BIM (Building Information Modeling) software automatic full-life-cycle management method for virtualized resource pool
Through the automated management method of virtualized resource pools, dynamic correlation adaptation and time-varying feedback-driven algorithms are used to solve the problems of low resource utilization and slow scheduling response in the BIM system, efficient allocation and recycling of resources are achieved, and project improvement overall efficiency and response speed.
Patent Information
- Application Number
- CN202411564480.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-05
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-11-05
AI Technical Summary
The existing BIM management system has low resource utilization, slow scheduling response, lacks flexibility, and cannot dynamically adjust resource allocation, resulting in resource waste and task delays, affecting project progress and efficiency.
Adopt the automated full life cycle management method of virtualized resource pools, and through dynamic correlation adaptive resource coordination algorithm and time-varying feedback-driven computing resource hierarchical multiplexing algorithm, we can realize efficient application, scheduling, allocation and recycling of resources to ensure resources High utilization rate and fast system response speed.
It significantly improves system resource utilization efficiency, reduces task delay and resource waste, improves project execution fluency and response speed, and is suitable for complex multi-task parallel BIM environments.
Smart Images

Figure CN119065859A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of virtualization technology, and in particular to a BIM software automated full life cycle management method for a virtualized resource pool. Background Art
[0002] In the current building information model (BIM) management process, with the widespread application of BIM technology and the continuous expansion of project scale, how to efficiently manage and schedule resources has become an important challenge. BIM software is used in the full life cycle management of building design, construction and maintenance. With the increasing demand for multi-task parallel processing, the shortcomings of existing technologies in resource management are becoming more and more apparent. Most of the current BIM resource management technologies rely on static configuration, and resource scheduling is relatively fixed. This method can be used in small-scale projects or when resources are sufficient, but in complex, multi-task parallel construction projects, it exposes serious problems such as low resource utilization efficiency and slow scheduling response.
[0003] In the existing BIM management system, resource allocation and scheduling mostly rely on pre-set algorithms or human intervention, which lacks flexibility. Especially when resource demand fluctuates greatly, the system is difficult to respond quickly. For example, when the resource demand for certain tasks decreases, the system cannot release the occupied resources in time, resulting in resource waste. At the same time, in the case of tight resources, the system cannot effectively allocate temporarily idle resources to other tasks through dynamic scheduling. This results in not only long-term idle resources in the resource pool, but also some tasks being delayed or interrupted due to insufficient resources, which seriously affects the overall progress and efficiency of the project.
[0004] More importantly, the resource scheduling method of existing BIM management technology is too rigid and cannot be adjusted in real time according to the changes in demand during task execution. Even if a certain dynamic adjustment mechanism is introduced into the scheduling algorithm, its adaptive ability is still limited. As the scale and complexity of the project increase, the resource requirements between tasks vary significantly, and the resource occupancy of different tasks also changes over time, which makes it difficult for static resource allocation strategies to cope with complex scenarios in actual operations. Low resource utilization and delayed scheduling ultimately lead to a decline in project execution efficiency.
[0005] At the same time, when processing concurrent tasks, the resource competition problem in the existing system is particularly prominent. When multiple tasks request the same resource at the same time, it is difficult for the system to coordinate effectively, resulting in intensified resource competition. This not only causes some tasks to fail to execute smoothly due to insufficient resource support, but also further prolongs the overall completion time of the project. In addition, in some cases, although the demand for resources of tasks has changed, due to the lack of a dynamic adjustment mechanism for task priorities in the system, some low-priority tasks may occupy a large amount of resources, while high-priority tasks are stagnant due to insufficient resources.
[0006] In addition to the problems of low resource utilization and insufficient scheduling efficiency, the existing BIM management system also has defects in the resource reuse mechanism. When certain tasks are completed, the system is often unable to recycle the released resources in time and allocate them to new tasks. Especially in the case of tight resources, this lag in resource recovery and reuse will result in the resources in the resource pool not being able to be used efficiently to the maximum extent. The current scheduling mechanism lacks intelligent resource control means, and is unable to effectively recycle and redistribute resources according to real-time needs, which seriously affects the overall resource management efficiency of the system. Therefore, how to provide an automated full life cycle management method for BIM software for virtualized resource pools is a problem that technicians in this field urgently need to solve. Summary of the invention
[0007] One purpose of the present invention is to propose a BIM software automated full life cycle management method for a virtualized resource pool, so as to realize efficient application, scheduling, allocation and recovery of BIM resources. The method makes full use of virtualized resource scheduling technology, optimizes resource utilization through intelligent algorithms, ensures efficient resource allocation, adapts to complex multi-task parallel processing environments, and has the advantages of high flexibility, high resource utilization and fast system response speed.
[0008] The BIM software automated full life cycle management method for a virtualized resource pool according to an embodiment of the present invention comprises the following steps: S1. Through the user self-service page, the user initiates an application for BIM resources through the WEB interface; S2. After receiving the BIM environment resource application request, the desktop cloud resource scheduler checks whether the physical resources in the cloud desktop resource pool meet the demand; S3. When the resources in the resource pool meet the demand, a virtual machine instance is automatically created based on the computing power demand of the BIM software application, and the BIM software is deployed on the virtual machine instance to complete resource delivery; S4. When the resources in the resource pool cannot meet the demand, a dynamic associated adaptive resource coordination algorithm is used to identify and adjust the resource allocation between different tasks and coordinate resources between tasks; S5. When resource coordination still cannot meet the demand, the BIM resource application is added to the task queue, and the computing resource layered reuse algorithm driven by time-varying feedback is adopted to dynamically adjust the computing resource allocation based on the feedback information during the task execution process; S6. Monitor the usage status of virtual machine instances, recycle unused BIM environment resources, and reallocate the reclaimed resources to new task requests; S7. Through the strategy of time-sharing reuse of computing resources, integrate the computing resources in the resource pool, allocate resources and perform scheduling.
[0009] Optionally, S3 includes the following steps: S31. When the desktop cloud resource scheduler confirms that the physical resources in the resource pool meet the user's application requirements, according to the BIM software application type Determine the computing power parameters required for the virtual machine instance, including CPU requirements Memory requirements Storage requirements and network bandwidth requirements ; S32: The desktop cloud resource scheduler calls the resource pool in the virtualization platform according to the computing power demand parameters to create a virtual machine instance ,in Indicates virtual machine instances; S33, virtual machine instance After successful creation, the required BIM software Deploy to the instance. The deployment process includes software installation, configuration, and import of related dependency files. S34. After the BIM software is deployed, the resource information of the virtual machine instance, the BIM software application information and the allocation status are recorded in the database. The record includes The unique identifier of the task, the number of resources assigned, and the task number ; S35, generate feedback information, and send the virtual machine instance Deliver relevant information to users; S36. During the deployment of the virtual machine instance, dynamically monitor the virtualized resource pool; S37. If resource conflicts or errors occur during the creation of virtual machine instances or the deployment of BIM software, record error logs and try to reschedule resources for repair or redeployment.
[0010] Optionally, S4 includes the following steps: S41. When the physical resources in the resource pool cannot meet the demand, the desktop cloud resource scheduler calls the dynamic association adaptive resource coordination algorithm; for each task Build time dependent resource requirement vector: ;in, For the task In time The number of CPU cores required; For the task In time The amount of memory required; For the task In time The amount of storage required; For the task In time Required network bandwidth; S42. Introducing the dynamic correlation matrix , combined with the changing trend of historical task requirements; used to measure the task and tasks The resource demand correlation between: ; The first measurement task and tasks The similarity of resource requirements at each moment captures the current resource requirements; the second measurement task and tasks The resource demand change trend is captured through the time derivative to capture the dynamic changes of history; Used to control the impact of historical demand change trends on relevance; S43. Calculate the dynamic priority coefficient for each task based on the dynamic relevance of the task , used to determine the priority of tasks in resource allocation; the calculation formula of the priority coefficient is: ; in, For the task In time Priority; Task and tasks Dynamic correlation of S44, introduce task adaptive adjustment coefficient , which is used to make real-time adjustments based on the actual resource usage and expected demand of the task. The adjustment formula is: ; in, For the task Adaptive adjustment coefficient of To adjust the difference between actual use and expected use; For the task In time The actual resource usage of For the task the expected resource requirements; To control the correlation between tasks Impact on adaptive regulation; S45. According to the priority coefficient of the task and adaptive adjustment coefficient , dynamically calculate the resource allocation for each task ; S46. To optimize resource allocation, a spatiotemporal feedback control mechanism is introduced to adjust the resource allocation of tasks through real-time feedback; S47. Based on the execution of the tasks, the spatiotemporal feedback control mechanism is used to continuously optimize resource allocation.
[0011] Optionally, S5 includes the following steps: S51. When resource coordination still cannot meet the demand, BIM resource application is added to the task queue; a hierarchical computing model is constructed according to the different resource requirements of the task; each task The computing power demand vector is decomposed into ;in, For the task In time CPU computing power requirements; For the task In time Memory computing power requirements; For the task In time Storage and computing power requirements; For the task In time The network computing power required; For the task In time The energy demand, which represents the energy consumed by the task; S52. Introduce a high-dimensional time-varying feedback driven model to dynamically adjust the hierarchical computing power requirements of tasks based on real-time feedback information; define the time-varying feedback coefficient matrix , which is used to regulate the dynamic changes of hierarchical requirements of tasks over time, is derived as follows: ; in, is the identity matrix, keeping the original demand unchanged; is the feedback adjustment coefficient, which controls the intensity of the feedback impact; is the feedback frequency, which indicates the periodicity of feedback fluctuations; is the phase offset, indicating the task Feedback response delay; is the nonlinear feedback attenuation coefficient, which is used to control the impact of resource demand volatility on feedback; S53, according to the time-varying feedback coefficient matrix , dynamically adjust the hierarchical computing power requirements of tasks; S54, further optimize the resource allocation of tasks through the dynamic collaborative hierarchical model; collaborative hierarchical matrix Indicates the task and tasks The synergy effect in computing resource stratification is derived as follows: ; in, is the eigenvalue of the synergy matrix, indicating that The strength of collaboration on the layer; the first item represents the task and tasks The synergistic effect of computing power requirements changing over time; the second term controls the differences in resource requirements between different tasks through nonlinear decay; To control the impact of differences in resource requirements on synergy; S55. Adjust requirements based on task feedback and the collaborative hierarchical matrix , calculate the final resource allocation , the allocation formula is: ; in, For the task The final resource allocation obtained; It is a collaborative hierarchical matrix, which is used to adjust the collaboration and resource requirements between tasks; S56. To further optimize resource utilization, a nonlinear spatiotemporal feedback control mechanism is introduced to dynamically regulate resource allocation by combining temporal and spatial feedback. The derived formula is as follows: ; in, is the spatiotemporal feedback adjustment coefficient, which controls the impact of spatiotemporal feedback on resource allocation; is the spatial control function, which reflects the resource requirements of the task in the spatial location; is the spatial position attenuation coefficient, which is used to control the resource distribution of tasks in different spatial positions; is the initial spatial position of the task; S57. Through the spatiotemporal feedback control mechanism, the resource allocation of tasks is dynamically monitored and adjusted, so that tasks can achieve efficient resource hierarchical reuse and regulation when the resource pool is limited.
[0012] Optionally, the S6 comprises the following steps: S61. Monitor the usage status of virtual machine instances and collect resource usage data of each virtual machine instance, defined as , which includes CPU usage , Memory usage , Storage usage and network bandwidth usage ; S62. By evaluating the resource usage data of the virtual machine instance , computing resource utilization , and analyze whether the resources in the virtual machine instance are fully utilized; S63, when the resource utilization of the virtual machine instance Below the preset threshold for a long time When the resource of the virtual machine instance enters the recycling state, it is determined whether the resource of the virtual machine instance enters the recycling state; S64: For virtual machine instances that meet the recycling conditions, unused resources are recycled from the instances through a resource recycling mechanism to generate a recycled resource vector , indicating a virtual machine instance The amount of recycled resources in S65. Recycled resources The resource pool is re-added and resources are dynamically allocated based on new task requirements. The allocation is based on the task priority and the resource requirements of the current task, and the amount of resources reallocated to the new task. ; Allocate based on available resources in the resource pool; S66. After the resources are reallocated, the resource usage of the new task is monitored in real time, and resource usage data of the new task instance is collected and analyzed; S67. Further optimize and dynamically adjust resource allocation between tasks according to the resource usage status of the new task.
[0013] Optionally, the S7 includes the following steps: S71, real-time collection task Resource requirements, building a task-resource bidirectional response matrix , used to capture tasks About Resources The time-varying effect of is defined as follows: ; in, For the task In time The amount of resources required; For the task In time The rate of change of resource demand; For resources The time-varying response function of To the task the responsiveness of demand, based on the current availability of resources; Indicates the change from the initial time to the current time; S72. Resources The time-varying response function It represents the dynamic response of resources to changes in task requirements, taking into account resource availability and load fluctuations. The derived formula is: ; in, For resources In time The actual amount of resources available; For resources In time The maximum available capacity; For resources The minimum resource limit indicates the minimum resource usage limit; To regulate the decay coefficient, control the decay speed of resource response changes. The larger the coefficient, the more sensitive the resource is to changes. S73, based on the two-way response of tasks and resources, using nonlinear resource control function Dynamically adjust resource allocation between tasks; resource allocation priority Defined as: ; in, It is a two-way response matrix of tasks and resources, reflecting the task About Resources The impact of For current time Total available resources in the resource pool; For the task Resource requirements at the current time; A small constant used to avoid zero denominator; S74. In the process of resource allocation, the actual usage of tasks is dynamically adjusted through the feedback control mechanism; feedback control function Adjust based on actual usage of tasks and expected demand; S75. Final resource allocation for the task Integrated feedback control function and nonlinear resource regulation functions , defined as: ; S76. Continuously adjust resource allocation priorities based on real-time feedback so that tasks can obtain sufficient resource support in a timely manner when resource requirements change; real-time adjustments are determined by task feedback information and the availability of resource pools; S77. Analyze the execution feedback of the task and further optimize the time-sharing reuse strategy of computing resources.
[0014] The beneficial effects of the present invention are: The present invention realizes dynamic adjustment of resources between tasks in the case of insufficient resources by introducing a dynamic association adaptive resource coordination algorithm. By intelligently analyzing the resource demand correlation between tasks, the system can effectively coordinate resource allocation between different tasks, avoiding the task delay and resource waste problems caused by insufficient resources in traditional methods, and significantly improving the utilization efficiency of system resources and the fluency of task execution.
[0015] The present invention adopts a time-varying feedback-driven computing resource hierarchical reuse algorithm, and dynamically adjusts the resource allocation strategy according to feedback information during task execution. The system can capture the changes in resource requirements during task execution in real time, and optimize resource allocation through feedback mechanism to ensure that resources in the resource pool can be reused flexibly and efficiently.
[0016] This mechanism not only improves the system's resource management capabilities, but also effectively reduces delays and efficiency losses caused by resource mismatches during task execution. It is particularly suitable for complex multi-task parallel BIM environments.
[0017] The present invention designs a resource monitoring and recycling mechanism. By real-time monitoring of the usage status of virtual machine instances, it can timely recycle the resources of completed tasks or low-utilization tasks and reallocate these resources to new task requirements. Unlike the problem of traditional systems that cannot recycle and allocate resources in a timely manner, this mechanism ensures that the resources in the resource pool are always efficiently utilized, reduces the idle time of resources, and further improves the system's response speed and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 The overall flow chart of the BIM software automated full life cycle management method for virtualized resource pool proposed by the present invention; Figure 2 It is a schematic diagram of using a dynamic associated adaptive resource coordination algorithm to adjust resources between tasks when resources are insufficient in the present invention; Figure 3This is a schematic diagram of the present invention using a time-varying feedback driven computing resource hierarchical reuse algorithm to perform resource allocation and reuse. DETAILED DESCRIPTION
[0019] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.
[0020] refer to Figure 1-3 , a BIM software automated full life cycle management method for a virtualized resource pool includes the following steps: S1. Through the user self-service page, the user initiates an application for BIM resources through the WEB interface; S2. After receiving the BIM environment resource application request, the desktop cloud resource scheduler checks whether the physical resources in the cloud desktop resource pool meet the demand; S3. When the resources in the resource pool meet the demand, a virtual machine instance is automatically created based on the computing power demand of the BIM software application, and the BIM software is deployed on the virtual machine instance to complete resource delivery; S4. When the resources in the resource pool cannot meet the demand, a dynamic associated adaptive resource coordination algorithm is used to identify and adjust the resource allocation between different tasks and coordinate resources between tasks; S5. When resource coordination still cannot meet the demand, the BIM resource application is added to the task queue, and the computing resource layered reuse algorithm driven by time-varying feedback is adopted to dynamically adjust the computing resource allocation based on the feedback information during the task execution process; S6. Monitor the usage status of virtual machine instances, recycle unused BIM environment resources, and reallocate the reclaimed resources to new task requests; S7. Through the strategy of time-sharing reuse of computing resources, integrate the computing resources in the resource pool, allocate resources and perform scheduling.
[0021] In this implementation, S3 includes the following steps: S31. When the desktop cloud resource scheduler confirms that the physical resources in the resource pool meet the user's application requirements, according to the BIM software application type Determine the computing power parameters required for the virtual machine instance, including CPU requirements Memory requirements Storage requirements and network bandwidth requirements ; S32: The desktop cloud resource scheduler calls the resource pool in the virtualization platform according to the computing power demand parameters to create a virtual machine instance ,in Indicates virtual machine instances; S33, virtual machine instance After successful creation, the required BIM software Deploy to the instance. The deployment process includes software installation, configuration, and import of related dependency files. S34. After the BIM software is deployed, the resource information of the virtual machine instance, the BIM software application information and the allocation status are recorded in the database. The record includes The unique identifier of the task, the number of resources assigned, and the task number ; S35, generate feedback information, and send the virtual machine instance Deliver relevant information to users; S36. During the deployment of the virtual machine instance, dynamically monitor the virtualized resource pool; S37. If resource conflicts or errors occur during the creation of virtual machine instances or the deployment of BIM software, record error logs and try to reschedule resources for repair or redeployment.
[0022] In this implementation, S4 includes the following steps: S41. When the physical resources in the resource pool cannot meet the demand, the desktop cloud resource scheduler calls the dynamic association adaptive resource coordination algorithm; for each task Build time dependent resource requirement vector: ;in, For the task In time The number of CPU cores required; For the task In time The amount of memory required; For the task In time The amount of storage required; For the task In time Required network bandwidth; S42. Introducing the dynamic correlation matrix , combined with the changing trend of historical task requirements; used to measure the task and tasks The resource demand correlation between: ; The first measurement task and tasks The similarity of resource requirements at each moment captures the current resource requirements; the second measurement task and tasks The resource demand change trend is captured through the time derivative to capture the dynamic changes of history; Used to control the impact of historical demand change trends on relevance; S43. Calculate the dynamic priority coefficient for each task based on the dynamic relevance of the task , used to determine the priority of tasks in resource allocation; the calculation formula of the priority coefficient is: ; in, For the task In time Priority; Task and tasks Dynamic correlation of S44, introduce task adaptive adjustment coefficient , which is used to make real-time adjustments based on the actual resource usage and expected demand of the task. The adjustment formula is: ; in, For the task Adaptive adjustment coefficient of To adjust the difference between actual use and expected use; For the task In time The actual resource usage of For the task the expected resource requirements; To control the correlation between tasks Impact on adaptive regulation; S45. According to the priority coefficient of the task and adaptive adjustment coefficient , dynamically calculate the resource allocation for each task ; S46. To optimize resource allocation, a spatiotemporal feedback control mechanism is introduced to adjust the resource allocation of tasks through real-time feedback; S47. Based on the execution of the tasks, the spatiotemporal feedback control mechanism is used to continuously optimize resource allocation.
[0023] In this implementation, S5 includes the following steps: S51. When resource coordination still cannot meet the demand, BIM resource application is added to the task queue; a hierarchical computing model is constructed according to the different resource requirements of the task; each task The computing power demand vector is decomposed into ;in, For the task In time CPU computing power requirements; For the task In time Memory computing power requirements; For the task In time Storage and computing power requirements; For the task In time The network computing power required; For the task In time The energy demand, which represents the energy consumed by the task; S52. Introduce a high-dimensional time-varying feedback driven model to dynamically adjust the hierarchical computing power requirements of tasks based on real-time feedback information; define the time-varying feedback coefficient matrix , which is used to regulate the dynamic changes of hierarchical requirements of tasks over time, is derived as follows: ; in, is the identity matrix, keeping the original demand unchanged; is the feedback adjustment coefficient, which controls the intensity of the feedback impact; is the feedback frequency, which indicates the periodicity of feedback fluctuations; is the phase offset, indicating the task Feedback response delay; is the nonlinear feedback attenuation coefficient, which is used to control the impact of resource demand volatility on feedback; S53, according to the time-varying feedback coefficient matrix , dynamically adjust the hierarchical computing power requirements of tasks; S54, further optimize the resource allocation of tasks through the dynamic collaborative hierarchical model; collaborative hierarchical matrix Indicates the task and tasks The synergy effect in computing resource stratification is derived as follows: ; in, is the eigenvalue of the synergy matrix, indicating that The strength of collaboration on the layer; the first item represents the task and tasks The synergistic effect of computing power requirements changing over time; the second term controls the differences in resource requirements between different tasks through nonlinear decay; To control the impact of differences in resource requirements on synergy; S55. Adjust requirements based on task feedback and the collaborative hierarchical matrix , calculate the final resource allocation , the allocation formula is: ; in, For the task The final resource allocation obtained; It is a collaborative hierarchical matrix, which is used to adjust the collaboration and resource requirements between tasks; S56. To further optimize resource utilization, a nonlinear spatiotemporal feedback control mechanism is introduced to dynamically regulate resource allocation by combining temporal and spatial feedback. The derived formula is as follows: ; in, is the spatiotemporal feedback adjustment coefficient, which controls the impact of spatiotemporal feedback on resource allocation; is the spatial control function, which reflects the resource requirements of the task in the spatial location; is the spatial position attenuation coefficient, which is used to control the resource distribution of tasks in different spatial positions; is the initial spatial position of the task; S57. Through the spatiotemporal feedback control mechanism, the resource allocation of tasks is dynamically monitored and adjusted, so that tasks can achieve efficient resource hierarchical reuse and regulation when the resource pool is limited.
[0024] In this implementation, S6 includes the following steps: S61. Monitor the usage status of virtual machine instances and collect resource usage data of each virtual machine instance, defined as , which includes CPU usage , Memory usage , Storage usage and network bandwidth usage ; S62. By evaluating the resource usage data of the virtual machine instance , computing resource utilization , and analyze whether the resources in the virtual machine instance are fully utilized; S63. When the resource utilization of the virtual machine instance Below the preset threshold for a long time When the resource of the virtual machine instance enters the recycling state, it is determined whether the resource of the virtual machine instance enters the recycling state; S64: For virtual machine instances that meet the recycling conditions, unused resources are recycled from the instances through a resource recycling mechanism to generate a recycled resource vector , indicating a virtual machine instance The amount of recycled resources in S65. Recycled resources The resource pool is re-added and resources are dynamically allocated based on new task requirements. The allocation is based on the task priority and the resource requirements of the current task, and the amount of resources reallocated to the new task. ; Allocate based on available resources in the resource pool; S66. After the resources are reallocated, the resource usage of the new task is monitored in real time, and resource usage data of the new task instance is collected and analyzed; S67. Further optimize and dynamically adjust resource allocation between tasks according to the resource usage status of the new task.
[0025] In this implementation, S7 includes the following steps: S71, real-time collection task Resource requirements, building a task-resource bidirectional response matrix , used to capture tasks About Resources The time-varying effect of is defined as follows: ; in, For the task In time The amount of resources required; For the task In time The rate of change of resource demand; For resources The time-varying response function of To the task the responsiveness of demand, based on the current availability of resources; Indicates the change from the initial time to the current time; S72. Resources The time-varying response function It represents the dynamic response of resources to changes in task requirements, taking into account resource availability and load fluctuations. The derived formula is: ; in, For resources In time The actual amount of resources available; For resources In time The maximum available capacity; For resources Minimum resource limit; To regulate the decay coefficient, control the decay speed of resource response changes. The larger the coefficient, the more sensitive the resource is to changes. S73, based on the two-way response of tasks and resources, using nonlinear resource control function Dynamically adjust resource allocation between tasks; resource allocation priority Defined as: ; in, It is a two-way response matrix of tasks and resources, reflecting the task About Resources The impact of For current time Total available resources in the resource pool; For the task Resource requirements at the current time; A small constant used to avoid zero denominator; S74. In the process of resource allocation, the actual usage of tasks is dynamically adjusted through the feedback control mechanism; feedback control function Adjust based on actual usage of tasks and expected demand; S75. Final resource allocation for the task Integrated feedback control function and nonlinear resource regulation functions , defined as: ; S76. Continuously adjust resource allocation priorities based on real-time feedback so that tasks can obtain sufficient resource support in a timely manner when resource requirements change; real-time adjustments are determined by task feedback information and the availability of resource pools; S77. Analyze the execution feedback of the task and further optimize the time-sharing reuse strategy of computing resources.
[0026] Embodiment 1: In modern transportation planning projects, with the rapid development of cities and the increasing complexity of transportation needs, transportation planning and management systems need to process a large amount of traffic data and perform multi-task parallel planning and scheduling. For example, multiple tasks such as urban road design, traffic flow prediction, and public transportation route optimization usually need to be carried out simultaneously, and the demand for computing resources for each task also changes over time. Due to the static nature of resource scheduling, traditional transportation planning systems are often unable to flexibly adjust resource allocation according to the real-time needs of tasks, resulting in problems such as low resource utilization and task delays.
[0027] In this embodiment, the present invention is applied to a traffic planning project in a large city. The core task of the project is to carry out multiple traffic planning sub-projects at the same time, including multiple concurrent tasks such as new road design, traffic flow prediction, and public transportation network optimization. These tasks require the use of BIM software in a virtualized environment for data analysis and model simulation, requiring the system to have highly flexible resource scheduling capabilities.
[0028] In this scenario, traditional traffic planning systems have difficulty coping with complex and changing resource demands. Taking new road design and traffic flow prediction as an example, the new road design task has a greater demand for computing power in the initial stage, while the traffic flow prediction task has a smaller demand for computing power due to the data collection and preprocessing stages. However, as the new road design task is gradually completed, its computing power demand gradually decreases, while the traffic flow prediction task has a surge in resource demand when it enters the model simulation and data analysis stage. If the traditional static resource scheduling method is relied upon, it will be difficult for the system to flexibly adjust resources according to task changes, which may cause some tasks to be delayed due to insufficient resources, while resource-occupying tasks will be idle.
[0029] To solve this problem, the virtualized resource pool management method proposed in the present invention automatically schedules resources and optimizes resource allocation between tasks when resources are tight by dynamically associating an adaptive resource coordination algorithm and a time-varying feedback-driven computing power resource hierarchical reuse algorithm. At the beginning of the project, the system first initiates a resource application for the traffic planning task through the user self-service page, and the scheduler checks the current resource pool. If the resources in the resource pool meet the demand, the system will automatically create virtual machine instances according to the computing power demand and assign them to each task to ensure the parallel execution of the tasks. When the new road design task reaches the mid-term stage, the resource demand decreases, and the system automatically adjusts the resources dynamically through the time-varying feedback-driven algorithm, and allocates some idle resources to the traffic prediction task, avoiding waste of resources. At the same time, the computing power demand of the traffic prediction task gradually increases with the increase of data simulation workload, and the system monitors feedback in real time to ensure that the resources required for the task can be allocated in place in the shortest time.
[0030] The data proves the superiority of the present invention in this traffic planning project. In the first month of the project, the system recorded the resource allocation of different tasks. The resource allocation of the new road design task from the 1st to the 5th day accounted for as much as 70%, while the allocated resources for the traffic flow forecasting task were only 30%. As the resource requirements of the design tasks gradually decreased, the system dynamically allocated these idle resources to the traffic flow forecasting task. By the 15th day, the resource allocation ratio of the traffic flow forecasting task increased to 65%, while that of the new road design task decreased to 35%. This flexible scheduling of resources greatly improves the overall resource utilization of the system.
[0031] Especially at the key nodes of the project, the traffic planning simulation task requires a lot of resource support. The system ensures that the traffic flow prediction task obtains 90% of the available resources through the adaptive resource allocation mechanism, while other low-priority tasks are adjusted through the dynamic association adaptive resource coordination algorithm. The resource utilization rate of the project reached 95% on the 30th day, which is 25% higher than the average utilization rate of the traditional static scheduling system.
[0032] In addition, the execution time of the project has been significantly shortened. Through data analysis, under the traditional scheduling method, the project is expected to take 60 days to complete, but after adopting the resource scheduling method of the present invention, the actual execution time is shortened to 52 days. The system not only improves resource utilization, but also effectively shortens the total time of the project, saving a lot of costs and human resources.
[0033] Table 1 Data table
[0034] The data table shows the resource scheduling effect of the present invention in practical applications, proving the outstanding performance of the present invention in optimizing resource utilization, shortening scheduling response time, and improving task execution efficiency.
[0035] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A BIM software automated full life cycle management method for a virtualized resource pool, characterized in that: The steps include: S1. Through the user self-service page, the user initiates an application for BIM resources through the WEB interface; S2. After receiving the BIM environment resource application request, the desktop cloud resource scheduler checks whether the physical resources in the cloud desktop resource pool meet the demand; S3. When the resources in the resource pool meet the demand, a virtual machine instance is automatically created based on the computing power demand of the BIM software application, and the BIM software is deployed on the virtual machine instance to complete resource delivery; S4. When the resources in the resource pool cannot meet the demand, a dynamic associated adaptive resource coordination algorithm is used to identify and adjust the resource allocation between different tasks and coordinate resources between tasks; S5. When resource coordination still cannot meet the demand, the BIM resource application is added to the task queue, and the computing resource layered reuse algorithm driven by time-varying feedback is adopted to dynamically adjust the computing resource allocation based on the feedback information during the task execution process; S6. Monitor the usage status of virtual machine instances, recycle unused BIM environment resources, and reallocate the reclaimed resources to new task requests; S7. Through the strategy of time-sharing reuse of computing resources, integrate the computing resources in the resource pool, allocate resources and perform scheduling.
2. The BIM software automated full life cycle management method for a virtualized resource pool according to claim 1, characterized in that: The S3 comprises the following steps: S31. When the desktop cloud resource scheduler confirms that the physical resources in the resource pool meet the user's application requirements, according to the BIM software application type Determine the computing power parameters required for the virtual machine instance, including CPU requirements Memory requirements Storage requirements and network bandwidth requirements ; S32: The desktop cloud resource scheduler calls the resource pool in the virtualization platform according to the computing power demand parameters to create a virtual machine instance ,in Indicates virtual machine instances; S33, virtual machine instance After successful creation, the required BIM software Deploy to the instance. The deployment process includes software installation, configuration, and import of related dependency files. S34. After the BIM software is deployed, the resource information of the virtual machine instance, the BIM software application information and the allocation status are recorded in the database. The record includes The unique identifier of the task, the number of resources assigned, and the task number ; S35, generate feedback information, and send the virtual machine instance Deliver relevant information to users; S36. During the deployment of the virtual machine instance, dynamically monitor the virtualized resource pool; S37. If resource conflicts or errors occur during the creation of virtual machine instances or the deployment of BIM software, record error logs and try to reschedule resources for repair or redeployment.
3. The BIM software automated full life cycle management method for a virtualized resource pool according to claim 1, characterized in that: The S4 comprises the following steps: S41. When the physical resources in the resource pool cannot meet the demand, the desktop cloud resource scheduler calls the dynamic association adaptive resource coordination algorithm; for each task Build time dependent resource requirement vector: ;in, For the task In time The number of CPU cores required; For the task In time The amount of memory required; For the task In time The amount of storage required; For the task In time Required network bandwidth; S42. Introducing the dynamic correlation matrix , combined with the changing trend of historical task requirements; used to measure the task and tasks The resource demand correlation between: ; The first measurement task and tasks The similarity of resource requirements at each moment captures the current resource requirements; the second measurement task and tasks The resource demand change trend is captured through the time derivative to capture the dynamic changes of history; Used to control the impact of historical demand change trends on relevance; S43. Calculate the dynamic priority coefficient for each task based on the dynamic relevance of the task , used to determine the priority of tasks in resource allocation; the calculation formula of the priority coefficient is: ; in, For the task In time Priority; Task and tasks Dynamic correlation of S44, introduce task adaptive adjustment coefficient , which is used to make real-time adjustments based on the actual resource usage and expected demand of the task. The adjustment formula is: ; in, For the task Adaptive adjustment coefficient of To adjust the difference between actual use and expected use; For the task In time The actual resource usage of For the task the expected resource requirements; To control the correlation between tasks Impact on adaptive regulation; S45. According to the priority coefficient of the task and adaptive adjustment coefficient , dynamically calculate the resource allocation for each task ; S46. To optimize resource allocation, a spatiotemporal feedback control mechanism is introduced to adjust the resource allocation of tasks through real-time feedback; S47. Based on the execution of the tasks, the spatiotemporal feedback control mechanism is used to continuously optimize resource allocation.
4. The BIM software automated full life cycle management method for a virtualized resource pool according to claim 1, characterized in that: The S5 comprises the following steps: S51. When resource coordination still cannot meet the demand, BIM resource application is added to the task queue; a hierarchical computing model is constructed according to the different resource requirements of the task; each task The computing power demand vector is decomposed into ;in, For the task In time CPU computing power requirements; For the task In time Memory computing power requirements; For the task In time Storage and computing power requirements; For the task In time The network computing power requirements; For the task In time The energy demand, which represents the energy consumed by the task; S52. Introduce a high-dimensional time-varying feedback driven model to dynamically adjust the hierarchical computing power requirements of tasks based on real-time feedback information; define the time-varying feedback coefficient matrix , which is used to regulate the dynamic changes of hierarchical requirements of tasks over time, is derived as follows: ; in, is the identity matrix; is the feedback adjustment coefficient; is the feedback frequency; is the phase offset; is the nonlinear feedback attenuation coefficient; S53, according to the time-varying feedback coefficient matrix , dynamically adjust the hierarchical computing power requirements of tasks; S54, further optimize the resource allocation of tasks through the dynamic collaborative hierarchical model; collaborative hierarchical matrix Indicates the task and tasks The synergy effect in computing resource stratification is derived as follows: ; in, is the eigenvalue of the collaboration matrix; the first term represents the task and tasks The synergistic effect of computing power requirements changing over time; the second term controls the differences in resource requirements between different tasks through nonlinear decay; To control the impact of differences in resource requirements on synergy; S55. Adjust requirements based on task feedback and the collaborative hierarchical matrix , calculate the final resource allocation , the allocation formula is: ; in, For the task The final resource allocation obtained; is the collaborative layering matrix; S56. To further optimize resource utilization, a nonlinear spatiotemporal feedback control mechanism is introduced to dynamically regulate resource allocation by combining temporal and spatial feedback. The derived formula is as follows: ; in, is the spatiotemporal feedback adjustment coefficient; is the spatial regulation function; is the spatial position attenuation coefficient; is the initial spatial position of the task; S57. Through the spatiotemporal feedback control mechanism, the resource allocation of tasks is dynamically monitored and adjusted, so that tasks can achieve efficient resource hierarchical reuse and regulation when the resource pool is limited.
5. The BIM software automated full life cycle management method for a virtualized resource pool according to claim 1, characterized in that: The S6 comprises the following steps: S61. Monitor the usage status of virtual machine instances and collect resource usage data of each virtual machine instance, defined as , which includes CPU usage , Memory usage , Storage usage and network bandwidth usage ; S62. By evaluating the resource usage data of the virtual machine instance , computing resource utilization , and analyze whether the resources in the virtual machine instance are fully utilized; S63. When the resource utilization of the virtual machine instance Below the preset threshold for a long time When the resource of the virtual machine instance enters the recycling state, it is determined whether the resource of the virtual machine instance enters the recycling state; S64: For virtual machine instances that meet the recycling conditions, unused resources are recycled from the instances through a resource recycling mechanism to generate a recycled resource vector , indicating a virtual machine instance The amount of recycled resources in S65. Recycled resources The resource pool is re-added and resources are dynamically allocated based on new task requirements. The allocation is based on the task priority and the resource requirements of the current task, and the amount of resources reallocated to the new task. ; Allocate based on available resources in the resource pool; S66. After the resources are reallocated, the resource usage of the new task is monitored in real time, and resource usage data of the new task instance is collected and analyzed; S67. Further optimize and dynamically adjust resource allocation between tasks according to the resource usage status of the new task.
6. The BIM software automated full life cycle management method for virtualized resource pool according to claim 1, characterized in that: The S7 comprises the following steps: S71, real-time collection task Resource requirements, building a task-resource bidirectional response matrix , used to capture tasks About Resources The time-varying effect of is defined as follows: ; in, For the task In time The amount of resources required; For the task In time The rate of change of resource demand; For resources The time-varying response function of Indicates the change from the initial time to the current time; S72. Resources The time-varying response function It represents the dynamic response of resources to changes in task requirements, taking into account resource availability and load fluctuations. The derived formula is: ; in, For resources In time The actual amount of resources available; For resources In time The maximum available capacity; For resources Minimum resource limit; To adjust the attenuation coefficient; S73, based on the two-way response of tasks and resources, using nonlinear resource control function Dynamically adjust resource allocation between tasks; resource allocation priority Defined as: ; in, It is a two-way response matrix of tasks and resources; For current time Total available resources in the resource pool; For the task Resource requirements at the current time; A small constant to avoid zero denominator; S74. In the process of resource allocation, the actual usage of tasks is dynamically adjusted through the feedback control mechanism; feedback control function Adjust based on actual usage of tasks and expected demand; S75. Final resource allocation for the task Integrated feedback control function and nonlinear resource regulation functions , defined as: ; S76. Continuously adjust resource allocation priorities based on real-time feedback so that tasks can obtain sufficient resource support in a timely manner when resource requirements change; real-time adjustments are determined by task feedback information and the availability of resource pools; S77. Analyze the execution feedback of the task and further optimize the time-sharing reuse strategy of computing resources.
Citation Information
Patent Citations
Multi-level task scheduling method based on multi-objective optimization strategy
CN115858110A
Cloud resource automatic allocation system
CN118363765A
Policy-based hierarchical management of shared resources in a grid environment
US20060294238A1
Data harmonization across building lifecycle
US20210055716A1
Cited By
Computing power arrangement method and device, equipment and storage medium
CN119356828A
