] The second determining submodule includes:

[0171] An acquisition unit, used to acquire the third-level cache utilization rate of the cloud host in the current cycle and the third-level cache utilization rate of the previous cycle;

[0172] A calculation unit, used to calculate the difference between the third-level cache utilization rate of the cloud host in the current cycle and the third-level cache utilization rate in the previous cycle;

[0173] a first determining unit, configured to determine that the cloud host is a noisy cloud host if the difference is less than a product of a second preset value and an L3 cache utilization rate of a previous cycle;

[0174] The second determining unit is configured to determine that the cloud host is a high priority cloud host if the difference is less than an inverse value of a product of the second preset value and the L3 cache utilization rate of a previous cycle.

[0175] Optionally, the computing module includes:

[0176] The summing operation submodule is used to obtain the resource load value of the cloud host to be migrated, and perform sum operation on the resource load value of the cloud host to be migrated and the resource load value of the candidate host to obtain the total resource load value of the candidate host;

[0177] A normalization processing submodule, used to normalize the total resource load value of the candidate host machine to obtain a normalized total resource load value;

[0178] The second calculation submodule is used to obtain the weight of the candidate host machine, and calculate the score of the candidate host machine according to the weight of the candidate host machine and the normalized total resource load value.

[0179] Optionally, the device further comprises:

[0180] A second determination module is used to determine the host machine that does not include the cloud host to be migrated as the host machine to be filtered;

[0181] The filtering module is used to filter the host machines to be filtered according to the types and properties of the cloud hosts running on the host machines to be filtered, so as to obtain the candidate host machines.

[0182] Optionally, the device further comprises:

[0183] A third adding module, configured to add the host machines to be migrated to the first list in descending order according to the outlet temperatures of the host machines to be migrated;

[0184] A fourth adding module, configured to add the candidate host machines to the second list in order from low to high according to the outlet temperatures of the candidate host machines;

[0185] A first sending module, configured to send a notification of failure to migrate if the first list is empty and the second list is empty;

[0186] The second sending module is configured to send an alarm notification if the first list is not empty and the second list is empty.

[0187] In an embodiment of the present invention, a host machine set is obtained, and a host machine in the host machine set that meets a first condition is added to a candidate host machine set, the first condition comprising: the host machine is in an enabled state and a cloud host is not running, the outlet temperature of the host machine in the host machine set that does not meet the first condition is obtained, a host machine whose outlet temperature meets a second condition is determined as a host machine to be migrated, and a cloud host to be migrated is determined from the cloud hosts running on the host machine to be migrated, and a host machine whose outlet temperature does not meet the second condition is added to the candidate host machine set, and the resource load values ​​of the candidate host machines in the candidate host machine set are obtained. According to the resource load values ​​of the candidate host machines, the scores of the candidate host machines are calculated, and the candidate host machine with the lowest score is selected as the target host machine, and the cloud hosts to be migrated in the host machines to be migrated are migrated to the target host machine. The embodiment of the present invention divides the host machine to be migrated and the candidate host machine according to the outlet temperature of the host machine. When the temperature of the host machine to be migrated is too high, a suitable target host machine is selected for the cloud host to be migrated running on the host machine to be migrated, so as to migrate the cloud host to be migrated to the target host machine. The outlet temperature of the host machine is used as a key factor to adjust the cloud host resources to reduce the overload of the host machine to be migrated, thereby reducing the outlet temperature of the host machine to be migrated.

[0188] The present invention also provides an electronic device, see Figure 7 , including: a processor 701, a memory 702, and a computer program 7021 stored in the memory and executable on the processor, wherein when the processor executes the program, the cloud host migration method of the aforementioned embodiment is implemented.

[0189] The present invention also provides a readable storage medium, when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the cloud host migration method of the aforementioned embodiment.

[0190] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0191] It should be noted that the various information and data obtained in the embodiments of the present invention are all obtained with the authorization of the information / data holder.

[0192] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming voice. It should be understood that various programming voices can be utilized to realize the content of the present invention described herein, and the description of the above specific voices is for the purpose of disclosing the best mode of the present invention.

[0193] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0194] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

[0195] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0196] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the sorting device according to the present invention. The present invention may also be implemented as a device or apparatus program for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0197] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0198] The user information (including but not limited to the user's device information, user personal information, etc.) and related data involved in the present invention are all information authorized by the user or by all parties.

[0199] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0200] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0201] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A cloud host migration method, characterized in that: The method comprises: Obtain a host set, and add a host in the host set that meets a first condition to a candidate host set; the first condition includes: the host is in an enabled state and no cloud host is running; Obtaining the outlet temperature of the host machine in the host machine set that does not meet the first condition, determining the host machine whose outlet temperature meets the second condition as the host machine to be migrated, and determining the cloud host to be migrated from the cloud hosts running on the host machine to be migrated; Adding the host machine whose air outlet temperature does not meet the second condition to the candidate host machine set; Obtain resource load values ​​of candidate host machines in the candidate host machine set, calculate scores of the candidate host machines according to the resource load values ​​of the candidate host machines, and use the candidate host machine with the lowest score as the target host machine; the scores of the candidate host machines are positively correlated with the resource load values ​​of the candidate host machines; Migrate the cloud host to be migrated in the host to be migrated to the target host.

2. The method according to claim 1, characterized in that The step of determining the host machine whose air outlet temperature satisfies the second condition as the host machine to be migrated includes: Calculate the average temperature of the host machine outlet within a preset time range; If the average value is greater than or equal to a first preset value, the host machine is determined as the host machine to be migrated; The adding the host machine whose air outlet temperature does not satisfy the second condition into the candidate host machine set includes: If the average value is less than the first preset value, the host machine is added to the candidate host machine set.

3. The method according to claim 1, characterized in that The step of determining the cloud host to be migrated from the cloud hosts running on the host to be migrated includes: Obtaining a level 3 cache utilization rate of the cloud host running on the host to be migrated; The cloud host to be migrated in the host to be migrated is determined according to the level 3 cache utilization of the cloud host running on the host to be migrated.

4. The method according to claim 3, characterized in that The cloud hosts to be migrated include: high-priority cloud hosts and noisy cloud hosts; the high-priority cloud hosts are cloud hosts whose L3 cache utilization rate decreases by a value exceeding a second preset value; the noisy cloud hosts are cloud hosts whose L3 cache utilization rate increases by a value exceeding the second preset value; The step of determining the cloud host to be migrated in the host to be migrated according to the utilization rate of the third-level cache of the cloud host running on the host to be migrated includes: Obtaining the third-level cache utilization rate of the cloud host in the current cycle and the third-level cache utilization rate of the previous cycle; Calculate the difference between the third-level cache utilization rate of the cloud host in the current cycle and the third-level cache utilization rate in the previous cycle; If the difference is less than the product of the second preset value and the L3 cache utilization rate of the previous cycle, the cloud host is determined to be a noisy cloud host; If the difference is smaller than the inverse value of the product of the second preset value and the L3 cache utilization rate of the previous cycle, the cloud host is determined to be a high-priority cloud host.

5. The method according to claim 1, characterized in that The calculating the score of the candidate host machine according to the resource load value of the candidate host machine includes: Obtaining a resource load value of the cloud host to be migrated, and performing a sum operation on the resource load value of the cloud host to be migrated and the resource load value of the candidate host to obtain a total resource load value of the candidate host; Normalizing the total resource load value of the candidate host machine to obtain a normalized total resource load value; The weight of the candidate host machine is obtained, and the score of the candidate host machine is calculated according to the weight of the candidate host machine and the normalized total resource load value.

6. The method according to claim 1, characterized in that The method further comprises: Determine the host machine that does not include the cloud host to be migrated as the host machine to be filtered; The host machines to be filtered are screened according to the resource load of the cloud host running on the host machines to be filtered to obtain the candidate host machines.

7. The method according to claim 1, characterized in that The method further comprises: Adding the host machines to be migrated to the first list in descending order according to the outlet temperatures of the host machines to be migrated; Adding the candidate host machines to the second list in order from low to high according to the outlet temperatures of the candidate host machines; If the first list is empty and the second list is empty, sending a notification that migration is impossible; If the first list is not empty and the second list is empty, an alarm notification is sent.

8. A resource scheduling device, characterized in that: The device comprises: A first adding module is used to obtain a host set, and add a host in the host set that meets a first condition to a candidate host set; the first condition includes: the host is in an enabled state and no cloud host is running; A first determination module is used to obtain the air outlet temperature of the host machine in the host machine set that does not meet the first condition, determine the host machine whose air outlet temperature meets the second condition as the host machine to be migrated, and determine the cloud host to be migrated from the cloud hosts running on the host machine to be migrated; A second adding module, configured to add the host machine whose air outlet temperature does not satisfy the second condition into the candidate host machine set; A calculation module, used to obtain the resource load value of the candidate host in the candidate host set, calculate the score of the candidate host according to the resource load value of the candidate host, and use the candidate host with the lowest score as the target host; the score of the candidate host is positively correlated with the resource load value of the candidate host; The migration module is used to migrate the cloud host to be migrated in the host to be migrated to the target host.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 8 when executing the program.

10. A readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute any one of the methods described in claims 1-8.

Images