A cloud storage optimization management system and method based on data analysis

Abstract

The application discloses a cloud storage optimization management system and method based on data analysis, relates to the technical field of cloud storage optimization management, and comprises a cloud server data acquisition module, a cloud server data processing module, a cloud server state judgment module, a cloud server label setting module, a cloud server data distribution module and a cloud server label updating module. Real-time data acquisition is performed on the cloud server, any cloud server is analyzed, the state of the cloud server is judged according to the processed data, a label is stamped, the cloud server with a fluctuation label is analyzed, the data in the cloud server with the fluctuation label is redistributed, the state of the cloud server is judged again after the resource distribution is completed, and the label of the cloud server is updated. Through intelligent scheduling based on data driving, the application effectively balances the resource utilization rates of the servers in the cluster, and reduces resource waste caused by lagging manual intervention or experience judgment deviation.

Description

Technical Field

[0001] This invention relates to the field of cloud storage optimization management technology, specifically to a cloud storage optimization management system and method based on data analysis. Background Technology

[0002] In existing cloud storage management, server resource allocation largely relies on preset rules or manual experience, making it difficult to accurately perceive dynamic load changes in individual nodes in real time. For example, when a server experiences a surge in stored data or a sharp fluctuation in user connections due to sudden traffic, traditional methods cannot quickly identify the abnormal state. This often leads to single-point overload due to delayed data migration, resulting in decreased service response speed or even interruption. Furthermore, the lack of comprehensive analysis of multi-dimensional parameters such as storage overload rate, user connection fluctuation, and packet loss rate easily causes resource imbalance within the cluster. Some servers operate at high load for extended periods, while others remain idle, making it difficult to optimize overall utilization. For servers experiencing fluctuations, the lack of real-time dynamic tagging management and automated data redistribution mechanisms can exacerbate hardware wear and tear, increasing maintenance costs. Conversely, servers that remain idle for extended periods cannot automatically enter low-power mode, resulting in energy waste. Especially in large-scale data centers, manual inspection and adjustment are inefficient and cannot meet the real-time monitoring needs of massive numbers of servers, leading to a dual challenge of operational costs and energy consumption control. Traditional cloud storage systems rely heavily on emergency handling after a failure occurs, lacking the ability to anticipate potential performance bottlenecks. When the amount of stored data on a server approaches its maximum capacity or user connection counts fluctuate abnormally, the risk level cannot be assessed in real time using quantitative indicators. Repair processes are often only initiated after issues such as increased packet loss rates and service timeouts become apparent, resulting in delayed fault handling. Furthermore, the lack of periodic re-inspection and closed-loop management mechanisms for server status makes it difficult to continuously track optimization effects. This makes the system susceptible to chain reactions caused by performance degradation of local nodes in high-concurrency or sudden data surge scenarios, resulting in insufficient overall stability and fault tolerance. Summary of the Invention

[0003] The purpose of this invention is to provide a cloud storage optimization management system and method based on data analysis to solve the problems mentioned in the background art.

[0004] To address the aforementioned technical problems, this invention provides the following technical solution: a cloud storage optimization management method based on data analysis, comprising the following steps:

[0005] S1. Monitor the status of the cloud server in real time and collect data from the cloud server in real time;

[0006] S2. Analyze any cloud server and process the data on any cloud server;

[0007] S3. After data processing, determine the status of the cloud server based on the processed data;

[0008] S4. Tag the cloud server to indicate its status;

[0009] S5. Analyze the cloud servers with fluctuation tags and redistribute the data in the cloud servers with fluctuation tags.

[0010] S6. After completing resource allocation, reassess the status of the cloud server and update the cloud server's label.

[0011] Furthermore, in step S1, any cloud server is analyzed, and its status is monitored in real time. The maximum storage capacity of the cloud server is β. When the cloud server is powered off, it is marked as an idle cloud server. When the cloud server is powered on, real-time data is collected from it, with a collection interval of α. Data is collected I times within one collection cycle, including the cloud server's stored data volume, packet loss rate, and number of adjacent users. The set of stored data volume is {A1, A2, ..., A...}. i ,…,A I The packet loss rate is R, and the set of adjacent user numbers is {B1, B2, ..., B}. i ,…,B I}, where the number of adjacent users represents the number of users interacting with the cloud server, and A i B represents the amount of data stored on the cloud server during the i-th data collection. i This represents the number of adjacent users of the cloud server at the time of the i-th data collection. By monitoring and accurately marking the cloud server status in real time, it clearly distinguishes between idle and running servers, providing an intuitive basis for the rational allocation of resources, avoiding waste of idle resources, and improving overall utilization efficiency. For running servers, key information such as stored data volume, packet loss rate, and the number of adjacent users are collected periodically, enabling real-time monitoring of server load, data storage status, and network interaction quality. By continuously tracking this dynamic data, administrators can promptly detect abnormal fluctuations in server operation, including potential problems such as storage approaching saturation, unstable network connections, or a sudden increase in user access. They can then take proactive measures such as capacity expansion, network configuration optimization, or load balancing to effectively ensure the stability and reliability of cloud services, providing users with a smoother interactive experience, while reducing the risk of system failures and achieving refined and dynamic management of cloud servers.

[0012] Furthermore, in step S2, an analysis is performed on any cloud server. Within any collection period, the average amount of stored data is A0. During the i-th data collection period, the cloud server's storage over-limit index is W. Ai W Ai =Ai / (W I *β), the W I For the predetermined optimal load percentage of the cloud server, W I *β represents the optimal load data volume for the cloud server, when W Ai When the value is greater than 1, increment the storage overrun number by one, and substitute each i = 1, 2, ..., I to obtain the storage overrun number of the cloud server as w. A This leads to the cloud server's storage over-limit rate being W. A =w A / I; Calculate the fluctuation level W of the number of user connections within any given collection period. B W B =σ / (B MAX -B MIN ), where σ is the set {B1, B2, ..., B}. i ,…,B I The standard deviation of B MAX For the set {B1, B2, ..., B... i ,…,B I The maximum value in}, B MIN For the set {B1, B2, ..., B... i ,…,B I The minimum value in the data storage and user connection status of cloud servers is used to accurately assess the server's operational status at the data storage and user interaction levels. For stored data, by comparing the actual storage volume with the optimal load level, it determines whether the storage is within a reasonable range, promptly identifies abnormal situations where storage continuously exceeds limits, and provides a basis for decisions such as data migration and capacity expansion, avoiding service performance impacts due to storage overload. For user connection counts, by measuring their fluctuations, the stability and balance of user access can be intuitively reflected, helping administrators identify abnormal fluctuations such as sudden increases or decreases in user access, and proactively optimize resource allocation or adjust service strategies to ensure the server can smoothly handle user access demands of different scales. The combination of these two methods achieves a multi-dimensional assessment of server operational status, helping administrators accurately locate potential problems, improve resource utilization efficiency, and ensure the stability and reliability of cloud services and continuous optimization of user experience.

[0013] Furthermore, in step S3, the fluctuation coefficient D of any cloud server is calculated, where D = (K A *W A / W A0 +K B *W B )*(1+R), where W A0 K represents the average storage over-limit rate of cloud servers that are currently powered on. A W A / WA0 The weighting of the impact of cloud server volatility coefficient, K B This represents the weight of the impact of the stability of user connection counts on the volatility coefficient of the cloud server, typically K. A and K B It can be directly set to 1, setting the fluctuation threshold D0 = (K A +K B The cloud server is considered to be in a stable state when D < D0 and R < R0; otherwise, it is considered to be in a fluctuating state. E is a pre-defined fluctuation tolerance coefficient, which can be set to 1.3. By comprehensively considering the cloud server's storage pressure, user connection fluctuations, and network quality, the stability of server operation can be comprehensively and accurately determined. By integrating key indicators such as storage exceeding limits, user access fluctuations, and packet loss rates, potential risks in data storage, user interaction, and network transmission can be effectively identified. When all indicators are within a reasonable range, the server's stable state can be confirmed in a timely manner, ensuring continuous service operation. When abnormal storage pressure, sudden changes in user access, or unstable network connections occur, a fluctuating state can be quickly identified, prompting administrators to intervene in advance and mitigate risks by adjusting resource allocation, optimizing network configuration, or increasing storage capacity. This dynamic and comprehensive evaluation method achieves precise control over the server's operating status, helps improve resource utilization efficiency, reduces the probability of system failure, ensures the continuous and stable operation of cloud services, and provides users with a more reliable service experience.

[0014] Furthermore, in step S4, the cloud servers in the power-on state are analyzed, and cloud servers in a stable state are labeled with a stable tag, while cloud servers in a fluctuating state are labeled with a fluctuating tag. The stable tag indicates that the cloud server can be used for resource allocation, that is, the cloud server can continue to add data, while the fluctuating tag indicates that the cloud server needs to allocate resources, that is, the cloud server needs to allocate resources to other cloud servers.

[0015] Furthermore, in step S5, the cloud server with the fluctuation tag is analyzed, and the neighboring cloud servers of the cloud server are counted. The neighboring cloud servers are connected to any cloud server with the fluctuation tag. There are a total of M neighboring cloud servers with the stable tag, and the set of neighboring cloud servers with the stable tag is {Z1, Z2, ..., Z...}. m ,…,Z M}, where Z m This represents the m-th neighboring cloud server with a stable label;

[0016] Extract the amount of data to be allocated, F, from cloud server Z0 in reverse chronological order, and allocate resources based on the amount of data to be allocated, where F = A. MAX -WI *β, the amount of data F allocated to the m-th neighboring cloud server with a stable label. m :

[0017]

[0018] Among them W m_B This represents the fluctuation level of the m-th adjacent cloud server with a stable label. By analyzing cloud servers with fluctuation labels and leveraging adjacent stable cloud servers, dynamic allocation of data resources can be achieved, effectively alleviating the operational pressure on fluctuating servers and improving the overall stability of the cloud service architecture. By identifying and utilizing adjacent stable servers as targets for data allocation, data exceeding reasonable load on fluctuating servers can be promptly transferred to nodes in good operating condition, avoiding performance degradation or service interruption due to excessive load on a single server. Differentiated data allocation based on the fluctuation level of stable servers ensures that resource allocation is more aligned with the actual carrying capacity of each server. While fully utilizing the idle resources of stable servers, it avoids placing additional burdens on them, achieving a balanced distribution of data storage and processing capabilities within the cluster. This resource scheduling mechanism based on the status of adjacent nodes can quickly respond to abnormal server fluctuations, reduce manual intervention through automated load shifting, enhance the system's self-healing capabilities, ensure that cloud services maintain overall stable operation even under localized pressure, optimize resource utilization efficiency, and reduce potential failure risks.

[0019] Furthermore, in step S6, after resource allocation is completed, if cloud server Z0 is in a stable state, its fluctuation tag is deleted and a stable tag is added; if cloud server Z0 is still in a fluctuating state, its fluctuation tag is saved, and resources are allocated to it again until it reaches a stable state, at which point a stable tag is added. By dynamically adjusting the cloud server's status tag and continuously allocating resources, a closed-loop service stability assurance mechanism is constructed. After data resource allocation is completed, the decision to retain or delete the fluctuation tag is based on the server's real-time status, enabling accurate identification and dynamic tracking of the server's operating status. For servers still in a fluctuating state, repeated resource allocation until they stabilize avoids the limitations of a single adjustment, ensuring that potential problems are thoroughly resolved and preventing systemic risks caused by accumulated local pressure.

[0020] A cloud storage optimization and management system based on data analysis, comprising: a cloud server data acquisition module, a cloud server data processing module, a cloud server status judgment module, a cloud server tag setting module, a cloud server data allocation module, and a cloud server tag update module;

[0021] The cloud server data acquisition module is used to monitor the status of the cloud server in real time and to collect data from the cloud server in real time.

[0022] The cloud server data processing module is used to analyze any cloud server and process the data of any cloud server.

[0023] The cloud server status judgment module is used to determine the status of the cloud server based on the processed data after data processing.

[0024] The cloud server tagging module is used to tag cloud servers to indicate their status.

[0025] The cloud server data allocation module is used to analyze cloud servers with fluctuation tags and redistribute the data in the cloud servers with fluctuation tags.

[0026] The cloud server tag update module is used to determine the status of the cloud server again after resource allocation is completed, and update the cloud server tag accordingly.

[0027] Furthermore, the cloud server data acquisition module, cloud server data processing module, cloud server status judgment module, cloud server tag setting module, cloud server data allocation module, and cloud server tag update module are connected to the system hard drive via network cable. After one acquisition cycle, the real-time updated data is backed up to the system hard drive via network cable.

[0028] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: On the one hand, by collecting key parameters such as the amount of stored data, the number of user connections, and the packet loss rate of cloud servers in real time, dynamic status tags are formed through data analysis. When abnormal server load is detected, the system automatically triggers a data redistribution mechanism to migrate overloaded data to adjacent servers with stable loads, avoiding performance degradation or service interruption caused by resource congestion on a single node. This data-driven intelligent scheduling effectively balances the resource utilization of each server in the cluster, reduces resource waste caused by delayed manual intervention or biased judgment, and enables the cloud storage system to maintain stable and efficient operation even under high-concurrency access or sudden data growth scenarios, significantly improving the overall service quality.

[0029] On the one hand, it accurately identifies devices in idle, stable, and fluctuating states. For servers in fluctuating states, it promptly initiates data migration processes to avoid hardware wear and tear caused by continuous high loads. This automated management mechanism not only reduces the time cost of manual inspection and fault handling but also achieves energy conservation and consumption reduction through refined resource scheduling. It is particularly suitable for the long-term operation of large-scale data centers, simultaneously optimizing both hardware maintenance and energy consumption to improve the economy and sustainability of cloud storage services.

[0030] On the other hand, by aiding in fluctuation coefficient calculation, this method can proactively identify potential performance bottlenecks or faults, transforming passive response into proactive intervention. When the server's storage over-limit rate or user connection fluctuation exceeds the set standard, the system immediately initiates resource reallocation, completing optimization adjustments before the problem escalates into a failure. Simultaneously, through periodic status reviews and tag updates, a closed-loop management mechanism is formed to continuously track optimization effects and ensure the server remains within a reasonable load range. This proactive prevention and control model effectively reduces uncertainty in system operation, enhances the fault tolerance of the cloud storage environment, and provides a solid guarantee for the secure storage and reliable access of user data. Attached Figure Description

[0031] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0032] Figure 1 This is a structural diagram of a cloud storage optimization management system based on data analysis according to the present invention;

[0033] Figure 2 This is a flowchart of a cloud storage optimization management method based on data analysis according to the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Please see Figure 1 and Figure 2 This invention provides a technical solution: a cloud storage optimization management method based on data analysis, comprising the following steps:

[0036] S1. Monitor the status of the cloud server in real time and collect data from the cloud server in real time;

[0037] S2. Analyze any cloud server and process the data on any cloud server;

[0038] S3. After data processing, determine the status of the cloud server based on the processed data;

[0039] S4. Tag the cloud server to indicate its status;

[0040] S5. Analyze the cloud servers with fluctuation tags and redistribute the data in the cloud servers with fluctuation tags.

[0041] S6. After completing resource allocation, reassess the status of the cloud server and update the cloud server's label.

[0042] In step S1, any cloud server is analyzed, and its status is monitored in real time. The maximum storage capacity of the cloud server is β. When the cloud server is powered off, it is marked as an idle cloud server. When the cloud server is powered on, real-time data is collected from it, with a collection interval of α. Data is collected I times within one collection cycle, including the cloud server's stored data volume, packet loss rate, and number of adjacent users. The set of stored data volume is {A1, A2, ..., A...}. i ,…,A I The packet loss rate is R, and the set of adjacent user numbers is {B1, B2, ..., B}. i ,…,B I}, where the number of adjacent users represents the number of users interacting with the cloud server, and A i B represents the amount of data stored on the cloud server during the i-th data collection. i This represents the number of adjacent users of the cloud server at the time of the i-th data collection. By monitoring and accurately marking the cloud server status in real time, it clearly distinguishes between idle and running servers, providing an intuitive basis for the rational allocation of resources, avoiding waste of idle resources, and improving overall utilization efficiency. For running servers, key information such as stored data volume, packet loss rate, and the number of adjacent users are collected periodically, enabling real-time monitoring of server load, data storage status, and network interaction quality. By continuously tracking this dynamic data, administrators can promptly detect abnormal fluctuations in server operation, including potential problems such as storage approaching saturation, unstable network connections, or a sudden increase in user access. They can then take proactive measures such as capacity expansion, network configuration optimization, or load balancing to effectively ensure the stability and reliability of cloud services, providing users with a smoother interactive experience, while reducing the risk of system failures and achieving refined and dynamic management of cloud servers.

[0043] In step S2, an analysis is performed on any cloud server. Within any collection period, the average amount of stored data is A0. During the i-th data collection period, the cloud server's storage over-limit index is W. Ai W Ai =A i / (W I *β), W I For the predetermined optimal load percentage of the cloud server, W I *β represents the optimal load data volume for the cloud server, when W AiWhen the value is greater than 1, increment the storage overrun number by one, and substitute each i = 1, 2, ..., I to obtain the storage overrun number of the cloud server as w. A This leads to the cloud server's storage over-limit rate being W. A =w A / I; Calculate the fluctuation level W of the number of user connections within any given collection period. B W B =σ / (B MAX -B MIN ), where σ is the set {B1, B2, ..., B}. i ,…,B I The standard deviation of B MAX For the set {B1, B2, ..., B... i ,…,B I The maximum value in}, B MIN For the set {B1, B2, ..., B... i ,…,B I The minimum value in the data storage and user connection status of cloud servers is used to accurately assess the server's operational status at the data storage and user interaction levels. For stored data, by comparing the actual storage volume with the optimal load level, it determines whether the storage is within a reasonable range, promptly identifies abnormal situations where storage continuously exceeds limits, and provides a basis for decisions such as data migration and capacity expansion, avoiding service performance impacts due to storage overload. For user connection counts, by measuring their fluctuations, the stability and balance of user access can be intuitively reflected, helping administrators identify abnormal fluctuations such as sudden increases or decreases in user access, and proactively optimize resource allocation or adjust service strategies to ensure the server can smoothly handle user access demands of different scales. The combination of these two methods achieves a multi-dimensional assessment of server operational status, helping administrators accurately locate potential problems, improve resource utilization efficiency, and ensure the stability and reliability of cloud services and continuous optimization of user experience.

[0044] In step S3, an analysis is performed on any cloud server, and the fluctuation coefficient D of the cloud server is calculated, D = (K A *W A / W A0 +K B *W B )*(1+R), where W A0 K represents the average storage over-limit rate of cloud servers that are currently powered on. A W A / W A0 The weighting of the impact of cloud server volatility coefficient, K B This represents the weight of the impact of the stability of user connection counts on the volatility coefficient of the cloud server, typically K. A and K BIt can be directly set to 1, setting the fluctuation threshold D0 = (K A +K B The algorithm calculates E*(1+R)*, where D < D0 and R < R0, indicating the cloud server is in a stable state; otherwise, it is in a fluctuating state. E is a pre-defined fluctuation tolerance coefficient, which can be set to 1.3. By comprehensively considering the cloud server's storage pressure, user connection fluctuations, and network quality, the stability of server operation can be comprehensively and accurately assessed. By integrating key indicators such as storage exceeding limits, user access fluctuations, and packet loss rates, potential risks in data storage, user interaction, and network transmission can be effectively identified. When all indicators are within a reasonable range, the server's stable state can be confirmed in a timely manner, ensuring continuous service operation. When abnormal storage pressure, sudden changes in user access, or unstable network connections occur, a fluctuating state can be quickly identified, prompting administrators to intervene in advance and mitigate risks by adjusting resource allocation, optimizing network configuration, or increasing storage capacity. This dynamic and comprehensive assessment method achieves precise control over the server's operating status, helps improve resource utilization efficiency, reduces the probability of system failure, ensures the continuous and stable operation of cloud services, and provides users with a more reliable service experience.

[0045] In step S4, the cloud servers in the power-on state are analyzed. Cloud servers in a stable state are labeled with a stable tag, and cloud servers in a fluctuating state are labeled with a fluctuating tag. The stable tag indicates that the cloud server can be used for resource allocation, and the fluctuating tag indicates that the cloud server needs to be allocated resources.

[0046] In step S5, the cloud server with the fluctuation label is analyzed, and the neighboring cloud servers are counted. Each neighboring cloud server is connected to any cloud server with the fluctuation label. There are a total of M neighboring cloud servers with the stable label, and the set of neighboring cloud servers with the stable label is {Z1, Z2, ..., Z...}. m ,…,Z M}, where Z m This represents the m-th neighboring cloud server with a stable label;

[0047] Extract the amount of data to be allocated, F, from cloud server Z0 in reverse chronological order, and allocate resources based on the amount of data to be allocated, where F = A. MAX -W I *β, the amount of data F allocated to the m-th neighboring cloud server with a stable label. m :

[0048]

[0049] Among them W m_BThis represents the fluctuation level of the m-th adjacent cloud server with a stable label. By analyzing cloud servers with fluctuation labels and leveraging adjacent stable cloud servers, dynamic allocation of data resources can be achieved, effectively alleviating the operational pressure on fluctuating servers and improving the overall stability of the cloud service architecture. By identifying and utilizing adjacent stable servers as targets for data allocation, data exceeding reasonable load on fluctuating servers can be promptly transferred to nodes in good operating condition, avoiding performance degradation or service interruption due to excessive load on a single server. Differentiated data allocation based on the fluctuation level of stable servers ensures that resource allocation is more aligned with the actual carrying capacity of each server. While fully utilizing the idle resources of stable servers, it avoids placing additional burdens on them, achieving a balanced distribution of data storage and processing capabilities within the cluster. This resource scheduling mechanism based on the status of adjacent nodes can quickly respond to abnormal server fluctuations, reduce manual intervention through automated load shifting, enhance the system's self-healing capabilities, ensure that cloud services maintain overall stable operation even under localized pressure, optimize resource utilization efficiency, and reduce potential failure risks.

[0050] In step S6, after resource allocation is completed, if cloud server Z0 is in a stable state, its fluctuation tag is deleted and a stable tag is added. If cloud server Z0 is still in a fluctuating state, its fluctuation tag is saved, and resources are allocated to it again until it reaches a stable state, at which point a stable tag is added. By dynamically adjusting the cloud server's status tag and continuously allocating resources, a closed-loop service stability assurance mechanism is constructed. After data resource allocation is completed, the decision to retain or delete the fluctuation tag is based on the server's real-time status, enabling accurate identification and dynamic tracking of the server's operating status. For servers still in a fluctuating state, repeated resource allocation until they stabilize avoids the limitations of a single adjustment, ensuring that potential problems are thoroughly resolved and preventing systemic risks caused by accumulated local pressure.

[0051] A cloud storage optimization and management system based on data analysis, comprising: a cloud server data acquisition module, a cloud server data processing module, a cloud server status judgment module, a cloud server tag setting module, a cloud server data allocation module, and a cloud server tag update module;

[0052] The cloud server data acquisition module is used to monitor the status of the cloud server in real time and to collect data from the cloud server in real time.

[0053] The cloud server data processing module is used to analyze and process the data of any cloud server.

[0054] The cloud server status determination module is used to determine the status of the cloud server based on the processed data.

[0055] The cloud server tagging module is used to tag cloud servers to indicate their status;

[0056] The cloud server data allocation module is used to analyze cloud servers with fluctuation tags and redistribute the data in cloud servers with fluctuation tags.

[0057] The cloud server tag update module is used to reassess the status of the cloud server after resource allocation is completed and update the cloud server's tag accordingly.

[0058] The cloud server data acquisition module, cloud server data processing module, cloud server status judgment module, cloud server tag setting module, cloud server data allocation module, and cloud server tag update module are connected to the system hard drive via network cable. After one acquisition cycle, the real-time updated data is backed up to the system hard drive via network cable.

[0059] Example 1: Step S1: Real-time status monitoring and data collection. The operation and maintenance system continuously scans all cloud servers. Servers that are powered off are automatically marked as idle and added to the resource pool for later use. For servers that are powered on and running, a data collection program is started at fixed time intervals. For example, during the operation of a server, the system periodically collects information such as its stored data volume, network packet loss rate, and number of adjacent users. Within each collection cycle, multiple data records are formed into a set. For example, the stored data volume reflects the current data storage pressure, the number of adjacent users reflects the real-time interaction load, and the packet loss rate is used to evaluate the network transmission quality.

[0060] Step S2: Quantitatively analyze storage pressure and user fluctuations. Based on the collected server data, the system first analyzes the storage status: comparing the amount of stored data collected each time with a preset reasonable load level, and calculating the percentage of times the reasonable load is exceeded to form a storage overload rate, which directly reflects the server's pressure level at the data storage level. Simultaneously, it calculates the fluctuations in the number of user connections: by analyzing the dispersion of the number of users, it determines whether user access is stable. If the number of users fluctuates frequently and significantly, it indicates that the server is facing an unstable challenge in terms of interactive load.

[0061] Steps S3-S4: Comprehensively assess the status and label it. The system comprehensively considers indicators such as storage over-limit rate, user fluctuation level, and packet loss rate to construct a multi-dimensional fluctuation coefficient model. Through preset weight allocation, factors such as storage pressure, user interaction stability, and network quality are quantified and integrated to calculate the overall fluctuation coefficient of the server. If the fluctuation coefficient is lower than the preset threshold and the packet loss rate is within the normal range, the server is determined to be in a stable state and marked as a node that can accept resource allocation; if it exceeds the threshold or the network quality is abnormal, it is marked as a fluctuating state and included in the optimization queue. If the server's storage is close to its limit due to recent business growth and a sudden increase in user access causes fluctuations, it is marked as a fluctuating state, triggering the resource allocation process.

[0062] Steps S5-S6: Dynamic resource allocation and closed-loop optimization. The system identifies adjacent nodes of the server and selects servers marked as stable as resource allocation targets. Based on the amount of data to be transferred from the fluctuating server and the load fluctuation of adjacent stable servers, data resources are allocated differentially: more data is allocated to adjacent servers with more stable operation and lower load to balance cluster pressure. After the initial allocation, the system re-evaluates the status: if storage pressure decreases, user fluctuations stabilize, and network quality meets standards, the fluctuating label is removed and updated to a stable label; if anomalies still exist, the allocation strategy is continuously adjusted until stability is restored.

[0063] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A cloud storage optimization management method based on data analysis, characterized in that: The method includes the following steps: S1. Monitor the status of the cloud server in real time and collect data from the cloud server in real time; S2. Analyze any cloud server and process the data on any cloud server; S3. After data processing, determine the status of the cloud server based on the processed data; S4. Tag the cloud server to indicate its status; S5. Analyze the cloud servers with fluctuation tags and redistribute the data in the cloud servers with fluctuation tags. S6. After completing resource allocation, check the status of the cloud server again and update the cloud server's label. In step S3, an analysis is performed on any cloud server to calculate its volatility coefficient D. W A0 K represents the average storage over-limit rate of cloud servers that are currently powered on. A W A / W A0 The weighting of the impact of cloud server volatility coefficient, K B This represents the weighting of the impact of the stability of user connection counts on the cloud server's volatility coefficient, and sets a volatility threshold. When D < D0 and R < R0, the cloud server is determined to be in a stable state; otherwise, the cloud server is determined to be in a fluctuating state. E is a pre-defined fluctuation tolerance coefficient; where W... A W represents the storage over-limit rate of the cloud server. B This indicates the degree of fluctuation in the number of user connections, where R represents the packet loss rate and R0 represents the preset packet loss rate threshold.

2. The cloud storage optimization management method based on data analysis according to claim 1, characterized in that: In step S1, any cloud server is analyzed and its status is monitored in real time. The maximum storage capacity of the cloud server is β. When the cloud server is in a powered-off state, it is marked as an idle cloud server. When the cloud server is powered on, real-time data is collected from it. The collection interval is set to α, and data is collected I times within one collection cycle. This includes the amount of data stored on the cloud server, the packet loss rate, and the number of adjacent users. The set of stored data is {A1, A2, ..., A...}. i ,…,A I The packet loss rate is R, and the set of adjacent user numbers is {B1, B2, ..., B}. i ,…,B I }, where the number of adjacent users represents the number of users interacting with the cloud server, and A i B represents the amount of data stored on the cloud server during the i-th data collection. i This represents the number of adjacent users of the cloud server during the i-th data collection.

3. The cloud storage optimization management method based on data analysis according to claim 2, characterized in that: In step S2, an analysis is performed on any cloud server. Within any collection period, the average amount of stored data is A0. During the i-th data collection period, the cloud server's storage over-limit index is W. Ai , The W I The optimal load percentage for the cloud server is set. This indicates the optimal load data volume for a cloud server, when W Ai When the value is greater than 1, increment the storage overrun number by one, and substitute each i=1,2,…,I to obtain the storage overrun number of the cloud server as w. A This leads to the cloud server's storage over-limit rate being W. A =w A / I; Calculate the fluctuation level W of the number of user connections within any given collection period. B W B =σ / (B MAX -B MIN ), where σ is the set {B1, B2, ..., B}. i ,…,B I The standard deviation of B MAX For the set {B1, B2, ..., B... i ,…,B I The maximum value in}, B MIN For the set {B1, B2, ..., B... i ,…,B I The minimum value in}.

4. The cloud storage optimization management method based on data analysis according to claim 3, characterized in that: In step S4, the cloud servers in the power-on state are analyzed, and cloud servers in a stable state are labeled with a stable tag, while cloud servers in a fluctuating state are labeled with a fluctuating tag. The stable tag indicates that the cloud server can be used for resource allocation, and the fluctuating tag indicates that the cloud server needs to be allocated resources.

5. The cloud storage optimization management method based on data analysis according to claim 4, characterized in that: In step S5, the cloud server with the fluctuation tag is analyzed, and the neighboring cloud servers of the cloud server are counted. The neighboring cloud servers are connected to any cloud server with the fluctuation tag. There are a total of M neighboring cloud servers with the stable tag, and the set of neighboring cloud servers with the stable tag is {Z1, Z2, ..., Z...}. m ,…,Z M }, where Z m This represents the m-th neighboring cloud server with a stable label.

6. The cloud storage optimization management method based on data analysis according to claim 5, characterized in that: Extract the amount of data to be allocated for F from cloud server Z0 in reverse chronological order, and then allocate the amount of data to be allocated for F using resources. Allocate data amount F to the m-th neighboring cloud server with a stable label. m : ； Among them W m_B This represents the fluctuation level of the m-th neighboring cloud server with a stable label.

7. The cloud storage optimization management method based on data analysis according to claim 6, characterized in that: In step S6, after resource allocation is completed, if cloud server Z0 is in a stable state, the fluctuation tag of cloud server Z0 is deleted and a stable tag is added; if cloud server Z0 is still in a fluctuating state, the fluctuation tag of cloud server Z0 is saved, and resources are allocated to cloud server Z0 again until cloud server Z0 is in a stable state, and a stable tag is added to cloud server Z0.

8. A cloud storage optimization management system based on data analysis, wherein the system is applied to the cloud storage optimization management method based on data analysis as described in any one of claims 1-7, characterized in that: The system includes: a cloud server data acquisition module, a cloud server data processing module, a cloud server status judgment module, a cloud server tag setting module, a cloud server data allocation module, and a cloud server tag update module; The cloud server data acquisition module is used to monitor the status of the cloud server in real time and to collect data from the cloud server in real time. The cloud server data processing module is used to analyze any cloud server and process the data of any cloud server. The cloud server status judgment module is used to determine the status of the cloud server based on the processed data after data processing. The cloud server tagging module is used to tag cloud servers to indicate their status. The cloud server data allocation module is used to analyze cloud servers with fluctuation tags and redistribute the data in the cloud servers with fluctuation tags. The cloud server tag update module is used to determine the status of the cloud server again after resource allocation is completed, and update the cloud server tag accordingly.

9. A cloud storage optimization management system based on data analysis according to claim 8, characterized in that: The cloud server data acquisition module, cloud server data processing module, cloud server status judgment module, cloud server tag setting module, cloud server data allocation module, and cloud server tag update module are connected to the system hard drive via a network cable. After one acquisition cycle, the real-time updated data is backed up to the system hard drive via the network cable.

Images