Recommendation method and device for server shelving positions and storage medium

Abstract

The invention discloses a recommendation method and device for server shelving positions and a storage medium, and relates to the technical field of intelligent planning and data center operation and maintenance management. The method comprises the following steps of: in a cabinet group which simultaneously meets a network requirement, a refrigeration requirement, a space requirement, an electric power requirement and a maximum resource requirement, determining a machine position distribution scheme which simultaneously meets a space requirement and an electric power requirement of each server to be racked; and according to the distribution income of each stand distribution scheme, determining a shelving stand recommendation scheme of the to-be-shelved server. According to the method and the device, on the basis of considering the multi-dimensional resource requirements of the to-be-racked server, the recommendation cost is reduced and the recommendation efficiency and the recommendation effect are improved while the optimal recommendation machine position is automatically searched under the condition that a large model is not needed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent planning and data center operation and maintenance management, in particular to a server rack position recommendation method, device and storage medium. BACKGROUND

[0002] Server rack optimization position planning refers to using artificial experience or advanced technical means such as artificial intelligence, automation tools, optimization algorithms and intelligent management systems to realize the automation, precision and high efficiency of the whole process from planning to installation of the server, so as to improve the overall performance, resource utilization, operation and maintenance efficiency of the data center and reduce the cost. At present, the main means of server optimization position planning include: (1) Artificial planning, which is strongly dependent on personal experience and lacks automated processes, with low recommendation efficiency; and different personnel will have different recommendation methods, which is not sustainable.

[0003] (2) Intelligent capacity management system, which can automatically recommend positions combined with cabinet capacity, has certain sustainability and universality, but does not consider other variables such as power load, maintenance cost and business requirements, and still needs manual intervention, otherwise the actual recommendation effect is not good.

[0004] (3) Artificial intelligence and machine learning method, which trains the model, not only with large training overhead and high cost, but also for small and medium-sized data centers, there is not enough data to support training, so it is difficult to get a perfectly fitted model, and the actual recommendation effect is not good. SUMMARY

[0005] In view of the defects in the prior art, the technical problem solved by the present application is how to automatically find the optimal position according to the parameters required by the server rack without using a large model, so as to reduce the recommendation cost while improving the recommendation efficiency and effect.

[0006] To achieve the above purpose, in a first aspect, the present application provides a server rack position recommendation method, which comprises the following steps: All cabinet groups that meet the feasibility conditions in terms of available network capacity and available cooling capacity in the cooling area are selected as a first cabinet group set; In the first cabinet group set, all cabinet groups that meet the cabinet capacity requirement in terms of available space capacity and the cabinet power requirement in terms of available power capacity are selected as a second cabinet group set; In the cabinet space capacity and power capacity consumed by each server to be installed, the maximum cabinet space capacity and the maximum power capacity are determined; In the secondary cabinet group set collection, all cabinet groups capable of simultaneously carrying the cabinet space capacity with the largest consumption and the power capacity with the largest consumption are determined as a tertiary cabinet group set collection; In the tertiary cabinet group set collection, all cabinet groups capable of simultaneously satisfying the space requirement and the power requirement of each server to be mounted are determined as a cabinet group to be recommended; In each cabinet group to be recommended, a seat allocation scheme simultaneously satisfying the space requirement and the power requirement of each server to be mounted is determined; An allocation benefit after the server to be mounted is allocated according to the seat allocation scheme is determined, the allocation benefit including a space benefit obtained according to the available space capacity, a power benefit obtained according to the available power capacity, a refrigeration benefit obtained according to the available refrigeration capacity, and a network benefit obtained according to the available network capacity; According to the allocation benefit of each seat allocation scheme, a mounting seat recommendation scheme of the server to be mounted is determined.

[0007] In combination with the first aspect, in an implementation manner, a calculation formula of the allocation benefit v is as follows: ; Wherein: represents the space benefit, represents a weight of the space benefit; ; K represents a cabinet quantity in a current cabinet group which needs to be calculated, and k represents a cabinet serial number; represents an available space capacity utilization rate of the cabinet k, is a ratio of a pre-allocated space capacity of the cabinet k to the available space capacity; represents an average available space capacity utilization rate, ; represents the power benefit, represents a weight of the power benefit; ; represents an available power capacity utilization rate of the cabinet, is a ratio of a pre-allocated power capacity of the cabinet k to the available power capacity; represents the refrigeration benefit, represents a weight of the refrigeration benefit; ; This is the ratio of the pre-allocated cooling capacity to the available cooling capacity of the cooling zone where the current server rack is located; Represents network revenue. represent The weights; ; This is the ratio of the pre-allocated network capacity to the available network capacity of the current rack group.

[0008] In conjunction with the first aspect, in one implementation, the process of determining the allocation revenue after allocating servers to be deployed according to the server allocation scheme, and determining the recommended server placement scheme for the servers to be deployed based on the allocation revenue of each server allocation scheme, includes: Step A: Select any one server that has not been assigned a rack. Among all the rack groups to be recommended, determine the rack that meets the current server's space and power requirements as a branch rack. Use the mapping relationship between the current server and each branch rack as a rack allocation scheme. Step B: Retain the seat allocation schemes that meet the continued recommendation criteria; Step C: Mark the current server allocation, add nodes for each reserved slot allocation scheme, and then proceed to Step B; Step D: After all servers to be racked are marked for allocation, the reserved rack allocation scheme will be used as the recommended rack allocation scheme for the servers to be racked.

[0009] In conjunction with the first aspect, in one implementation, the continuous recommendation conditions include: When a rack allocation scheme is adopted, the rack group to which the rack allocation scheme belongs must simultaneously meet the feasibility conditions, the rack capacity requirements, and the rack power requirements. and: When adopting a racking allocation scheme, the upper bound of the revenue from the racking allocation scheme is above the lower bound of the revenue. The upper bound of the revenue is obtained based on the revenue allocated by the server allocation scheme and the maximum revenue allocated to the remaining servers to be racked when this server allocation scheme is adopted. The lower bound of the revenue is obtained by allocating all remaining servers to be racked to racks that simultaneously meet the space and power requirements when adopting this rack allocation scheme.

[0010] In conjunction with the first aspect, in one implementation, the upper bound of the revenue is... ; ; The process for determining the lower bound of revenue includes: determining the subsequent branch scheme for each rack allocation scheme, wherein the subsequent branch scheme is: when adopting this rack allocation scheme, all other servers to be racked are allocated to racks that simultaneously meet the space and power requirements. Each station allocation scheme and its corresponding subsequent branch scheme are combined into a complete branch scheme; Determine the distribution revenue for each complete branch scheme, and use the maximum distribution revenue as the lower bound of the revenue.

[0011] In conjunction with the first aspect, in one implementation, the process of adding a node includes: selecting any one server that has not been assigned a flag, determining, within the rack group to which the rack allocation scheme belongs, a rack that meets the current server's space and power requirements, and using the mapping relationship between the rack and the current server as a node in the current rack allocation scheme.

[0012] In conjunction with the first aspect, in one implementation, the feasibility conditions include: The available network capacity of the rack group is greater than the pre-allocated network capacity, which is the sum of the network capacity consumed by all servers to be deployed to the rack group. and: The available cooling capacity of the cooling area where the rack group is located is above the pre-allocated cooling capacity. The pre-allocated cooling capacity is the sum of the cooling capacity consumed by all servers to be racked when all servers to be racked are deployed to this rack group. The rack capacity requirements include: The available space capacity of each rack in the rack group is greater than the pre-allocated space capacity. The pre-allocated space capacity is the sum of the rack space capacity consumed by all the servers to be racked when they are deployed to the rack group. The power requirements for the cabinet include: The available power capacity of each rack in the rack group is greater than the pre-allocated power capacity, which is the sum of the power capacity consumed by all servers to be deployed to the rack group.

[0013] In conjunction with the first aspect, in one implementation, the criterion for determining whether the system can simultaneously handle the maximum rack space consumption and the maximum power consumption is: There are racks in the rack group whose maximum available power capacity exceeds the maximum power consumption capacity. and: There are racks in the rack group whose maximum available space capacity exceeds the rack space capacity that consumes the most space. The criteria for determining all rack groups that can simultaneously meet the space and power requirements of each server to be racked are as follows: After all servers to be racked are assigned to the same rack group, the available space capacity and available power capacity of each rack in that rack group are both positive numbers.

[0014] Secondly, embodiments of this application provide a server racking recommendation management device, which includes a processor, a memory, and a server racking recommendation program stored in the memory and executable by the processor. When the server racking recommendation program is executed by the processor, it implements the method provided in the first aspect.

[0015] Thirdly, embodiments of this application provide a computer-readable storage medium storing a recommendation program for server rack positions, wherein the computer program, when executed, implements the method provided in the first aspect.

[0016] Compared with the prior art, the advantages of this application are: This application first filters the rack groups from large to small "granularity" based on the resource requirements (space, power, cooling, and network) of the servers to be racked, thereby gradually eliminating rack groups that do not meet the requirements.

[0017] Furthermore, this application obtains multiple rack allocation schemes that simultaneously meet the space and power requirements of each server to be racked in the remaining rack groups through an "exhaustive" approach. Based on the allocation benefits (space benefits, power benefits, cooling benefits, and network benefits) of each rack allocation scheme, a relatively preferred rack allocation scheme for the servers to be racked is obtained.

[0018] Therefore, this application, taking into account the resource requirements (space, power, cooling, and network) of the servers to be racked, combines the above-mentioned reasonable screening method with the above-mentioned racking location recommendation scheme determined by allocating revenue; thereby achieving automatic finding of optimal recommended racking locations without requiring the large models of existing technologies, while reducing recommendation costs and improving recommendation efficiency and effectiveness, specifically: (1) It does not require the large models found in existing technologies; (2) Compared with the manual planning method in the existing technology, it significantly improves recommendation efficiency and recommendation effect; (3) Compared with the intelligent capacity management system in the prior art, it significantly improves the recommendation efficiency. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the rack group architecture in an embodiment of this application; Figure 2 This is a flowchart illustrating the recommended method for server rack space allocation in an embodiment of this application. Figure 3 This is a schematic diagram of the execution scheme of S5 in the embodiments of this application; Figure 4 This is a schematic diagram of the hardware structure of the recommended management device for server rack positions involved in the embodiments of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0024] Firstly, embodiments of this application provide a method for recommending server rack locations, wherein the server rack location in this method is a data center, see [link to relevant documentation]. Figure 1 As shown, the data center architecture is as follows: (1) A data center includes several computer rooms, each corresponding to one or more columns (columns 1 to 8 in the figure). (2) Each computer room includes several rack groups; (3) Each rack group includes several racks (rack 1 to rack 20 in each column of the figure). (4) Each computer room has multiple cooling zones equipped with necessary equipment (cold aisles in the diagram); all cabinets in the same cabinet group belong to the same cooling zone; (5) Servers from the same batch will be placed in the same rack group and connected to the same network switch. The number of network switch ports in the rack group has a margin to meet long-term rack requirements and will not become a resource bottleneck; (6) Servers of the same type consume the same rack space capacity, power capacity, cooling capacity and network capacity (switch port capacity).

[0025] Based on this, see Figure 2 As shown, the steps of the above method include: S1: All rack groups whose available network capacity and available cooling capacity in their respective cooling zones both meet the feasibility requirements are considered as a primary rack group set.

[0026] S1 performs a "large-particle" filter by traversing the rack groups, eliminating rack groups that do not meet the network and cooling requirements.

[0027] In one embodiment, the feasibility conditions for the rack group in S1 include (i.e., all conditions must be met simultaneously): The available network capacity of the rack group (i.e. the remaining available network capacity) is greater than the pre-allocated network capacity. The pre-allocated network capacity is the sum of the network capacity consumed by all servers to be deployed to the rack group. and: The available cooling capacity (i.e. the remaining available cooling capacity) of the cooling area where the rack group is located is above the pre-allocated cooling capacity. The pre-allocated cooling capacity is the sum of the cooling capacity consumed by all servers to be deployed to the rack group.

[0028] S2: In the set of primary rack groups, all rack groups whose available space capacity meets the rack capacity requirements and whose available power capacity meets the rack power requirements are considered as the set of secondary rack groups.

[0029] S2 uses a granular filtering method, traversing the racks to remove rack groups that do not meet the space and power requirements.

[0030] In one embodiment, the rack capacity requirement in S2 includes: The available space capacity (i.e. the remaining available space capacity) of each rack in the rack group is greater than the pre-allocated space capacity. The pre-allocated space capacity is the sum of the rack space capacity consumed by all servers to be deployed to the rack group.

[0031] The cabinet power requirements in S2 include: The available power capacity (i.e. the remaining available power capacity) of each rack in the rack group is above the pre-allocated power capacity, which is the sum of the power capacity consumed by all servers to be deployed to the rack group.

[0032] S3: Among the rack space and power capacity required for each server to be racked, determine the rack space and power capacity that consume the most. In the secondary rack group set, all rack groups that can simultaneously accommodate the rack space capacity and the power capacity with the highest consumption mentioned above will be considered as the tertiary rack group set.

[0033] S3 further filters out rack groups that do not meet the maximum requirements of the servers to be installed based on the "highest requirements" of the servers to be installed.

[0034] In one embodiment, the determination condition in S3 for being able to simultaneously handle the rack space capacity with the highest consumption and the power capacity with the highest consumption is as follows: There are cabinets in the cabinet group whose maximum available power capacity exceeds the aforementioned maximum power consumption capacity. and: There are racks in the rack group whose maximum available space capacity exceeds the rack space capacity that consumes the most space mentioned above.

[0035] S4: In the set of three-tier rack groups, identify all rack groups that can simultaneously meet the space and power requirements of each server to be racked, and use them as the set of rack groups to be recommended.

[0036] In one embodiment, the criterion for determining all rack groups in S4 that can simultaneously meet the space and power requirements of each server to be racked is: After all servers to be racked are assigned to the same rack group, the available space capacity and available power capacity of each rack in that rack group will be positive. The specific execution method can be as follows: First, sort the servers in descending order of rack space and power consumption. Then, attempt to assign each server to a rack in the same rack group that meets its space requirements (rack space consumption) and power requirements (power consumption). When each server can be successfully assigned, add the rack group to the set of rack groups to be recommended.

[0037] At this point, it is already possible to recommend a suitable server rack for each server to be installed, while simultaneously meeting the requirements for server space, power, cooling, and network. However, this application also needs to consider the subsequent allocation of servers, so it is necessary to select a more optimal server rack from the aforementioned set of recommended racks, considering the four dimensions of space, power, cooling, and network.

[0038] S5: In each rack group to be recommended, determine a rack allocation scheme that simultaneously meets the space and power requirements of each server to be racked (each rack group to be recommended may correspond to multiple schemes). The revenue allocation after allocating servers to be racked according to the rack allocation scheme is determined. The revenue allocation includes space revenue based on available space capacity, power revenue based on available power capacity, cooling revenue based on available cooling capacity, and network revenue based on available network capacity. Based on the allocation benefits of each rack allocation scheme, a recommended rack allocation scheme for the servers to be racked is determined.

[0039] Therefore, this application first filters the rack groups from large to small "granularity" according to the resource requirements (space, power, cooling and network) of the servers to be racked, so as to gradually eliminate rack groups that do not meet the requirements.

[0040] Furthermore, this application obtains multiple rack allocation schemes that simultaneously meet the space and power requirements of each server to be racked in the remaining rack groups through an "exhaustive" approach. Based on the allocation benefits (space benefits, power benefits, cooling benefits, and network benefits) of each rack allocation scheme, a relatively preferred rack allocation scheme for the servers to be racked is obtained.

[0041] Therefore, this application, taking into account the resource requirements (space, power, cooling, and network) of the servers to be racked, combines the above-mentioned reasonable screening method with the above-mentioned racking position recommendation scheme determined by allocating revenue; thereby achieving automatic finding of optimal recommended racking positions without the need for the large models in existing technologies, while reducing recommendation costs and improving recommendation efficiency and effectiveness.

[0042] In one embodiment, the formula for calculating the distribution benefit v of S5 is:

[0043] in: Represents spatial benefits, represent The weights; ; K represents the number of racks in the current rack group that needs to be calculated, and k represents the rack number; This represents the utilization rate of the available space capacity of rack k. The ratio of the pre-allocated space capacity (see above for how the pre-allocated space capacity is determined) to the available space capacity of rack k; This represents the average available space capacity utilization rate. .

[0044] Represents electricity revenue, represent The weights; ; This represents the utilization rate of the available power capacity of the server rack. The ratio of the pre-allocated power capacity (see above for how the pre-allocated power capacity is determined) to the available power capacity for cabinet k.

[0045] Represents cooling revenue. represent The weights; ; The ratio of the pre-allocated cooling capacity (see above for how to determine the pre-allocated cooling capacity) to the available cooling capacity of the cooling zone where the current cabinet group is located.

[0046] Represents network revenue. represent The weights; ; The ratio of the pre-allocated network capacity (see above for how the pre-allocated network capacity is determined) to the available network capacity for the current rack group.

[0047] The above weights can be set as needed, and must meet the following requirements: .

[0048] In one embodiment, the specific process of S5 includes: S501: "Branch" operation: Select any one server that has not been marked for allocation and select the rack that meets the space and power requirements of the current server (i.e., the server selected above) from all recommended rack groups. Then, select the rack that meets the space and power requirements of the current server (i.e., the server selected above) as the branch rack and assign the mapping relationship between the current server and each branch rack (i.e., assign the current server to each branch rack) as a rack allocation scheme.

[0049] S502: "Pruning" operation: Retain the stand allocation schemes that meet the continuous recommendation criteria, remove the stand allocation schemes that do not meet the continuous recommendation criteria, and proceed to S503.

[0050] The continuing recommendation conditions in S502 include (simultaneous fulfillment of): (1) When the rack allocation scheme is adopted (that is, after the current server is allocated to the branch rack), the rack group to which the rack allocation scheme belongs must simultaneously meet the above feasibility conditions (corresponding to available network capacity and available cooling capacity), rack capacity requirements (corresponding to available space capacity) and rack power requirements (corresponding to available power capacity). (2) When adopting the station allocation scheme, the upper limit of the revenue of the station allocation scheme is above the lower limit of the revenue.

[0051] The upper bound of revenue is obtained based on the revenue from the rack allocation scheme and the maximum revenue from the remaining servers to be racked when this rack allocation scheme is adopted (optimistic estimate calculation); The lower bound of revenue is obtained by allocating all remaining servers to be racked to racks that simultaneously meet both space and power requirements when this rack allocation scheme is adopted.

[0052] Specifically, the upper limit of the profit. That is, V corresponds to the allocation revenue of the above-mentioned machine station allocation scheme. This corresponds to the maximum benefit of allocating the remaining servers to be racked when adopting this rack allocation scheme; .

[0053] The process for determining the lower bound of the above-mentioned revenue includes: determining the subsequent branch schemes for each rack allocation scheme. The subsequent branch schemes are: when adopting this rack allocation scheme, all other servers to be racked are allocated to racks that simultaneously meet the space and power requirements; there may be multiple schemes, that is, one rack allocation scheme may correspond to multiple subsequent branch schemes.

[0054] Each station allocation scheme and its corresponding subsequent branch scheme are combined to form a complete branch scheme.

[0055] Determine the allocation payoff for each complete branching scheme, and use the maximum allocation payoff as the lower bound of the payoff; for example: Let y be the station allocation scheme, and let the complete set of branch schemes for y be... Then the lower bound of the benefit .

[0056] S503: Mark the current server for allocation, add nodes for each reserved server allocation scheme, and continue to execute the pruning operation of S502 (that is, the server allocation scheme in S502 at this time is the server allocation scheme reserved by S503) until all servers to be racked are marked for allocation.

[0057] The process of adding a node in S503 includes: selecting any server that has not been assigned a flag, identifying the rack that meets the current server's space and power requirements in the rack group to which the rack allocation scheme belongs, and using the mapping relationship between the rack and the current server as the node of the current rack allocation scheme.

[0058] The timing sequence of S503 is as follows: Mark the current server as assigned; Determine whether all servers to be deployed have been marked for allocation. If so, it means that all servers to be deployed have been allocated and proceed to S504; otherwise, add nodes for each reserved server allocation scheme and continue to perform the pruning operation in S502.

[0059] S504: The reserved rack allocation scheme will be used as the recommended rack allocation scheme for servers to be racked.

[0060] According to the operations of S501~S504, there may be one or a group (i.e. multiple) of airport stand allocation schemes that are retained until the end. At this time, the allocation revenue of the airport stand allocation scheme will be the same as the lower bound of the revenue.

[0061] In practical applications, the S5 workflow can be achieved by building a tree structure and continuously performing branching and pruning operations, for example, see [link to example]. Figure 3 As shown, the root node represents an empty allocation, internal nodes represent partial allocations, and leaf nodes represent complete allocation schemes. The set of all branches of this tree structure is denoted as... Each of the incomplete branches This represents a partial allocation scheme, where each complete branch... This represents a complete allocation scheme, which is also a solution to this problem.

[0062] Based on the above considerations, the branch and bound phase includes the following three steps.

[0063] Branch operations: For the current branch For any given server, select each available rack to branch out and generate new child nodes.

[0064] Bounding operation: For each branch The upper and lower bounds of revenue for the currently allocated server slots are calculated based on the above calculation method.

[0065] Pruning operation: For each branch If it meets the above continued recommendation criteria, it will be retained; otherwise, it will be pruned.

[0066] The application scenario of the above method is illustrated below through a specific embodiment.

[0067] Step A: During the initial planning of the data center, the data center cabinets are modularly grouped according to business needs. At the same time, the deployment plan for data center switches is determined based on network requirements, thus completing the initial setup of the cabinet groups and switches.

[0068] Step B: When planning the racking of data center servers, for a certain batch of server racking requirements, based on the data center rack group, switch and existing server data, the racking location planning process of the servers is mathematically abstracted and mathematically modeled according to the above method. Based on the proposed mathematical model, a server optimal racking location selection algorithm is further proposed, and the optimal racking location of the servers is calculated.

[0069] Specifically, this embodiment has two rack groups. and , Includes two racks and , Includes two racks and .

[0070] rack group The remaining network capacity is 10Gbps, and the remaining cooling capacity is 8kW; rack group The remaining network capacity is 8Gbps, and the remaining cooling capacity is 10kW.

[0071] server rack The remaining space capacity is 10U, and the remaining power capacity is 5kW; the rack The remaining space capacity is 8U, and the remaining power capacity is 6kW; the rack The remaining space capacity is 12U, and the remaining power capacity is 7kW; the rack The remaining space capacity is 9U, and the remaining power capacity is 5kW.

[0072] Objective function weights .

[0073] Two servers are to be deployed. and ,in It consumes 2U of space, generates 3kW of power, has a cooling capacity of 2kW, and a port capacity of 1Gbps. It consumes 3U of space, consumes 4kW of power, has a cooling capacity of 3kW, and a port capacity of 2Gbps.

[0074] The recommended rack space allocation scheme (preferred scheme) is as follows: .

[0075] Step C: When deploying the data center server, based on the preferred mounting location and the wiring specifications of the data center server network card and the uplink switch, use network cables to connect the server network port and the corresponding port of the switch to complete the wiring work. Step D: After the data center servers are deployed, the out-of-band management network of the data center is used to periodically acquire the traffic and power data of the uplink switch ports on the servers through the out-of-band monitoring platform. The power supply capacity and cooling capacity information are obtained through the data center power and environmental monitoring system. This acquired information is stored to provide data support for subsequent server deployment planning.

[0076] Specifically, the monitoring platform obtains configuration information such as device model, CPU, memory, hard drive, PCIe device, and power supply, as well as power data, from the server side via the out-of-band management network; it obtains port traffic data from the switch side; and it obtains power supply system capacity and estimated cooling capacity data from the power and environmental monitoring system. Combined with the service information carried by the servers, it further updates the consumption data of rack capacity, power capacity, cooling capacity, and network capacity (switch port capacity) for different server types, providing data support for subsequent server deployment planning.

[0077] Secondly, embodiments of this application provide a recommended management device for server rack positions. The recommended management device for server rack positions can be a personal computer (PC), a laptop computer, a server, or other devices with data processing capabilities.

[0078] Reference Figure 4 , Figure 4 This is a schematic diagram of the hardware structure of the server rack-mount recommendation management device involved in the embodiments of this application. In this embodiment, the hardware structure of the server rack-mount recommendation management device may include a processor, memory, communication interface, and communication bus.

[0079] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0080] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting internal components of the server rack-mount recommendation management device, as well as interfaces used for interconnecting the server rack-mount recommendation management device with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0081] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0082] The processor can be a general-purpose processor, which can call the server rack slot recommendation program stored in memory and execute the server rack slot recommendation method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the server rack slot recommendation program is called can refer to the various embodiments of the server rack slot recommendation method of this application, which will not be repeated here.

[0083] Those skilled in the art will understand that Figure 4 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0084] Thirdly, embodiments of this application also provide a computer-readable storage medium.

[0085] The computer-readable storage medium of this application stores a server rack position recommendation program, wherein when the server rack position recommendation program is executed by a processor, it implements the steps of the server rack position recommendation method as described above.

[0086] The method implemented when the server rack space recommendation program is executed can be referred to in the various embodiments of the server rack space recommendation method of this application, and will not be repeated here.

[0087] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0088] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0089] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0090] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0091] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0092] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, 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 is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0093] The above are merely specific embodiments of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the scope of the claims.

Claims

1. A method for recommending server rack positions, characterized in that, The method comprises the following steps: All rack groups meeting the feasibility conditions in terms of available network capacity and available refrigeration capacity of the refrigeration area are selected as a first-level rack group set; All rack groups meeting the rack capacity requirement in terms of available space capacity and meeting the rack power requirement in terms of available power capacity are selected as a second-level rack group set from the first-level rack group set; The maximum rack space capacity and the maximum power capacity consumed by each server to be mounted are determined from the rack space capacity and the power capacity consumed by each server to be mounted; All rack groups capable of simultaneously carrying the maximum rack space capacity and the maximum power capacity are selected as a third-level rack group set from the second-level rack group set; All rack groups capable of simultaneously meeting the space requirement and the power requirement of each server to be mounted are determined as recommended rack groups from the third-level rack group set; A rack allocation scheme meeting the space requirement and the power requirement of each server to be mounted is determined for each recommended rack group; An allocation benefit after the servers to be mounted are allocated according to the rack allocation scheme is determined, and the allocation benefit comprises a space benefit obtained according to the available space capacity, a power benefit obtained according to the available power capacity, a refrigeration benefit obtained according to the available refrigeration capacity, and a network benefit obtained according to the available network capacity; A mounting rack recommendation scheme for the servers to be mounted is determined according to the allocation benefit of each rack allocation scheme.

2. The method of claim 1, wherein, The distribution of the proceeds ; Wherein: representing spatial gains, representing weights; ； K represents the number of racks in the current rack group to be calculated, and k represents the rack serial number; a utilization of the available space capacity representing the cabinet k, a ratio of the pre-allocated space capacity to the available space capacity for the cabinet k; representing the average available space capacity utilization, ; representing the electric power benefit, representing the weight of the electric power benefit; ； a utilization of the available power capacity representing the cabinet, a ratio of the pre-allocated power capacity to the available power capacity for the cabinet k; representing a refrigeration benefit, representing a weight of the refrigeration benefit; ； a ratio of the pre-allocated refrigeration capacity to the available refrigeration capacity for the refrigeration area in which the current cabinet group is located; representing the network revenue, representing the weight of the network. ； The ratio of the pre-allocated network capacity for the current cabinet group to the available network capacity. 3.The method of claim 1, wherein, The process of determining the allocation benefit after the servers to be mounted are allocated according to the rack allocation scheme and determining the mounting rack recommendation scheme for the servers to be mounted according to the allocation benefit of each rack allocation scheme comprises: Step A: selecting any one server to be mounted that is not marked for allocation, and determining a rack meeting the space requirement and the power requirement of the current server as a branch rack in all recommended rack groups, and taking the mapping relationship between the current server and each branch rack as a rack allocation scheme; Step B: retaining the rack allocation scheme meeting the continuous recommendation condition; Step C: marking the current server for allocation, adding a node to each retained rack allocation scheme, and then going to Step B; Step D: after all servers to be mounted are marked for allocation, retaining the rack allocation scheme as the mounting rack recommendation scheme for the servers to be mounted.

4. The method for recommending server rack positions as described in claim 3, characterized in that, The continuous recommendation condition comprises: The rack group to which the rack allocation scheme belongs meets the feasibility condition, the rack capacity requirement and the rack power requirement at the same time when the rack allocation scheme is adopted; And: The upper bound of the benefit of the rack allocation scheme is above the lower bound of the benefit when the rack allocation scheme is adopted; The upper bound of the benefit is obtained according to the allocation benefit of the rack allocation scheme and the maximum allocation benefit of the remaining servers to be mounted when the rack allocation scheme is adopted; The lower bound of the benefit is obtained according to the allocation benefit of all remaining servers to be mounted when the rack allocation scheme is adopted, which are allocated to the racks meeting the space requirement and the power requirement at the same time.

5. The method for recommending server rack positions as described in claim 4, characterized in that, The revenue upper bound ; ； The determining process of the lower bound of the revenue comprises: determining a subsequent branch scheme of each rack allocation scheme, the subsequent branch scheme being a scheme of allocating all the remaining to-be-shelved servers to the cabinets that meet the space requirement and the power requirement simultaneously when the rack allocation scheme is adopted; forming each rack allocation scheme and the corresponding subsequent branch scheme of each rack allocation scheme into a complete branch scheme; determining the allocation revenue of each complete branch scheme, and taking the maximum allocation revenue as the lower bound of the revenue.

6. The method for recommending server rack positions as described in claim 3, characterized in that, The process of adding nodes comprises: selecting an arbitrary one of the to-be-shelved servers that are not marked for allocation, and determining, in the cabinet group to which the rack allocation scheme belongs, a cabinet that meets the space requirement and the power requirement of the current server, and taking the mapping relationship between the cabinet and the current server as a node of the current rack allocation scheme.

7. The method of recommending stands for a server according to any one of claims 1 to 6, wherein, The feasibility condition comprises: the available network capacity of the cabinet group is above the pre-allocation network capacity, the pre-allocation network capacity being the sum of the network capacities consumed by all the to-be-shelved servers when the to-be-shelved servers are deployed to the cabinet group; and: the available refrigeration capacity of the refrigeration area where the cabinet group is located is above the pre-allocation refrigeration capacity, the pre-allocation refrigeration capacity being the sum of the refrigeration capacities consumed by all the to-be-shelved servers when the to-be-shelved servers are deployed to the cabinet group. The cabinet capacity requirement comprises: the available space capacity of each cabinet in the cabinet group is above the pre-allocation space capacity, the pre-allocation space capacity being the sum of the cabinet space capacities consumed by all the to-be-shelved servers when the to-be-shelved servers are deployed to the cabinet group. The cabinet power requirement comprises: the available power capacity of each cabinet in the cabinet group is above the pre-allocation power capacity, the pre-allocation power capacity being the sum of the power capacities consumed by all the to-be-shelved servers when the to-be-shelved servers are deployed to the cabinet group.

8. The method of recommending on-board server locations according to any one of claims 1 to 6, wherein, The determination condition that the cabinet can simultaneously accommodate the maximum consumed cabinet space capacity and the maximum consumed power capacity comprises: there is a cabinet in the cabinet group whose maximum available power capacity is above the maximum consumed power capacity; and: there is a cabinet in the cabinet group whose maximum available space capacity is above the maximum consumed cabinet space capacity. The determination condition that all the cabinet groups can simultaneously meet the space requirement and the power requirement of each to-be-shelved server comprises: after all the to-be-shelved servers are allocated to the same cabinet group, the available space capacity and the available power capacity of each cabinet in the cabinet group are both positive numbers. 9.A server-based recommendation management device for a shelf position, characterized by comprising: The server-shelving rack recommendation management device comprises a processor, a memory, and a server-shelving rack recommendation program stored in the memory and executable by the processor, wherein the server-shelving rack recommendation program, when executed by the processor, implements the steps of the server-shelving rack recommendation method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a server-shelving rack recommendation program, wherein the server-shelving rack recommendation program, when executed, implements the steps of the server-shelving rack recommendation method according to any one of claims 1 to 8.

Images