CN118092817A - Intelligent management method and system for space of tablet personal computer - Google Patents
Intelligent management method and system for space of tablet personal computer Download PDFInfo
- Publication number
- CN118092817A CN118092817A CN202410504055.XA CN202410504055A CN118092817A CN 118092817 A CN118092817 A CN 118092817A CN 202410504055 A CN202410504055 A CN 202410504055A CN 118092817 A CN118092817 A CN 118092817A
- Authority
- CN
- China
- Prior art keywords
- memory
- application
- adjustment
- time
- resource
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims description 43
- 230000015654 memory Effects 0.000 claims abstract description 276
- 238000000034 method Methods 0.000 claims abstract description 28
- 230000000694 effects Effects 0.000 claims description 30
- 230000004044 response Effects 0.000 claims description 25
- 238000004364 calculation method Methods 0.000 claims description 22
- 238000012544 monitoring process Methods 0.000 claims description 18
- 230000006399 behavior Effects 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 230000007787 long-term memory Effects 0.000 claims description 6
- 230000035945 sensitivity Effects 0.000 claims description 6
- 230000006403 short-term memory Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 4
- 238000013480 data collection Methods 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 239000002699 waste material Substances 0.000 description 10
- 230000009286 beneficial effect Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 6
- 238000013468 resource allocation Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000008014 freezing Effects 0.000 description 2
- 238000007710 freezing Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000005303 weighing Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The invention provides a method and a system for intelligently managing space of a tablet personal computer. The method comprises the following steps: acquiring the use condition of the storage space of the tablet personal computer; evaluating the storage resource state of the tablet computer through a dynamic resource index model; predicting the memory and external memory requirements of the ith application of the tablet computer at a future time point; and adjusting the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at a future time point. By the method and the corresponding system, extra storage space can be allocated in advance, so that the increase of storage hardware loss caused by frequent internal and external memory call is avoided, and the use requirement of a user can be responded more quickly.
Description
Technical Field
The invention provides a method and a system for intelligently managing space of a tablet personal computer, and relates to the technical field of storage processing.
Background
With the popularity of tablet computers, users have placed higher demands on storage and memory management of devices. The existing management method often cannot fully utilize limited resources, so that the performance of equipment is reduced and the user experience is damaged. For example, a user uses an application frequently in a certain period of time, but because the memory is insufficient, application data is stored in the external memory, and when the user uses the application data, the data is called into the memory from the external memory, so that the loss of storage hardware is increased. Therefore, there is an urgent need for a more efficient, adaptive management method to optimize resource usage.
Disclosure of Invention
The invention provides a method and a system for intelligently managing space of a tablet personal computer, which are used for solving the problems mentioned above:
The invention provides a method for intelligently managing space of a tablet personal computer, which comprises the following steps:
Acquiring the use condition of the storage space of the tablet personal computer;
evaluating the storage resource state of the tablet computer through a dynamic resource index model;
Predicting the memory and external memory requirements of the ith application of the tablet computer at a future time point;
And adjusting the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at a future time point.
Further, the method for intelligently managing the space of the tablet personal computer, which is used for acquiring the use condition of the storage space of the tablet personal computer, comprises the following steps:
Acquiring the available memory capacity and the available memory capacity of the tablet personal computer at a time point t;
Acquiring the number of applications in the tablet personal computer, and the memory capacity occupied by each application at a time point t and the liveness index at the time point t;
And acquiring the data exchange amount from the memory to the external memory and the data exchange amount from the external memory to the memory of the tablet personal computer at time t.
Further, the method for intelligently managing the space of the tablet personal computer evaluates the storage resource state of the tablet personal computer through a dynamic resource index model comprises the following steps:
specifically, the dynamic resource index model is:
；
Wherein, Representing the resource competition coefficient between the ith and jth applications, N representing the total number of applications in the tablet computer,/>And/>Weight coefficients of resource contention and data exchange, respectively,/>Representing the amount of data exchange from memory to memory at time t,/>Representing the amount of data exchange from memory to memory at time t,/>Representing the available memory at time t,/>Representing the available memory at time t,/>Indicating the liveness index of the ith application at time point t,And/>Respectively representing the memory and the memory space occupied by the ith application at time t,/>Is a coefficient for adjusting the specific gravity of the application liveness and the resource demand,/>Is/>Is the maximum possible value of (a).
Further, the intelligent management method for the space of the tablet personal computer predicts the requirement of the ith application of the tablet personal computer on the memory and the external memory at a future time point, and comprises the following steps:
Collecting historical data for an ith application, including memory usage, external memory usage, application activity and user usage behavior;
Removing abnormal values in the data, and converting the data format into a time sequence format;
Selecting relevant features from the collected data, the relevant features including frequency of use, period of activity and amount of data exchange of the application;
Creating a new feature, and then creating a trend feature according to the increasing trend of the memory usage after creating a time feature according to the time period of application usage;
Training a long-term and short-term memory network model using the selected features and the historical data;
and predicting the memory and external memory requirements of the ith application at a specific future time point by using the trained model.
Further, the method for intelligently managing the space of the tablet personal computer adjusts the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at the future time point, and comprises the following steps:
When the DRI value calculation result is 0,0.3, not adjusting;
when the DRI value calculation result is [0.3,0.6], distributing storage space for the application according to the adjustment model;
specifically, the adjustment model is:
；
Wherein, Indicating the memory footprint of the ith application at the next time point t +1,Representing the current time point t moment, the memory usage amount of the ith application, alpha is an adjustment speed parameter, the speed and the sensitivity of controlling resource adjustment, sigma is a memory adjustment preference coefficient, and reflects the preference degree of the memory when the resource is adjusted,/>Representing the memory predicted usage of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present. 0 < alpha is less than or equal to 1, and alpha determines the response speed of the resource adjustment process to the difference between the predicted value and the current value. A smaller value of alpha means that the adjustment process is smoother and slower, while a larger value makes the adjustment faster and more sensitive, 0.ltoreq.σ.ltoreq.1, σ controls the balance of memory and external memory requirements, and when σ is closer to 1, the system is more biased to preferentially meet the memory requirements; when approaching 0, the memory is biased to store,/>The value range is as follows: /(I)Represents a higher/>, of importance of time and cost factors in adjusting resourcesThe value means that the system takes more into account response time and cost when making resource adjustments.
Meaning: the step length of the deep learning model in the optimization process determines the parameter updating speed. A smaller learning rate means that the learning process is more stable but may be slower, and a larger learning rate may accelerate the learning speed but risk unstable convergence.
；
Wherein,Representing the memory footprint of the ith application at the next point in time t +1,Representing the memory usage of the ith application at the current time point tRepresenting the predicted usage of the memory of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present.
And when the DRI value calculation result is larger than 0.6, allocating storage space for the application according to the adjustment model.
The invention provides an intelligent management system for a space of a tablet personal computer, which comprises the following components:
The monitoring module is used for acquiring the use condition of the storage space of the tablet personal computer;
The storage resource state evaluation module is used for evaluating the storage resource state of the tablet computer through the dynamic resource index model;
The prediction module is used for predicting the memory and external memory requirements of the ith application of the tablet personal computer at a future time point;
the adjusting module is used for adjusting the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at the future time point.
Further, a panel computer space intelligent management system, the monitoring module includes:
The current available capacity module is used for acquiring the available external memory capacity and the available internal memory capacity of the tablet personal computer at a time point t;
the method comprises the steps of acquiring an application condition module, and acquiring the number of applications in a tablet personal computer, the memory and the memory capacity occupied by each application at a time point t and the liveness index at the time point t;
The internal and external memory interaction condition acquisition module is used for acquiring the data exchange amount from the internal memory to the external memory and the data exchange amount from the external memory to the internal memory of the tablet personal computer at time t.
Further, a tablet computer space intelligent management system, the prediction module includes:
the historical data collection module is used for collecting historical data for the ith application, including memory usage, external memory usage, application activity and user usage behavior;
The preprocessing module is used for removing abnormal values in the data and converting the data format into a time sequence format;
a selection feature module for selecting relevant features from the collected data, the relevant features including frequency of use, period of activity and amount of data exchange of the application;
the new feature creation module is used for creating new features, creating time features according to the time period of application use, and then creating trend features according to the increasing trend of memory use;
A training model module for training the long-term and short-term memory network model using the selected features and the historical data;
and the prediction demand module is used for predicting the memory and external memory demands of the ith application at a specific future time point by using the trained model.
Further, a panel computer space intelligent management system, its characterized in that, the adjustment module includes:
the first adjusting module is used for not adjusting when the DRI value calculation result is 0, 0.3;
the second adjusting module is used for distributing storage space to the application according to the adjusting model when the DRI value calculation result is [0.3,0.6 ];
specifically, the adjustment model is:
；
Wherein, Indicating the memory footprint of the ith application at the next time point t +1,Representing the current time point t moment, the memory usage amount of the ith application, alpha is an adjustment speed parameter, the speed and the sensitivity of controlling resource adjustment, sigma is a memory adjustment preference coefficient, and reflects the preference degree of the memory when the resource is adjusted,/>Representing the memory predicted usage of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present.
；
Wherein,Representing the memory footprint of the ith application at the next point in time t +1,Representing the memory usage of the ith application at the current time point tRepresenting the predicted usage of the memory of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present.
And the third adjustment module is used for distributing storage space to the application according to the adjustment model when the DRI value calculation result is larger than 0.6.
The invention has the beneficial effects that: by monitoring the DRI model in real time, resource management can be more accurate and effective, so that each application can be ensured to allocate resources according to needs, and resource waste is avoided; the application resource demand is predicted in advance, and the system can adjust the resource on the premise of not influencing the user experience, so that the performance of important applications at key moments is ensured; the user reasonably distributes and adjusts the resources, so that the response speed and performance of the application are improved, and the overall user experience is improved; predicting future resource demands can help the system take action to avoid potential resource shortage problems, the system can allocate additional storage space in advance, avoid frequent internal and external memory calls to increase storage hardware loss, and simultaneously can respond to user use demands more quickly; by intelligently scheduling the resources, enough resource support is ensured when high-priority or critical tasks are executed, and the reliability and stability of the tablet personal computer are improved; according to the technical scheme, the resource management of the tablet personal computer is more intelligent, the use habit and the application requirement of a user can be adapted, the overall performance is optimized, the resource waste is avoided, and the smoother and stable user experience can be provided.
Drawings
Fig. 1 is a schematic diagram of a method for intelligently managing space of a tablet computer.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
According to one embodiment of the invention, a method for intelligently managing space of a tablet personal computer comprises the following steps:
Acquiring the use condition of the storage space of the tablet personal computer;
evaluating the storage resource state of the tablet computer through a dynamic resource index model;
Predicting the memory and external memory requirements of the ith application of the tablet computer at a future time point;
And adjusting the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at a future time point.
The working principle of the technical scheme is as follows: the method comprises the steps that a tablet personal computer monitors the use condition of a built-in storage space in real time, including but not limited to storage occupation, available space and the like of each application, a dynamic resource index model is used for evaluating and quantifying the storage resource state of the tablet personal computer, the current occupation condition and the use efficiency of resources are included, and the memory and external memory requirements of an ith application at a future time point are predicted according to historical use data and behavior analysis. The prediction model can help the system to predict the resource requirement of a specific application, so that corresponding optimization and adjustment can be made in advance; and combining the real-time monitoring data and the prediction result, and dynamically adjusting the storage space of the application by the system. For example, if it is predicted that an application will require more resources in the future, the system may allocate additional memory in advance, avoiding frequent memory calls from increasing memory hardware consumption, and simultaneously responding to user usage needs more quickly.
The technical scheme has the effects that: by monitoring the DRI model in real time, resource management can be more accurate and effective, so that each application can be ensured to allocate resources according to needs, and resource waste is avoided; the application resource demand is predicted in advance, and the system can adjust the resource on the premise of not influencing the user experience, so that the performance of important applications at key moments is ensured; the user reasonably distributes and adjusts the resources, so that the response speed and performance of the application are improved, and the overall user experience is improved; predicting future resource demands can help the system take action to avoid potential resource shortage problems, the system can allocate additional storage space in advance, avoid frequent internal and external memory calls to increase storage hardware loss, and simultaneously can respond to user use demands more quickly; by intelligently scheduling the resources, enough resource support is ensured when high-priority or critical tasks are executed, and the reliability and stability of the tablet personal computer are improved; according to the technical scheme, the resource management of the tablet personal computer is more intelligent, the use habit and the application requirement of a user can be adapted, the overall performance is optimized, the resource waste is avoided, and the smoother and stable user experience can be provided.
According to one embodiment of the invention, a method for intelligently managing the space of a tablet personal computer, which is used for acquiring the use condition of the storage space of the tablet personal computer, comprises the following steps:
Acquiring the available memory capacity and the available memory capacity of the tablet personal computer at a time point t;
Acquiring the number of applications in the tablet personal computer, and the memory capacity occupied by each application at a time point t and the liveness index at the time point t;
And acquiring the data exchange amount from the memory to the external memory and the data exchange amount from the external memory to the memory of the tablet personal computer at time t.
The working principle of the technical scheme is as follows: the system periodically detects and records the available external memory (such as SD card, built-in storage and the like) capacity and the available internal memory capacity of the tablet personal computer at the current time point t, which is helpful for evaluating the use condition of the whole resources and the residual resources;
Periodically collecting data about applications within the tablet computer: the quantity, the memory and the external memory capacity occupied by each application and the activity index of the application, wherein the activity index can consider parameters such as the starting frequency of the application, the foreground activity time and the like so as to quantify the using activity degree of the application; at a certain time t, detecting the amount of data exchange from memory to external memory (such as moving the data during running to flash memory for storage) and from external memory to internal memory (such as reading the data stored on external memory to internal memory for calculation by a CPU) performed by the tablet computer is helpful for understanding the I/O load and data flow conditions of the current system.
The technical scheme has the effects that: the system can better manage and optimize the allocation of the memory and the external memory through detailed real-time data, so that the resource waste is avoided; according to the resource occupation and the liveness index of the application, the system can provide more resources for the most commonly used and most important application, so that the performance of the system is improved; analyzing the data exchange quantity, the system can intelligently prefetch and buffer memory and adjust the data, reduce the delay and improve the response speed; the risk of resource shortage is foreseen, and the system can make adjustment or inform the user in advance to avoid overload condition; the reasonable memory and external memory management is beneficial to maintaining the stable operation of the tablet personal computer, and reduces the occurrence of application breakdown and system freezing; based on the liveness of the application, the system can more accurately allocate the resources in a targeted manner, i.e. the more active applications are prioritized; through the continuously collected data, a dynamic adjustment strategy can be formulated and updated to adapt to the changes of user behaviors and application demands, and the technical scheme intelligently manages the memory and the external memory resources of the tablet personal computer by comprehensively considering the use condition, the application demands and the system performances of various resources, so that the optimal system performance and user experience are maintained.
According to one embodiment of the invention, a method for intelligently managing space of a tablet personal computer, which evaluates the storage resource state of the tablet personal computer through a dynamic resource index model, comprises the following steps:
specifically, the dynamic resource index model is:
；
Wherein, Representing the resource competition coefficient between the ith and jth applications, N representing the total number of applications in the tablet computer,/>And/>Weight coefficients of resource contention and data exchange, respectively,/>，/>，/>Representing the amount of data exchange from memory to memory at time t,/>Representing the amount of data exchange from memory to memory at time t,/>Representing the available memory at time t,/>Representing the available memory at time t,/>Index of liveness indicating the ith application at time point t,/>And/>Respectively representing the memory and the memory space occupied by the ith application at time t,/>The coefficient of the specific gravity of the application activity and the resource demand is adjusted, and the value range is [0,1].
The resource competition coefficient quantifies the competition degree of two applications on resource use, the system performance monitoring tool can collect the resource use data (such as CPU, memory and I/O use condition) of the applications, analyze the application running log, especially the items recording the resource request and conflict condition, such as two applications accessing the same memory block or CPU core at the same time, divide the conflicting items by the total resource request items by a certain application, and calculate a quantified competition coefficient.
The working principle and the effect of the technical scheme are as follows: This reflects in part the resource requirements or usage of application i at point in time t, where/> Is a weight coefficient that may be used to balance the relationship between application liveness and its resource requirements. This can help identify which applications occupy the most resources,/>The ratio of the allocated resource amount to the total available resource amount of each application program is considered, and the memory ratio occupied by the application/>And memory ratio/>Is used for adjusting dynamic resource index, and reflects the utilization efficiency of resources,/>This portion represents the sum of the resource competition coefficients between applications, taking into account not only the resource usage of a single application, but also its interactions and competition with other applications. In a multitasking environment, applications may need to compete for CPU time slices, memory space, etc., resources,/>Is the weight coefficient of the competition in the resource index, and the data exchange frequency between the memory and the external memory influences the overall performance of the system, especially the I/O performance,/>This part represents the amount of data exchanged between memory and external memory at point in time t,Is the weight coefficient of the data exchange in the resource index. By comprehensively considering various factors, the design provides a comprehensive resource assessment index, so that resource monitoring and management is not limited to a single dimension; the index is dynamically calculated, can reflect the use condition of the resource changing along with time, and provides decision support for dynamic resource adjustment; taking resource competition and data exchange frequency between applications into consideration, the system can predict bottlenecks possibly occurring in the future and make adjustment in advance; the monitoring of the data exchange amount is beneficial to optimizing the data flow and the buffer strategy, reducing the I/O waiting time and improving the overall system performance; by quantifying the resource competition, the system can more reasonably allocate the resources and avoid performance degradation caused by resource competition; the index can be used as a powerful basis for system resource management decision-making to help a system administrator or an automation tool to make more reasonable resource scheduling decision-making; the design can provide more accurate resource state evaluation and management capability for tablet computers or other similar devices, which is important to ensure stable operation and high-efficiency performance of the devices in complex multi-task environments.
According to one embodiment of the invention, a method for intelligently managing space of a tablet personal computer predicts the memory and external memory requirements of an ith application of the tablet personal computer at a future time point, and comprises the following steps:
Collecting historical data for an ith application, including memory usage, external memory usage, application activity and user usage behavior;
Removing abnormal values in the data, and converting the data format into a time sequence format;
Selecting relevant features from the collected data, the relevant features including frequency of use, period of activity and amount of data exchange of the application;
Creating a new feature, and then creating a trend feature according to the increasing trend of the memory usage after creating a time feature according to the time period of application usage;
Training a long-term and short-term memory network model using the selected features and the historical data;
and predicting the memory and external memory requirements of the ith application at a specific future time point by using the trained model.
The working principle of the technical scheme is as follows: the step needs to record the information such as the memory and the memory usage of the application at regular intervals, and the data is the basis for creating a model and needs to ensure the accuracy and the integrity of the model; data cleansing, outliers in the data may affect the accuracy of the model. By removing abnormal values, a data set which reflects the application behaviors more cleanly and more truly can be obtained; converting the data format into a time sequence so that the long-short-term memory network can understand and learn the time-varying pattern of the data; feature selection, namely selecting features related to future resource demands from historical data, such as the use frequency of an application, an active time period and the like; creating new features, such as temporal features, to reflect usage patterns of the application at different time periods, and trend features to predict future demands from the trend of memory usage; training a long-short term memory network using the features and data, the LSTM model being capable of capturing long-term dependencies and complex patterns in time series data; and predicting the memory and external memory requirements of the ith application at a specific time point in the future by using the trained model, and providing support for resource management.
The technical scheme has the effects that: by predicting the application resource demand, allocation adjustments can be made in advance before the resources become scarce to avoid performance bottlenecks; if a certain application is expected to have high resource requirements, the system can make resource adjustment for the application in advance, so that the application performance and the user experience are improved; the prediction can be carried out at different time points to ensure that the resource management system can dynamically allocate the memory and the external memory resources; knowing the future resource demands of an application can help reduce over-allocation and idle resources, making resource allocation more efficient; the LSTM model provides powerful support for the automation of resource management, and more intelligent and accurate resource management decisions can be made through machine learning; by the technical scheme, a system administrator or an automatic management tool can better understand the use mode of the application and forecast the future resource demand, so that the resources on the tablet personal computer or other devices can be managed more effectively and intelligently, and the method is particularly suitable for environments requiring high reliability and performance.
An embodiment of the present invention provides a method for intelligently managing a space of a tablet computer, wherein the adjusting an application storage space occupation condition according to a current storage resource state of the tablet computer and a requirement condition of an ith application for a memory and an external memory at a future time point includes:
When the DRI value calculation result is 0,0.3, not adjusting;
when the DRI value calculation result is [0.3,0.6], distributing storage space for the application according to the adjustment model;
specifically, the adjustment model is:
；
Wherein, Indicating the memory footprint of the ith application at the next time point t +1,Representing the current time point t moment, the memory usage amount of the ith application, alpha is an adjustment speed parameter, the speed and the sensitivity of controlling resource adjustment, sigma is a memory adjustment preference coefficient, and reflects the preference degree of the memory when the resource is adjusted,/>Representing the memory predicted usage of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present.
；
Wherein,Representing the memory footprint of the ith application at the next point in time t +1,Representing the memory usage of the ith application at the current time point tRepresenting the predicted usage of the memory of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present.
When the DRI value calculation result is larger than 0.6, storage space is allocated to the application according to the adjustment model, and the value of alpha is larger.
The working principle and the effect of the technical scheme are as follows: the adjustment model is not just passive in response to current demand, but rather is able to predict future trends and make the most beneficial resource adjustments based thereon. The model realizes multi-objective optimization by calculating resource use, adjustment cost and expected benefit, which means that the model can predict future demands and measure the input-output ratio of resource adjustment while meeting the current demands; the model can adapt to the continuously-changing use mode by introducing the cost and benefit consideration of dynamic adjustment and predicting the future demand, and the flexibility and the response rapidity of resource management are maintained; weighing the adjustment cost and the adjusted expected benefit, when the DRI value calculation result is 0,0.3, indicating that the current system resource has a margin, and not adjusting at this time, ensuring that the system can not make excessive resource adjustment for a small amount of performance improvement; by utilizing deep learning, the model iterates on the basis of continuously collected data, continuously improving the accuracy of prediction and the effectiveness of resource allocation; the performance is improved, the delay can be reduced by the predictive resource adjustment, and the application response speed is improved, so that the interactive experience of a user is improved; the resource utilization rate is improved, the waste can be reduced by reasonable resource allocation, and the idle resources are ensured to be fully utilized; unnecessary resource adjustment is reduced, and excessive or uneconomical resource investment is avoided; intelligent prediction and dynamic adjustment enable the system to adapt to changing environments and requirements; continuous performance monitoring, system performance and user behavior, continuous monitoring facilitates real-time tuning and improvement of policies. The design provides a self-learning and self-adapting resource management framework, aims at reducing resource adjustment cost and improving user satisfaction and system stability while maintaining high-efficiency performance.
In one embodiment of the invention, a system for intelligently managing space of a tablet computer comprises:
The monitoring module is used for acquiring the use condition of the storage space of the tablet personal computer;
The storage resource state evaluation module is used for evaluating the storage resource state of the tablet computer through the dynamic resource index model;
The prediction module is used for predicting the memory and external memory requirements of the ith application of the tablet personal computer at a future time point;
the adjusting module is used for adjusting the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at the future time point.
The working principle of the technical scheme is as follows: the method comprises the steps that a tablet personal computer monitors the use condition of a built-in storage space in real time, including but not limited to storage occupation, available space and the like of each application, a dynamic resource index model is used for evaluating and quantifying the storage resource state of the tablet personal computer, the current occupation condition and the use efficiency of resources are included, and the memory and external memory requirements of an ith application at a future time point are predicted according to historical use data and behavior analysis. The prediction model can help the system to predict the resource requirement of a specific application, so that corresponding optimization and adjustment can be made in advance; and combining the real-time monitoring data and the prediction result, and dynamically adjusting the storage space of the application by the system. For example, if it is predicted that an application will require more resources in the future, the system may allocate additional memory in advance, avoiding frequent memory calls from increasing memory hardware consumption, and simultaneously responding to user usage needs more quickly.
The technical scheme has the effects that: through real-time monitoring and a DRI model, resource management can be more accurate and effective, so that each application can be ensured to allocate resources according to needs, and resource waste is avoided; the application resource demand is predicted in advance, and the system can adjust the resource on the premise of not influencing the user experience, so that the performance of important applications at key moments is ensured; the user reasonably distributes and adjusts the resources, so that the response speed and performance of the application are improved, and the overall user experience is improved; predicting future resource demands can help the system take action to avoid potential resource shortage problems, the system can allocate additional storage space in advance, avoid frequent internal and external memory calls to increase storage hardware loss, and simultaneously can respond to user use demands more quickly; by intelligently scheduling the resources, enough resource support is ensured when high-priority or critical tasks are executed, and the reliability and stability of the tablet personal computer are improved; according to the technical scheme, the resource management of the tablet personal computer is more intelligent, the use habit and the application requirement of a user can be adapted, the overall performance is optimized, the resource waste is avoided, and the smoother and stable user experience can be provided.
In one embodiment of the present invention, a system for intelligently managing space of a tablet computer, the monitoring module includes:
The current available capacity module is used for acquiring the available external memory capacity and the available internal memory capacity of the tablet personal computer at a time point t;
the method comprises the steps of acquiring an application condition module, and acquiring the number of applications in a tablet personal computer, the memory and the memory capacity occupied by each application at a time point t and the liveness index at the time point t;
The internal and external memory interaction condition acquisition module is used for acquiring the data exchange amount from the internal memory to the external memory and the data exchange amount from the external memory to the internal memory of the tablet personal computer at time t.
The working principle of the technical scheme is as follows: the system periodically detects and records the available external memory (such as SD card, built-in storage and the like) capacity and the available internal memory capacity of the tablet personal computer at the current time point t, which is helpful for evaluating the use condition of the whole resources and the residual resources;
Periodically collecting data about applications within the tablet computer: the quantity, the memory and the external memory capacity occupied by each application and the activity index of the application, wherein the activity index can consider parameters such as the starting frequency of the application, the foreground activity time and the like so as to quantify the using activity degree of the application; at a certain time t, detecting the amount of data exchange from memory to external memory (such as moving the data during running to flash memory for storage) and from external memory to internal memory (such as reading the data stored on external memory to internal memory for calculation by a CPU) performed by the tablet computer is helpful for understanding the I/O load and data flow conditions of the current system.
The technical scheme has the effects that: the system can better manage and optimize the allocation of the memory and the external memory through detailed real-time data, so that the resource waste is avoided; according to the resource occupation and the liveness index of the application, the system can provide more resources for the most commonly used and most important application, so that the performance of the system is improved; analyzing the data exchange quantity, the system can intelligently prefetch and buffer memory and adjust the data, reduce the delay and improve the response speed; the risk of resource shortage is foreseen, and the system can make adjustment or inform the user in advance to avoid overload condition; the reasonable memory and external memory management is beneficial to maintaining the stable operation of the tablet personal computer, and reduces the occurrence of application breakdown and system freezing; based on the liveness of the application, the system can more accurately allocate the resources in a targeted manner, i.e. the more active applications are prioritized; through the continuously collected data, a dynamic adjustment strategy can be formulated and updated to adapt to the changes of user behaviors and application demands, and the technical scheme intelligently manages the memory and the external memory resources of the tablet personal computer by comprehensively considering the use condition, the application demands and the system performances of various resources, so that the optimal system performance and user experience are maintained.
An embodiment of the invention provides an intelligent management system for a tablet personal computer space, wherein the prediction module comprises:
the historical data collection module is used for collecting historical data for the ith application, including memory usage, external memory usage, application activity and user usage behavior;
The preprocessing module is used for removing abnormal values in the data and converting the data format into a time sequence format;
a selection feature module for selecting relevant features from the collected data, the relevant features including frequency of use, period of activity and amount of data exchange of the application;
the new feature creation module is used for creating new features, creating time features according to the time period of application use, and then creating trend features according to the increasing trend of memory use;
A training model module for training the long-term and short-term memory network model using the selected features and the historical data;
and the prediction demand module is used for predicting the memory and external memory demands of the ith application at a specific future time point by using the trained model.
The working principle of the technical scheme is as follows: the step needs to record the information such as the memory and the memory usage of the application at regular intervals, and the data is the basis for creating a model and needs to ensure the accuracy and the integrity of the model; data cleansing, outliers in the data may affect the accuracy of the model. By removing abnormal values, a data set which reflects the application behaviors more cleanly and more truly can be obtained; converting the data format into a time sequence so that the long-short-term memory network can understand and learn the time-varying pattern of the data; feature selection, namely selecting features related to future resource demands from historical data, such as the use frequency of an application, an active time period and the like; creating new features, such as temporal features, to reflect usage patterns of the application at different time periods, and trend features to predict future demands from the trend of memory usage; training a long-short term memory network using the features and data, the LSTM model being capable of capturing long-term dependencies and complex patterns in time series data; and predicting the memory and external memory requirements of the ith application at a specific time point in the future by using the trained model, and providing support for resource management.
The technical scheme has the effects that: by predicting the application resource demand, allocation adjustments can be made in advance before the resources become scarce to avoid performance bottlenecks; if a certain application is expected to have high resource requirements, the system can make resource adjustment for the application in advance, so that the application performance and the user experience are improved; the prediction can be carried out at different time points to ensure that the resource management system can dynamically allocate the memory and the external memory resources; knowing the future resource demands of an application can help reduce over-allocation and idle resources, making resource allocation more efficient; the LSTM model provides powerful support for the automation of resource management, and more intelligent and accurate resource management decisions can be made through machine learning; by the technical scheme, a system administrator or an automatic management tool can better understand the use mode of the application and forecast the future resource demand, so that the resources on the tablet personal computer or other devices can be managed more effectively and intelligently, and the method is particularly suitable for environments requiring high reliability and performance.
In one embodiment of the present invention, a system for intelligently managing space of a tablet computer, the adjustment module includes:
the first adjusting module is used for not adjusting when the DRI value calculation result is 0, 0.3;
the second adjusting module is used for distributing storage space to the application according to the adjusting model when the DRI value calculation result is [0.3,0.6 ];
specifically, the adjustment model is:
；
Wherein, Indicating the memory footprint of the ith application at the next time point t +1,Representing the current time point t moment, the memory usage amount of the ith application, alpha is an adjustment speed parameter, the speed and the sensitivity of controlling resource adjustment, sigma is a memory adjustment preference coefficient, and reflects the preference degree of the memory when the resource is adjusted,/>Representing the memory predicted usage of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present.
；
Wherein,Representing the memory footprint of the ith application at the next point in time t +1,Representing the memory usage of the ith application at the current time point tRepresenting the predicted usage of the memory of the ith application at time t,/>Is the adjustment factor for response time and adjustment cost, and Δt is the time difference from the last adjustment to the present.
And the third adjusting module is used for distributing storage space to the application according to the adjusting model when the DRI value calculation result is larger than 0.6, and the value of alpha is larger at the moment.
The working principle and the effect of the technical scheme are as follows: the adjustment model is not just passive in response to current demand, but rather is able to predict future trends and make the most beneficial resource adjustments based thereon. The model realizes multi-objective optimization by calculating resource use, adjustment cost and expected benefit, which means that the model can predict future demands and measure the input-output ratio of resource adjustment while meeting the current demands; the model can adapt to the continuously-changing use mode by introducing the cost and benefit consideration of dynamic adjustment and predicting the future demand, and the flexibility and the response rapidity of resource management are maintained; weighing the adjustment cost and the adjusted expected benefit, when the DRI value calculation result is 0,0.3, indicating that the current system resource has a margin, and not adjusting at this time, ensuring that the system can not make excessive resource adjustment for a small amount of performance improvement; by utilizing deep learning, the model iterates on the basis of continuously collected data, continuously improving the accuracy of prediction and the effectiveness of resource allocation; the performance is improved, the delay can be reduced by the predictive resource adjustment, and the application response speed is improved, so that the interactive experience of a user is improved; the resource utilization rate is improved, the waste can be reduced by reasonable resource allocation, and the idle resources are ensured to be fully utilized; unnecessary resource adjustment is reduced, and excessive or uneconomical resource investment is avoided; intelligent prediction and dynamic adjustment enable the system to adapt to changing environments and requirements; continuous performance monitoring, system performance and user behavior, continuous monitoring facilitates real-time tuning and improvement of policies. The design provides a self-learning and self-adapting resource management framework, aims at reducing resource adjustment cost and improving user satisfaction and system stability while maintaining high-efficiency performance.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (9)
1. The intelligent management method for the space of the tablet personal computer is characterized by comprising the following steps:
Acquiring the use condition of the storage space of the tablet personal computer;
evaluating the storage resource state of the tablet computer through a dynamic resource index model;
Predicting the memory and external memory requirements of the ith application of the tablet computer at a future time point;
And adjusting the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at a future time point.
2. The method for intelligently managing space of a tablet computer according to claim 1, wherein the obtaining the usage condition of the storage space of the tablet computer comprises:
Acquiring the available memory capacity and the available memory capacity of the tablet personal computer at a time point t;
Acquiring the number of applications in the tablet personal computer, and the memory capacity occupied by each application at a time point t and the liveness index at the time point t;
And acquiring the data exchange amount from the memory to the external memory and the data exchange amount from the external memory to the memory of the tablet personal computer at time t.
3. The method for intelligently managing space of a tablet computer according to claim 1, wherein the evaluating the storage resource status of the tablet computer by the dynamic resource index model comprises:
specifically, the dynamic resource index model is:
；
Wherein, Represents the resource competition coefficient between the ith and jth applications, N represents the total number of applications in the tablet computer,And/>Weight coefficients of resource contention and data exchange, respectively,/>Representing the amount of data exchange from memory to memory at time t,/>Representing the amount of data exchange from memory to memory at time t,/>Indicating the available memory at point in time t,Representing the available memory at time t,/>Index of liveness indicating the ith application at time point t,/>And/>Respectively representing the memory and the memory space occupied by the ith application at time t,/>Is a coefficient for adjusting the specific gravity of the application liveness and the resource demand,/>Is/>Is the maximum possible value of (a).
4. The intelligent management method of tablet computer space according to claim 1, wherein predicting the memory and external memory requirements of the ith application of the tablet computer at a future point in time comprises:
Collecting historical data for an ith application, including memory usage, external memory usage, application activity and user usage behavior;
Removing abnormal values in the data, and converting the data format into a time sequence format;
Selecting relevant features from the collected data, the relevant features including frequency of use, period of activity and amount of data exchange of the application;
Creating a new feature, and then creating a trend feature according to the increasing trend of the memory usage after creating a time feature according to the time period of application usage;
Training a long-term and short-term memory network model using the selected features and the historical data;
and predicting the memory and external memory requirements of the ith application at a specific future time point by using the trained model.
5. The intelligent management method of the space of the tablet computer according to claim 1, wherein the adjusting the occupation of the application storage space according to the current storage resource state of the tablet computer and the requirement of the ith application for the memory and the external memory at the future time point comprises:
When the DRI value calculation result is 0,0.3, not adjusting;
when the DRI value calculation result is [0.3,0.6], distributing storage space for the application according to the adjustment model;
specifically, the adjustment model is:
；
Wherein, Indicating the memory footprint of the ith application at the next time point t +1,Representing the current time point t moment, the memory usage amount of the ith application, alpha is an adjustment speed parameter, the speed and the sensitivity of controlling resource adjustment, sigma is a memory adjustment preference coefficient, and reflects the preference degree of the memory when the resource is adjusted,/>Representing the memory predicted usage of the ith application at time t,/>An adjustment coefficient which is a response time and an adjustment cost, and Δt is a time difference from the last adjustment to the present;
；
Wherein, Representing the memory occupancy of the ith application at the next time point t+1,/>Representing the memory usage of the ith application at the current time point tRepresenting the predicted usage of the memory of the ith application at time t,/>An adjustment coefficient which is a response time and an adjustment cost, and Δt is a time difference from the last adjustment to the present;
And when the DRI value calculation result is larger than 0.6, allocating storage space for the application according to the adjustment model.
6. An intelligent management system for a tablet computer space, the system comprising:
The monitoring module is used for acquiring the use condition of the storage space of the tablet personal computer;
The storage resource state evaluation module is used for evaluating the storage resource state of the tablet computer through the dynamic resource index model;
The prediction module is used for predicting the memory and external memory requirements of the ith application of the tablet personal computer at a future time point;
the adjusting module is used for adjusting the occupation condition of the application storage space according to the current storage resource state of the tablet personal computer and the requirement condition of the ith application on the memory and the external memory at the future time point.
7. The intelligent management system of tablet computer space of claim 6, wherein the monitoring module comprises:
The current available capacity module is used for acquiring the available external memory capacity and the available internal memory capacity of the tablet personal computer at a time point t;
the method comprises the steps of acquiring an application condition module, and acquiring the number of applications in a tablet personal computer, the memory and the memory capacity occupied by each application at a time point t and the liveness index at the time point t;
The internal and external memory interaction condition acquisition module is used for acquiring the data exchange amount from the internal memory to the external memory and the data exchange amount from the external memory to the internal memory of the tablet personal computer at time t.
8. The intelligent management system of tablet computer space of claim 6, wherein the prediction module comprises:
the historical data collection module is used for collecting historical data for the ith application, including memory usage, external memory usage, application activity and user usage behavior;
The preprocessing module is used for removing abnormal values in the data and converting the data format into a time sequence format;
a selection feature module for selecting relevant features from the collected data, the relevant features including frequency of use, period of activity and amount of data exchange of the application;
the new feature creation module is used for creating new features, creating time features according to the time period of application use, and then creating trend features according to the increasing trend of memory use;
A training model module for training the long-term and short-term memory network model using the selected features and the historical data;
and the prediction demand module is used for predicting the memory and external memory demands of the ith application at a specific future time point by using the trained model.
9. The intelligent management system of tablet computer space of claim 6, wherein the adjustment module comprises:
the first adjusting module is used for not adjusting when the DRI value calculation result is 0, 0.3;
the second adjusting module is used for distributing storage space to the application according to the adjusting model when the DRI value calculation result is [0.3,0.6 ];
specifically, the adjustment model is:
；
Wherein, Indicating the memory footprint of the ith application at the next time point t +1,Representing the current time point t moment, the memory usage amount of the ith application, alpha is an adjustment speed parameter, the speed and the sensitivity of controlling resource adjustment, sigma is a memory adjustment preference coefficient, and reflects the preference degree of the memory when the resource is adjusted,/>Representing the memory predicted usage of the ith application at time t,/>An adjustment coefficient which is a response time and an adjustment cost, and Δt is a time difference from the last adjustment to the present;
；
Wherein, Representing the memory occupancy of the ith application at the next time point t+1,/>Representing the memory usage of the ith application at the current time point tRepresenting the predicted usage of the memory of the ith application at time t,/>An adjustment coefficient which is a response time and an adjustment cost, and Δt is a time difference from the last adjustment to the present;
And the third adjustment module is used for distributing storage space to the application according to the adjustment model when the DRI value calculation result is larger than 0.6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410504055.XA CN118092817A (en) | 2024-04-25 | 2024-04-25 | Intelligent management method and system for space of tablet personal computer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410504055.XA CN118092817A (en) | 2024-04-25 | 2024-04-25 | Intelligent management method and system for space of tablet personal computer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118092817A true CN118092817A (en) | 2024-05-28 |
Family
ID=91160316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410504055.XA Pending CN118092817A (en) | 2024-04-25 | 2024-04-25 | Intelligent management method and system for space of tablet personal computer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118092817A (en) |
-
2024
- 2024-04-25 CN CN202410504055.XA patent/CN118092817A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110096349B (en) | Job scheduling method based on cluster node load state prediction | |
| WO2021179462A1 (en) | Improved quantum ant colony algorithm-based spark platform task scheduling method | |
| EP4068090A1 (en) | Container scheduling method and apparatus, and non-volatile computer-readable storage medium | |
| CN108770017B (en) | Dynamic equalization method and system for wireless resources | |
| CN111414070B (en) | Case power consumption management method and system, electronic device and storage medium | |
| CN105607952B (en) | Method and device for scheduling virtualized resources | |
| WO2020172852A1 (en) | Computing resource scheduling method, scheduler, internet of things system, and computer readable medium | |
| Li et al. | Cost-aware automatic scaling and workload-aware replica management for edge-cloud environment | |
| CN116974768A (en) | Calculation power scheduling method based on deep learning | |
| CN117349026B (en) | Distributed computing power scheduling system for AIGC model training | |
| CN103442087A (en) | Web service system access volume control device and method based on response time trend analysis | |
| CN117076882A (en) | Dynamic prediction management method for cloud service resources | |
| CN115145709B (en) | Low-carbon big data artificial intelligence method and medical health state system | |
| CN118092817A (en) | Intelligent management method and system for space of tablet personal computer | |
| CN115658319A (en) | Resource scheduling method, system, device and storage medium | |
| KR102556093B1 (en) | Reward generation method to reduce peak load of electric power and action control apparatus performing the same method | |
| CN115168042A (en) | Management method and device of monitoring cluster, computer storage medium and electronic equipment | |
| CN114860449A (en) | Data processing method, device, equipment and storage medium | |
| Lili et al. | A Markov chain based resource prediction in computational grid | |
| CN112052087A (en) | Deep learning training system and method for dynamic resource adjustment and migration | |
| CN118227291A (en) | Multi-task scheduling method and device based on heterogeneous distributed cluster | |
| CN117519913B (en) | Method and system for elastically telescoping scheduling of container memory resources | |
| CN117972367B (en) | Data storage prediction method, data storage subsystem and intelligent computing platform | |
| Milocco et al. | An Evaluation Method of Optimal Cost Saving in a Data Center with Proactive Management | |
| Prangchumpol et al. | Resource Allocation with Exponential Model Prediction for Server Virtualization. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |