Modularized computing power and storage integrated server

Abstract

The invention relates to the field of computer servers, and discloses a modularized computing power and storage integrated server which comprises a computing module, a storage module, a bus module, a control module and an optimization module. The calculation module and the storage module support modular hot plug through a standardized interface, and can dynamically adjust configuration according to task requirements; the bus module is used for realizing high-speed data transmission between the calculation module and the storage module and supporting dynamic allocation of bandwidth priority; the control module generates a resource extension suggestion and a bandwidth allocation strategy based on fuzzy reasoning; and the optimization module performs offline optimization on the fuzzy rule base and membership function parameters of the control module through a genetic algorithm so as to improve the overall performance of system regulation and control. According to the method, elastic expansion and dynamic management of computing and storage resources are achieved, the resource utilization rate, the bandwidth allocation efficiency and the system operation stability are improved, and the method is suitable for high-performance computing and large-scale storage tasks with complex loads and multi-scene requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of computer servers, and in particular to a modular computing power and storage integrated server. Background Art

[0002] With the rapid development of big data and artificial intelligence technologies, the demand for computing power and storage resources in data centers and high-performance computing fields is increasing. However, existing server architectures generally have problems such as high coupling between computing resources and storage resources and lack of flexibility in resource allocation. Traditional servers usually adopt a fixed integration design, that is, the computing unit and storage unit are directly integrated on the same motherboard. Although this tightly coupled design can meet the needs under a single task load, it is easy to lead to inefficient use of resources in dynamic load changes or complex task scenarios.

[0003] Specifically, it is difficult for traditional server architectures to elastically expand resources under the load of computing-intensive or storage-intensive tasks. When computing tasks increase, storage units may be redundant, but computing units cannot be expanded in time; conversely, in storage-intensive tasks, computing units are idle, but storage units cannot meet demand, resulting in a waste of resources. In addition, the server expansion capabilities in the prior art are often limited by hardware design, and resource expansion can only be achieved by upgrading the server hardware as a whole, which not only increases operation and maintenance costs, but also affects the flexibility of the system.

[0004] At the same time, the data transmission bandwidth allocation of traditional servers lacks dynamism. In the collaborative work of computing units and storage units, the fixed bandwidth allocation method often cannot adapt to the real-time changes in task loads. For example, in storage-intensive tasks, since bandwidth resources are evenly distributed, the storage unit may cause performance bottlenecks due to insufficient bandwidth, while the bandwidth resources of the computing unit may be idle. In addition, existing bandwidth control technologies are usually not targeted and cannot prioritize bandwidth resources according to different task requirements, thereby reducing the overall system operation efficiency.

[0005] In addition, most resource expansion and allocation strategies in existing technologies use static rules, which cannot adjust resource configuration according to real-time load and are difficult to adapt to dynamic needs in complex scenarios. The optimization of the rule base also mainly relies on manual experience and lacks intelligent global optimization capabilities, which makes the expansion and control strategies perform poorly in complex and changeable operating scenarios, further limiting the improvement of system performance.

[0006] Based on the above problems, it is difficult for existing technologies to achieve dynamic management and efficient utilization of resources in complex task scenarios. There is an urgent need for a modular, scalable, and intelligent server design and resource control method to meet the current diversified task requirements and improve the system's resource utilization and operating efficiency. Summary of the invention

[0007] In view of the deficiencies in the prior art, the present invention provides a modular computing power and storage integrated server, which solves the problems of tight coupling of computing resources and storage resources, inflexible expansion, and unreasonable allocation of data transmission bandwidth in existing servers.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: a modular computing power storage integrated server, including a computing module, a storage module, a bus module, a control module and an optimization module; The computing module is used to provide high-performance computing capabilities and supports modular hot plugging; The storage module is used to provide storage resources and supports modular hot plugging; The bus module is used to realize high-speed data transmission between the computing module and the storage module, and supports dynamic allocation of bandwidth priority; The control module collects the operating status of the computing module, the storage module and the bus module in real time, and generates resource expansion decisions and bandwidth allocation strategies based on fuzzy reasoning; The optimization module optimizes the fuzzy reasoning rules and membership function parameters of the control module through a genetic algorithm to achieve global optimal control of system performance.

[0009] Preferably, the control module includes: a data acquisition unit to monitor computational load, storage utilization, I / O throughput, and bus bandwidth usage; The fuzzy reasoning unit is used to reason about the monitoring data based on the fuzzy rule base and generate resource expansion decisions and bandwidth allocation strategies.

[0010] Preferably, the fuzzy reasoning unit is based on the following input variables: Calculate the load, which represents the ratio of the current CPU usage to the total capacity; Storage utilization, which represents the ratio of used storage capacity to total storage capacity; I / O throughput, which represents the ratio of the current I / O operation rate to the maximum supported rate; Bus bandwidth requirement, which indicates the ratio of the current bus bandwidth to the maximum bus bandwidth.

[0011] Preferably, the fuzzy rule base of the fuzzy reasoning unit includes the following rules: When the computing load is high and the storage utilization is high, an expansion recommendation for adding computing modules and storage modules is generated; When the computing load is moderate and the storage utilization is moderate, an expansion recommendation is generated to maintain the existing module configuration; When the computing load is low and the storage utilization is low, a scaling recommendation is generated to reduce computing modules or storage modules.

[0012] Preferably, the optimization module optimizes the fuzzy reasoning rule base and parameter settings of the control module through a genetic algorithm, and the optimization process includes: The initial population is randomly generated, and each individual represents a combination of the fuzzy rule base and membership function parameters; The performance of individuals in the population is evaluated through a fitness function, which is based on resource utilization, bandwidth allocation efficiency, and extended response time; The population is optimized through selection, crossover and mutation operations to generate an optimized fuzzy rule base and parameter settings.

[0013] Preferably, the fitness function includes the following performance indicators: Resource utilization, which represents the ratio of actual resource usage to the total amount of allocated resources; Bandwidth allocation efficiency, which represents the ratio of effective transmission bandwidth to total allocated bandwidth; The extended response time indicates the delay time from load change to the generation of extended instructions.

[0014] Preferably, the bus module is designed based on the PCIe or CXL protocol, supports high-speed data interaction between the computing module and the storage module, and dynamically adjusts the bandwidth priority through the control module.

[0015] Preferably, the extension suggestion generated by the control module includes: Dynamically increase computing modules to cope with peak computing demands; Dynamically increase storage modules to meet data-intensive tasks; Dynamically reduce idle modules to reduce resource energy consumption.

[0016] Preferably, the optimization module optimizes the fuzzy inference rule base by offline operation, and the optimized rule base adapts to different computing and storage requirements in real time under complex load scenarios, ensuring that the resource allocation of the server is close to the global optimum.

[0017] The present invention also provides a method for resource expansion and bandwidth allocation based on the modular computing power and storage integrated server, comprising the following steps: The control module collects the operation data of the computing module, storage module and bus module in real time, including computing load, storage utilization and bandwidth requirements; Based on the collected operation data, the fuzzy reasoning unit is used to generate resource expansion suggestions and bandwidth allocation strategies; Adjust the configuration of the computing module or storage module according to the resource expansion suggestions and initialize the new module; The bus module implements the bandwidth priority allocation strategy to optimize the data flow between the computing module and the storage module. The fuzzy rule base and parameters are optimized offline through the optimization module, and the reasoning logic of the control module is updated to improve the accuracy of future decisions.

[0018] The present invention provides a modular computing power storage integrated server. It has the following beneficial effects: (1) The present invention dynamically adjusts the number and configuration of computing modules and storage modules through real-time monitoring and fuzzy reasoning of the control module, thus avoiding the waste caused by the fixed allocation of computing resources and storage resources in traditional servers. By expanding and recycling resources on demand, the present invention can efficiently utilize existing computing and storage resources in both high-load and low-load scenarios.

[0019] (2) The present invention uses the dynamic bandwidth allocation mechanism of the bus module, combined with the bandwidth allocation strategy generated by the control module, to adjust the bandwidth priority in real time according to the computationally intensive or storage intensive characteristics of different tasks, thereby reducing data transmission bottlenecks. This mechanism effectively improves the transmission efficiency and overall performance of the system under complex load scenarios.

[0020] (3) The present invention adopts a modular hot-swappable design. The computing module and the storage module are connected through a standardized interface, and the modules can be quickly added or removed according to demand during operation. The offline control of the optimization module further improves the system's adaptability to complex and variable loads, making the server have good scalability and scenario adaptability.

[0021] (4) The present invention optimizes the fuzzy rule base and membership function parameters of the control module through the genetic algorithm of the optimization module, so that the system can improve the dynamic decision logic according to the historical operation data. The optimized rule base significantly improves the accuracy and efficiency of expansion decision-making and bandwidth allocation, thereby ensuring the global performance of the system in long-term operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic diagram of the server architecture of the present invention; Figure 2 The figure is a flow chart of the resource expansion and bandwidth allocation method of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Please see attached Figure 1 The present invention provides a modular computing power and storage integrated server, which realizes dynamic expansion and global optimization of computing power and storage resources through modular hardware design and intelligent software control. Specifically, the server includes a computing module, a storage module, a bus module, a control module and an optimization module. Through the modular hot-swap mechanism and the coordinated work of the control module and the optimization module, efficient utilization and dynamic regulation of resources are achieved. The various modules of the modular computing power and storage integrated server of the present invention are described in detail below.

[0025] Computing Module In this embodiment, the computing module is an important component of the modular computing power storage integrated server, and is mainly used to provide high-performance computing capabilities. The computing module is connected to other modules (such as bus modules) through standardized interfaces and supports modular hot-swappable design, so that the computing power can be dynamically adjusted according to the operating status and computing needs of the server.

[0026] As an option, the core components of the computing module may include a high-performance processor (such as a CPU, GPU, or TPU) and a cache. The processor is responsible for executing complex computing tasks, while the cache is used to accelerate temporary access to data, thereby improving overall computing efficiency. It should be noted that the number and specific configuration of the computing modules can be adjusted according to actual business needs.

[0027] Specifically, in this embodiment, the computing module is connected to the bus module through a standardized physical interface (such as PCIe or CXL). When the module is inserted into the server, the bus module can identify the module status and add it to the computing resource pool of the server through an initialization process.

[0028] In this embodiment, the operating status of the computing module is collected and monitored in real time by the control module, specifically including: Defining and Monitoring Computational Load: Computational Load L The ratio of the actual resource usage of the current computing module to its total computing capacity can be expressed as:

[0029] in, is the current processor usage, The total processing power of the computing module (such as the maximum frequency or number of cores of the CPU). ,The control module can determine whether the utilization of the computing module meets the current demand.

[0030] Hot-swap support and status feedback: When a computing module is inserted or removed, the bus module notifies the control module through a status signal. As an implementation method, when a computing module is inserted, the bus module detects the connection status through the interface and sends an insertion event to the control module; the control module then completes the module initialization, including module resource registration and computing task scheduling.

[0031] For example, in some operating scenarios, such as when L >0.8 (i.e., CPU usage exceeds 80%), the control module may generate an expansion suggestion to add a new computing module to balance the system load. L When <0.2 (that is, the CPU usage is less than 20%), the system may suggest removing some idle computing modules to reduce power consumption.

[0032] As a possible implementation method, resource expansion of the computing module can be achieved through the following process: First, after the computing module is inserted, the bus module recognizes the new module and sends an event to the control module; Secondly, the control module registers the new module to the resource pool and allocates appropriate tasks or computing loads according to the current state of the system; Finally, the control module monitors the working status of the new module in real time to ensure that it can operate efficiently after joining the system.

[0033] It should be noted that the tasks supported by the computing module can be diverse, including but not limited to: General computing tasks: such as integer operations or floating-point operations performed by the CPU; Parallel computing tasks: such as large-scale matrix operations or deep learning inference tasks processed by GPUs or TPUs.

[0034] In some embodiments, in order to ensure the operating efficiency of the computing module, an independent power supply and cooling system may be designed for each module to ensure that the module does not overheat under high load or that performance degrades due to insufficient power supply.

[0035] In one possible implementation, the computing module can also support workload isolation. Specifically, the module can divide different computing resource pools according to task types (such as high-priority tasks and low-priority tasks), thereby preventing low-priority tasks from interfering with critical tasks.

[0036] In this embodiment, the computing module also has fault detection and recovery functions. For example, when a hardware fault occurs in the module, the bus module can detect the abnormal signal in time and notify the control module to perform isolation operations, thereby ensuring the stability of the overall operation of the server.

[0037] It should be further explained that the design of the computing module adopts a modular concept, which enables users to dynamically adjust the computing power of the server according to specific needs. For example, in data analysis or deep learning scenarios, more computing modules can be added to meet high-load computing needs; while in low-load scenarios (such as non-peak task processing), some computing modules can be removed to save energy.

[0038] In this embodiment, the implementation of the computing module ensures high flexibility and scalability of the server while meeting diverse business needs. As an option, the standardized interface and modular design of the computing module also facilitate subsequent hardware upgrades or maintenance, extending the life cycle of the system.

[0039] Storage Module In this embodiment, the storage module is an important component of the modular computing power storage integrated server, and is mainly used to provide the storage capacity of the server. The storage module supports modular hot-swappable design, and is connected to other modules of the server through standardized interfaces to achieve dynamic adjustment of data storage resources to meet the storage requirements in different task scenarios.

[0040] As an option, the main hardware of the storage module includes a high-speed SSD (solid state drive) or a large-capacity HDD (mechanical hard disk). Storage modules can be configured as high-performance (such as NVMe SSD) or large-capacity (such as SATA HDD) according to different application scenarios. It should be noted that different types of storage modules can be interchanged and combined through a unified physical interface (such as PCIe or SAS interface).

[0041] Specifically, in this embodiment, the operating status of the storage module is collected and monitored in real time by the control module, and the collected content includes but is not limited to storage utilization, data read and write rate, and module operating status. S It is the core monitoring indicator of the storage module, which is used to indicate the proportion of currently used storage resources to the total storage capacity. It can be expressed as:

[0042] in, Indicates the current used storage capacity. Indicates the total capacity of the storage module. S ,The control module can determine the occupancy of storage resources and generate ,expansion or reduction suggestions.

[0043] In one possible implementation, when the storage utilization S >0.7 (i.e., storage occupancy exceeds 70%), the control module may prompt to add new storage modules to meet the needs of storage-intensive tasks; SWhen <0.2 (i.e. storage occupancy is less than 20%), the system can prompt you to remove idle storage modules to optimize resource utilization.

[0044] It should be noted that during the insertion or removal of a storage module, the bus module is responsible for detecting the connection status of the storage module and sending events to the control module. For example, when a new storage module is inserted, the bus module detects the hardware connection signal and notifies the control module to perform the initialization operation of the module, including the registration of storage capacity and the configuration of storage paths.

[0045] Exemplarily, in some embodiments, the storage module can support partitioned storage and task isolation. Specifically, the storage module can allocate independent storage partitions according to the different priorities of the tasks. For example, high-priority tasks can give priority to high-speed storage modules, while low-priority tasks can be allocated to large-capacity storage modules. This partition mechanism can avoid resource competition between high-priority tasks and low-priority tasks.

[0046] In this embodiment, the storage module also supports fault detection and isolation functions. During the operation of the module, the bus module can monitor the operation status of the storage module through hardware signals. For example, when an abnormal I / O response of the storage module is detected, the bus module will feed back the information to the control module, and the control module will further stop the task allocation to the faulty module through isolation operation to ensure the stability of the overall storage system of the server.

[0047] As an option, the storage module can support different storage modes, such as cache mode and pass-through mode. In cache mode, the storage module can work with the computing module to store frequently used data in high-speed SSDs to improve access efficiency; in pass-through mode, the storage module directly responds to large-scale data read and write requirements, such as cold data archiving or data backup tasks.

[0048] In another possible implementation, the storage module supports the dynamic migration function of the storage load. Specifically, when the utilization rate of a storage module is too high, the control module can schedule data migration operations to distribute part of the storage load to other storage modules to reduce the pressure of a single module. This dynamic migration function can improve the balance and reliability of the overall storage system of the server.

[0049] It should be further explained that the design of the storage module adopts a modular concept, so that users can flexibly adjust the storage capacity according to actual needs. For example, when large-scale historical data needs to be processed, storage modules can be quickly added to expand storage capacity; in low-load scenarios, some storage modules can be removed to save energy.

[0050] In some embodiments, in order to improve the operating efficiency of the storage module, a data compression or encryption function can be integrated into the module. For example, the storage module can first perform real-time compression when receiving data to reduce the actual storage space occupied; in sensitive data scenarios, the storage module can perform hardware-level encryption on the data to improve data security.

[0051] It is understandable that the initialization process of the storage module has an important impact on the overall performance of the server. In this embodiment, when a new storage module is inserted, the control module first checks the hardware compatibility of the module, and then completes the allocation of storage paths and resource registration operations to ensure that the module can operate normally in a short time.

[0052] In this embodiment, the design and implementation of the storage module ensure that the storage resources of the server have high scalability and flexibility.

[0053] Bus module In this embodiment, the bus module is one of the core components of the modular computing power and storage integrated server, and is mainly used to achieve high-speed data transmission between the computing module and the storage module. The bus module is connected to the computing module and the storage module through a standardized interface that supports modular hot plugging, and provides the necessary channels for efficient data interaction.

[0054] As an option, the bus module is designed with high-performance interface protocols, such as PCIe (Peripheral Component Interconnect Express) or CXL (Compute Express Link). These protocols can provide high-bandwidth and low-latency communication capabilities for data transmission between modules, and ensure the reliability and consistency of data transmission when dynamically adjusting module configurations.

[0055] Specifically, in this embodiment, the functions of the bus module include data transmission management, dynamic bandwidth allocation and module status detection. By integrating these functions, the bus module realizes efficient regulation of the internal data flow of the server and supports the flexible combination of the computing module and the storage module.

[0056] In one possible implementation, the bus module's dynamic bandwidth allocation function is one of its core features. Bandwidth allocation is adjusted based on the current operating status and module requirements. It should be noted that dynamic bandwidth allocation is based on the strategy provided by the control module, and adjusts the data transmission priority and bandwidth share to meet the resource requirements of different modules in different scenarios.

[0057] For example, in data-intensive tasks, such as when the storage module needs to perform large-scale data backup, the bus module can prioritize allocating more bandwidth to the storage module; and in compute-intensive tasks, such as when the computing module needs to process complex high-performance calculations, the bus module will prioritize meeting the data transmission needs of the computing module.

[0058] It is understood that the bandwidth requirement allocation of the bus module can be expressed by the following formula:

[0059] in, Indicates the bandwidth allocated to a module. Represents the total available bandwidth, Indicates the bandwidth weight of the current module. By dynamically adjusting the weight , which can realize on-demand allocation of bandwidth resources.

[0060] As an implementation, the bus module further includes a module status detection unit. The unit can monitor the physical status of the computing module and the storage module connected to the bus module in real time. For example, when a module is connected to the bus module via a hot-swap interface, the status detection unit can capture the hardware connection signal and notify the control module of the event. The control module then completes the initialization configuration of the new module.

[0061] It should be noted that the module status detection of the bus module is not limited to the hardware connection status, but can also include the monitoring of the module operation status. For example, when the data transmission delay of a module is abnormal, the bus module can capture the abnormality through the operation status monitoring function and send an alarm signal to the control module. The control module takes further measures based on the alarm signal, such as reallocating bandwidth or isolating the faulty module.

[0062] In this embodiment, the bus module also has the function of error detection and recovery of data transmission. For example, when an error occurs during data transmission (such as data packet loss or verification failure), the bus module can find the problem through the error detection mechanism and automatically trigger the data retransmission operation to ensure the integrity of data transmission.

[0063] In some embodiments, the bus module supports multi-channel parallel transmission to further improve the efficiency of data transmission. Specifically, each computing module or storage module can be assigned an independent data channel to avoid data flow interference between different modules. This multi-channel transmission design is particularly suitable for high-performance computing and large-scale data processing scenarios.

[0064] In another possible implementation, the bus module supports dynamic bandwidth scheduling and priority management. For example, the bus module can set a higher transmission priority for critical tasks and allocate lower bandwidth resources for low-priority tasks. The priority management strategy is provided by the control module and implemented through the hardware scheduling mechanism of the bus module.

[0065] It should be further explained that the bus module is designed with a modular concept, and its standardized interface is compatible with different types of computing modules and storage modules. This design not only improves the flexibility of the server, but also facilitates the upgrade and maintenance of the module. For example, when higher transmission performance is required, the demand can be met by replacing a higher-specification bus module without changing the hardware configuration of other modules.

[0066] In this embodiment, the initialization process of the bus module is crucial. When the server is started or a new module is inserted, the bus module will first detect the type of connected module and complete the necessary initialization operations, including data path configuration and bandwidth resource registration. This initialization process ensures that the module can quickly integrate into the server's resource pool and participate in data interaction.

[0067] In summary, the bus module in this embodiment provides reliable communication guarantee for the modular computing power and storage integrated server by supporting high-speed data transmission, dynamic bandwidth allocation and module status detection.

[0068] Control Module In this embodiment, the control module is the core management unit of the modular computing power storage integrated server, which is responsible for coordinating the functional modules of the server, including the computing module, storage module and bus module. The main function of the control module is to collect the system operation status in real time, generate resource expansion decisions and bandwidth allocation strategies using fuzzy reasoning, and dynamically control the operation status of each module to achieve efficient operation of the system.

[0069] As an option, the functional structure of the control module may include the following sub-units: a data acquisition unit, a fuzzy reasoning unit, and an instruction issuing unit. The data acquisition unit is responsible for obtaining the server operation status in real time; the fuzzy reasoning unit performs logical analysis and processing on the collected data; and the instruction issuing unit is used to transmit the generated expansion decision and bandwidth strategy to the corresponding module.

[0070] Specifically, in this embodiment, the data acquisition unit monitors the operating status of the computing module and the storage module in real time through the communication interface with the bus module, including computing load, storage utilization and bus bandwidth usage.

[0071] As a possible implementation method, the control module can also monitor the bandwidth occupancy rate B of the bus module, that is, the ratio of the currently used bandwidth to the total bandwidth, which is expressed by the following formula:

[0072] in, The bandwidth used for the current bus, is the maximum bandwidth of the bus module.

[0073] In a possible implementation, the fuzzy reasoning unit generates resource expansion decisions and bandwidth allocation strategies based on the above-collected indicators and a preset fuzzy rule base. For example, the fuzzy reasoning unit can generate expansion suggestions based on the following rules: When the computing load is high and the storage utilization is moderate, a prompt is given to add computing modules; When storage utilization is high and bus bandwidth occupancy is low, a prompt is given to add storage modules; When the computational load is low and the storage utilization is low, it prompts to remove redundant modules to reduce power consumption.

[0074] Exemplarily, the input variables of the rules in the fuzzy rule base include the computational load , Storage Utilization and bus bandwidth requirements The fuzzy reasoning unit calculates the output membership by the minimum-maximum reasoning method:

[0075] The output results are clarified through weighted average method to generate expansion decisions and bandwidth allocation strategy , the formula is:

[0076] It should be noted that the rule base and membership function parameters of the fuzzy reasoning unit are generated by offline optimization of the optimization module. The optimization module optimizes the reasoning logic of the control module through genetic algorithms, so that it can generate more accurate expansion suggestions and bandwidth strategies in complex load scenarios.

[0077] In another possible implementation, the instruction issuing unit of the control module transmits the decision and strategy generated by the fuzzy reasoning unit to the computing module and the storage module, and dynamically adjusts the configuration status of the modules. For example, when prompted to add a computing module, the instruction issuing unit completes the registration operation of the new computing module through the bus module; when prompted to reduce the storage module, the instruction issuing unit notifies the storage module to stop the current task and complete the safe uninstallation.

[0078] It is understandable that the control module not only considers the current operating status in resource expansion decisions, but also predicts future resource requirements through fuzzy reasoning. For example, when the peak load is about to arrive, the control module can increase module resources in advance to avoid system performance limitations; when the load decreases significantly, the control module can release excess resources in advance.

[0079] As an implementation method, the control module also has a fault detection function. For example, when the status signal of the computing module or storage module is abnormal (such as too high response delay or data transmission interruption), the control module can identify the abnormality and generate an alarm signal to prompt the relevant modules to perform fault isolation or task migration.

[0080] It should be further explained that the design of the control module ensures that it can adapt to various types of hardware modules. Through standardized interface protocols (such as I²C or PCIe), the control module can be compatible with computing modules and storage modules from different manufacturers, thereby improving the hardware flexibility and scalability of the server.

[0081] In this embodiment, the control module realizes dynamic resource regulation and intelligent management of the modular computing power and storage integrated server through real-time monitoring, fuzzy reasoning and instruction issuance.

[0082] Optimization Module In this embodiment, the optimization module is an important component of the modular computing power and storage integrated server, and is mainly used to dynamically adjust the fuzzy reasoning rule base and membership function parameters of the control module through offline optimization, thereby improving the accuracy and adaptability of the control module in generating expansion decisions and bandwidth allocation strategies.

[0083] As an option, the optimization module uses genetic algorithms to optimize the fuzzy inference rule base and parameters. Genetic algorithms have global search capabilities, can effectively avoid local optimal problems, and are suitable for complex and variable load scenarios. The optimization results of the optimization module can be applied to the control module after periodic updates to ensure that the system's intelligent control capabilities are continuously improved.

[0084] Specifically, in this embodiment, the workflow of the optimization module includes four stages: data collection, fitness evaluation, genetic operation, and parameter update. The following are detailed descriptions: In a possible implementation, the optimization module first collects historical data during the operation of the server through interaction with the control module, including load fluctuations of the computing module, changes in the utilization of the storage module, and bandwidth occupancy of the bus module. These data serve as the basis of the optimization process and provide a reliable basis for subsequent fitness evaluation.

[0085] It should be noted that fitness evaluation is the core step of the optimization module. The optimization module calculates the fitness value based on the following formula :

[0086] in: It represents resource utilization, which is defined as the ratio of actual resource usage to total allocated resources; represents the scaling response time, which is the time from load change to scaling decision generation; It represents the bandwidth allocation efficiency, which is defined as the ratio of the effective allocated bandwidth to the total allocated bandwidth; , , It is a weight parameter, which is set according to the specific application scenario.

[0087] Exemplarily, the optimization module calculates the fitness of different combinations of fuzzy rule bases and membership function parameters, and selects the combination with the best performance according to the fitness value.

[0088] As an implementation method, genetic operations include three sub-steps: selection, crossover, and mutation. Selection is to select better individuals from the population based on fitness values ​​to generate the next generation. Crossover is to partially exchange two selected fuzzy rule bases or parameter sets to generate new rule combinations. Mutation is to randomly adjust some membership function parameters to increase the diversity of the population.

[0089] It should be further explained that the design of crossover and mutation operations ensures that the optimization module can explore new rule combinations to a certain extent while maintaining the inheritance of existing high fitness individuals. For example, for the fuzzy rule base of computing load, the crossover operation may combine the rule of "adding computing modules when the load is high" with the rule of "maintaining the existing configuration when the load is medium" to generate a new rule combination.

[0090] In another possible implementation, the optimization module will specifically optimize the membership function parameters. The optimization of the membership function parameters aims to adjust the fuzzy membership of the input variables to make them more consistent with the actual business scenario. For example, the membership function for calculating load may change from "low load" to "low load". Optimized for low load ”, making the inference results more responsive to rapid changes in load.

[0091] It is understandable that the optimization process of the optimization module is performed offline, which means that the optimization operation will not affect the real-time performance of the server. When the optimization process is completed, the optimization module updates the optimal fuzzy rule base and membership function parameters to the control module, and the control module applies these optimization results in subsequent real-time reasoning.

[0092] As an option, the optimization cycle of the optimization module can be adjusted according to the actual situation of system operation. For example, in the case of frequent load fluctuations or low system performance, the optimization cycle can be shortened to adapt to changes more quickly; in the scenario of stable load, the optimization cycle can be appropriately extended to reduce the resource usage of the optimization process.

[0093] In a possible implementation, the optimization module can also perform multi-objective optimization. For example, in some scenarios, resource utilization can be used as the main optimization goal, while in other scenarios, bandwidth allocation efficiency or extended response time may be more important. This multi-objective optimization is achieved by adjusting the weight parameters in the fitness function to meet the needs of different scenarios.

[0094] It should be further explained that the design of the optimization module adopts a modular concept and can run independently of other parts of the server. This design not only improves the flexibility of the optimization module, but also enables it to perform continuous optimization without affecting the normal operation of the server. For example, when the control module is performing real-time control tasks, the optimization module can use the idle resources of the server to complete the optimization task.

[0095] In this embodiment, the optimization module optimizes the fuzzy rule base and membership function parameters through a genetic algorithm, thereby achieving continuous improvement of the intelligent control capability of the control module.

[0096] In general, the present invention realizes elastic expansion and dynamic management of computing resources and storage resources through modular hot-swappable design and intelligent control methods. The computing module and storage module support standardized interface connection and can flexibly adjust the configuration according to load requirements; the bus module is responsible for realizing high-speed data transmission and dynamic bandwidth allocation; the control module generates resource expansion decisions and bandwidth allocation strategies based on fuzzy reasoning by real-time monitoring of system status; the optimization module optimizes the fuzzy rule base and membership function parameters offline through genetic algorithms to improve the global performance of system control. The overall architecture design of the present invention improves the resource utilization, response efficiency and operation stability of the server, is suitable for complex computing and large-scale storage scenarios, and has high practical value and flexibility.

[0097] Please see attached Figure 2 The present invention also provides a resource expansion and bandwidth allocation method based on a modular computing power and storage integrated server. The specific implementation method is described below.

[0098] In this embodiment, the core of the method is to achieve dynamic expansion of the computing module and the storage module and optimized bandwidth allocation of the bus module through the collaborative work of the control module and the optimization module, thereby meeting the resource requirements of the server in different scenarios.

[0099] S1. Collect the operation data of the computing module, storage module and bus module in real time through the control module; The operation data of the computing module, storage module and bus module are collected in real time through the control module. In this embodiment, the operation data includes but is not limited to computing load, storage utilization and bandwidth demand. The acquisition unit of the control module obtains these data through the communication interface with the bus module, and the data content and its definition are as described above. For example, the computing load can be represented by the ratio of the current computing module usage to the total computing capacity, and the storage utilization and bandwidth demand are defined in the same way as described above.

[0100] S2, based on the collected operation data, using the fuzzy reasoning unit to generate resource expansion suggestions and bandwidth allocation strategies; Based on the collected operation data, a fuzzy reasoning unit is used to generate resource expansion suggestions and bandwidth allocation strategies. In this embodiment, the fuzzy reasoning unit performs reasoning through the fuzzy rule base and input data to generate expansion suggestions (such as increasing or decreasing computing modules / storage modules) and bandwidth allocation strategies (such as preferentially allocating bandwidth resources of computing modules or storage modules). The reasoning process and rule base structure have been given in the control module part and will not be repeated here.

[0101] S3. Adjust the configuration of the computing module or storage module according to the resource expansion suggestion, and initialize the new module; Adjust the configuration of the computing module or storage module according to the resource expansion suggestion, and complete the initialization of the new module. In this embodiment, when the reasoning result of the control module recommends adding a new computing module or storage module, the system will detect the hot plug status of the physical module and complete the resource initialization. For example, when a new module is detected, the bus module will notify the control module to complete the registration and configuration of the module to ensure that it can join the existing resource pool and participate in task scheduling.

[0102] S4, implementing bandwidth priority allocation strategy through bus module to optimize data flow between computing module and storage module; The bandwidth priority allocation strategy is executed through the bus module to achieve data flow optimization between the computing module and the storage module. In this embodiment, the bus module receives the bandwidth allocation strategy generated by the control module and dynamically adjusts the bandwidth priority according to the strategy. For example, in storage-intensive tasks, the bus module will prioritize allocating more bandwidth to the storage module; in computing-intensive tasks, higher priority bandwidth resources will be allocated to the computing module. The specific bandwidth allocation method can refer to the description of the bus module part.

[0103] S5, perform offline optimization of the fuzzy rule base and parameters through the optimization module, and update the reasoning logic of the control module to improve the accuracy of future decisions; The optimization module optimizes the fuzzy rule base and membership function parameters offline and updates the reasoning logic of the control module. In this embodiment, the optimization module regularly improves the reasoning logic of the control module to improve its decision-making accuracy and adaptability. The optimization module analyzes historical operation data through a genetic algorithm, and the optimized rule base and parameters are loaded by the control module in the next operation cycle to guide future resource expansion and bandwidth allocation.

[0104] This embodiment realizes intelligent expansion and optimized management of server resources through the above method, which can effectively cope with changes in resource requirements in complex load scenarios and improve the flexibility and performance of the system.

[0105] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A modular computing power and storage integrated server, characterized in that: It includes a computing module, a storage module, a bus module, a control module and an optimization module; The computing module is used to provide high-performance computing capabilities and supports modular hot plugging; The storage module is used to provide storage resources and supports modular hot plugging; The bus module is used to realize high-speed data transmission between the computing module and the storage module, and supports dynamic allocation of bandwidth priority; The control module collects the operating status of the computing module, the storage module and the bus module in real time, and generates resource expansion decisions and bandwidth allocation strategies based on fuzzy reasoning; The optimization module optimizes the fuzzy reasoning rules and membership function parameters of the control module through a genetic algorithm to achieve global optimal control of system performance.

2. The modular computing power and storage integrated server according to claim 1, characterized in that: The control module comprises: a data acquisition unit to monitor computational load, storage utilization, I / O throughput, and bus bandwidth usage; The fuzzy reasoning unit is used to reason about the monitoring data based on the fuzzy rule base and generate resource expansion decisions and bandwidth allocation strategies.

3. The modular computing power and storage integrated server according to claim 2, characterized in that: The fuzzy inference unit is based on the following input variables: Calculate the load, which represents the ratio of the current CPU usage to the total capacity; Storage utilization, which represents the ratio of used storage capacity to total storage capacity; I / O throughput, which represents the ratio of the current I / O operation rate to the maximum supported rate; Bus bandwidth requirement, which indicates the ratio of the current bus bandwidth to the maximum bus bandwidth.

4. The modular computing power and storage integrated server according to claim 2, characterized in that: The fuzzy rule base of the fuzzy reasoning unit includes the following rules: When the computing load is high and the storage utilization is high, an expansion recommendation for adding computing modules and storage modules is generated; When the computing load is moderate and the storage utilization is moderate, an expansion recommendation is generated to maintain the existing module configuration; When the computing load is low and the storage utilization is low, a scaling recommendation is generated to reduce computing modules or storage modules.

5. The modular computing power and storage integrated server according to claim 1, characterized in that: The optimization module optimizes the fuzzy reasoning rule base and parameter settings of the control module through a genetic algorithm. The optimization process includes: The initial population is randomly generated, and each individual represents a combination of the fuzzy rule base and membership function parameters; The performance of individuals in the population is evaluated through a fitness function, which is based on resource utilization, bandwidth allocation efficiency, and extended response time; The population is optimized through selection, crossover and mutation operations to generate an optimized fuzzy rule base and parameter settings.

6. The modular computing power and storage integrated server according to claim 5, characterized in that: The fitness function includes the following performance indicators: Resource utilization, which represents the ratio of actual resource usage to the total amount of allocated resources; Bandwidth allocation efficiency, which represents the ratio of effective transmission bandwidth to total allocated bandwidth; The extended response time indicates the delay time from load change to the generation of extended instructions.

7. The modular computing power and storage integrated server according to claim 1, characterized in that: The bus module is designed based on the PCIe or CXL protocol, supports high-speed data interaction between the computing module and the storage module, and dynamically adjusts the bandwidth priority through the control module.

8. The modular computing power and storage integrated server according to claim 1, characterized in that: The expansion suggestions generated by the control module include: Dynamically increase computing modules to cope with peak computing demands; Dynamically increase storage modules to meet data-intensive tasks; Dynamically reduce idle modules to reduce resource energy consumption.

9. The modular computing power and storage integrated server according to claim 1, characterized in that: The optimization module optimizes the fuzzy inference rule base by running it offline. The optimized rule base adapts to different computing and storage requirements in real time under complex load scenarios, ensuring that the resource allocation of the server is close to the global optimum.

10. A method for resource expansion and bandwidth allocation based on the modular computing power and storage integrated server according to any one of claims 1 to 9, characterized in that: The following steps are involved: The control module collects the operation data of the computing module, storage module and bus module in real time, including computing load, storage utilization and bandwidth requirements; Based on the collected operation data, the fuzzy reasoning unit is used to generate resource expansion suggestions and bandwidth allocation strategies; Adjust the configuration of the computing module or storage module according to the resource expansion suggestions and initialize the new module; The bus module implements the bandwidth priority allocation strategy to optimize the data flow between the computing module and the storage module. The fuzzy rule base and parameters are optimized offline through the optimization module, and the reasoning logic of the control module is updated to improve the accuracy of future decisions.

Images