Distributed computing system based on fungal hypha network topology structure

Abstract

The invention discloses a distributed computing system based on a fungal hypha network topology structure. The distributed computing system comprises a topology management module which generates a dynamic connection signal by simulating biological behaviors of fungal hypha growth and fusion; the signal diffusion module receives the dynamic connection signal and generates a routing diffusion signal, and the routing diffusion signal asynchronously propagates a state and an instruction between nodes by simulating a chemical signal transduction mode in a hypha network; the resource scheduling module receives the route diffusion signal and generates a task scheduling signal; the node collaboration module receives the task scheduling signal and generates a collaboration computing signal, and the collaboration computing signal is used for coordinating a plurality of computing nodes to jointly execute a single distributed task; the topology management module also receives a node state feedback signal from the node cooperation module. According to the distributed computing system based on the fungal hypha network topology structure, the problem that a distributed system lacks self-adaption, high robustness and low-energy-consumption consensus can be solved.

Description

Technical Field

[0001] This invention relates to the fields of distributed computing and biomimetic computing technology, specifically to a distributed computing system based on fungal hyphal network topology. Background Technology

[0002] Traditional distributed computing systems, such as client-server architectures, heavily rely on a central node for management and scheduling, which exposes them to single points of failure and limits scalability. While subsequent peer-to-peer network technologies have enabled interconnection between nodes, their network topologies are often relatively static or have simple adjustment strategies, making it difficult to maintain high efficiency and stability in dynamic environments with frequent node joining and leaving. In recent years, blockchain technology has provided a decentralized and trusted collaboration paradigm, but its consensus mechanisms, such as proof-of-work, result in significant energy consumption and low transaction processing efficiency, failing to meet the demands of high-frequency applications. Existing resource allocation and task scheduling algorithms are mostly based on global information or complex predefined strategies. In large-scale, rapidly changing network environments, their computational overhead is enormous and their response is slow, making real-time and precise resource allocation impossible. While self-organizing networks such as wireless sensor networks possess some adaptive capabilities, their topology changes are primarily driven by physical location movement, lacking proactive and intelligent growth and optimization capabilities.

[0003] Meanwhile, in the field of biology, research on fungal hyphal networks has revealed their extraordinary characteristics: they are decentralized systems without central control, capable of autonomously constructing efficient and redundant mesh topologies through the growth and fusion of hyphal tips; they can transmit information within the network through chemical and electrical signals and collectively respond to resource-rich areas or threats; they possess extremely strong robustness, with local damage not affecting overall functionality and enabling rapid self-healing. These characteristics precisely address the core challenges faced by existing distributed computing systems. Therefore, applying the intelligent core mechanisms of fungal hyphal networks to the architecture design of distributed computing systems holds promise for pioneering a new generation of computing paradigms with biological-grade adaptability and resilience, effectively overcoming many inherent shortcomings of existing technologies. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of this invention is to provide a distributed computing system based on fungal hyphal network topology to solve the problems of distributed systems lacking adaptability, high robustness, and low-energy consensus. This invention abstracts and transforms the core biological mechanisms of fungal hyphal networks, such as growth, signal transmission, and resource allocation, into a complete technical solution. The system uses a topology management module to dynamically self-organize network connections, a signal diffusion module to achieve efficient asynchronous communication, and a resource scheduling module to biomimeticly guide task flow, ultimately forming a decentralized, environmentally adaptable, and fault-tolerant and self-healing distributed computing system.

[0005] This invention provides a distributed computing system based on fungal hyphal network topology, comprising:

[0006] The topology management module generates dynamic connection signals by simulating the biological behavior of fungal hyphae growth and fusion. These dynamic connection signals are used to indicate the establishment, enhancement, or dissolution of connections between computing nodes.

[0007] The signal diffusion module receives dynamic connection signals and generates routing diffusion signals. The routing diffusion signals asynchronously propagate status and instructions between nodes by simulating the chemical signal transmission in a mycelial network.

[0008] The resource scheduling module receives the routing diffusion signal and generates a task scheduling signal. The task scheduling signal directs the computing tasks to resource-rich nodes by simulating the mycelial nutrient transport mechanism.

[0009] The node coordination module receives task scheduling signals and generates coordination computing signals, which are used to coordinate multiple computing nodes to jointly execute a single distributed task.

[0010] The topology management module also receives node status feedback signals from the node coordination module and adjusts the dynamic connection signal generation strategy according to the signals, thereby forming a closed-loop adaptive network topology control system.

[0011] In one embodiment of the present invention, the topology management module generates dynamic connection signals in the following ways: simulating the vegetative growth behavior of mycelia, continuously receiving resource status broadcast signals from each computing node in the system, wherein the resource status broadcast signals include at least the CPU utilization, memory idle rate, and network bandwidth redundancy; the topology management module calculates the resource enrichment direction based on the resource status broadcast signals, and preferentially generates connection establishment signals pointing to high resource nodes, while generating connection weakening or interruption signals for resource overloaded nodes, thereby realizing the biomimetic growth and expansion of the computing network topology towards resource-rich areas.

[0012] In one embodiment of the present invention, the specific method by which the signal diffusion module generates the routing diffusion signal includes: simulating the multipath transmission mechanism of cytoplasmic flow and vesicle transport in a mycelial network, splitting the computing task or data packet to be distributed into multiple sub-signal units; each sub-signal unit is independently transmitted asynchronously along the network path defined by the dynamic connection signal, and carries the source node identifier and sequence information during the transmission process; finally, all sub-signal units are reassembled and restored at the target node according to the sequence information they carry. This method avoids the blocking risk of a single path in traditional networks and significantly improves the parallel efficiency and overall robustness of data transmission.

[0013] In one embodiment of the present invention, the resource scheduling module generates task scheduling signals in the following ways: simulating the stress avoidance response of a mycelial network when encountering obstacles or toxins, continuously monitoring the error rate and response delay signals of each computing node; when the error rate of a specific node exceeds a preset threshold, the resource scheduling module determines that the area is an unhealthy network area and generates a task avoidance signal; the task avoidance signal will guide new computing tasks to bypass the unhealthy node and its closely related neighboring nodes, and through the topology management module, trigger the process of opening alternative paths in the network, thereby ensuring the continuity and success rate of computing tasks.

[0014] In one embodiment of the present invention, the specific method by which the node collaboration module generates collaborative computing signals includes: simulating the swarm intelligence decision-making behavior of mycelial networks, decomposing a large computing task into multiple interconnected sub-task blocks; the collaborative computing signals are distributed to a group of computing nodes, the signals encoding the execution logic of each sub-task block and the necessary interaction protocols between nodes; each node collaborates by exchanging intermediate computing result signals during the execution of sub-tasks, and finally one node or a dedicated node is responsible for integrating all intermediate results and outputting the final result, thereby realizing decentralized collaborative problem solving.

[0015] In one embodiment of the present invention, the system further includes a network memory learning module, which connects the topology management module and the node collaboration module, and is used to receive and store historical dynamic connection signals and collaborative computing signals. The network memory learning module trains an internal decision-making model by analyzing the success and failure patterns of historical signals and generates network optimization prediction signals to be fed back to the topology management module. The topology management module then uses the prediction signals to pre-build and optimize the connection relationships that may be formed in the future, so that the entire system has the ability to learn from historical experience and continuously improve itself.

[0016] In one embodiment of the present invention, the sub-signal unit transmission mechanism of the signal diffusion module further includes a priority marking function; each sub-signal unit is assigned a priority mark when it is created, which is generated based on the urgency of the computing task to which it belongs and the importance of the data; during transmission, when network path congestion occurs, the sub-signal unit holding the high priority mark will obtain the priority forwarding right and occupy network bandwidth resources for transmission first, thereby ensuring that critical tasks can be delivered and processed with low latency and achieving differentiated service quality assurance.

[0017] In one embodiment of the present invention, the task avoidance mechanism of the resource scheduling module also includes a node recovery monitoring function; after generating a task avoidance signal and guiding the task to detour, the resource scheduling module will not permanently abandon the unhealthy node, but will continuously send lightweight probe task signals to the node at a low frequency and monitor its response; once the error rate and response latency of the node are detected to return to normal levels, the resource scheduling module generates a node recovery signal and gradually reschedules new computing tasks to the node, thereby realizing the full and flexible utilization of network resources.

[0018] In one embodiment of the present invention, the dynamic connection strategy of the topology management module includes a periodic connection optimization function; the module evaluates the current topology connection relationship of the entire network at fixed time intervals, and the evaluation basis includes connection utilization signals and path delay signals; for connections that are in a low utilization state for a long time, the topology management module generates a connection reclamation signal to release the connection and thus save resources; at the same time, for critical path areas with high load and high latency, a connection enhancement signal is generated to incentivize the establishment of new redundant connections around them to share the load, thereby realizing the periodic self-optimization of the network.

[0019] In one embodiment of the present invention, the collaborative computing signal of the node collaboration module also integrates a result verification mechanism. When the node responsible for integrating the final result completes the result calculation, it will broadcast the result summary generation verification request signal to other nodes that participated in the collaborative computing. After receiving the verification request, the participating nodes will use the intermediate calculation results stored locally to perform consistency verification on the final result and generate a verification result signal through a voting mechanism. Only when the verification result signal indicates that the consistency has passed will the final result be confirmed and output. Otherwise, the task recalculation process will be triggered to ensure the correctness and reliability of the distributed computing results.

[0020] This invention provides a distributed computing system based on fungal hyphal network topology. By simulating the core biological mechanisms of fungal hyphal networks, such as growth, signal transmission, and resource allocation, this invention abstracts and transforms them into a complete technical solution. The system achieves dynamic self-organizing network connections through a topology management module, efficient asynchronous communication through a signal diffusion module, and biomimetic task flow guidance through a resource scheduling module. Ultimately, it forms a decentralized, environmentally adaptable, and fault-tolerant distributed computing system with strong self-healing capabilities. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a system architecture diagram of a distributed computing system based on fungal hyphal network topology. Detailed Implementation

[0023] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0024] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0025] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0026] Please see Figure 1The diagram illustrates a distributed computing system based on a fungal hyphal network topology, comprising: a topology management module that generates dynamic connection signals by simulating the biological behavior of fungal hyphal growth and fusion, used to indicate the establishment, enhancement, or dissolution of connections between computing nodes; a signal diffusion module that receives the dynamic connection signals and generates routing diffusion signals, which asynchronously propagate states and instructions between nodes by simulating chemical signal transmission in the hyphal network; a resource scheduling module that receives the routing diffusion signals and generates task scheduling signals, which guide computing tasks to resource-rich nodes by simulating the nutrient transport mechanism of hyphae; a node collaboration module that receives the task scheduling signals and generates collaborative computing signals, used to coordinate multiple computing nodes to jointly execute a single distributed task; and a topology management module that also receives node status feedback signals from the node collaboration module and adjusts the generation strategy of the dynamic connection signals accordingly, thereby forming a closed-loop adaptive network topology control system.

[0027] Figure 1 As shown, its core feature lies in the construction of a highly biomimetic, decentralized, adaptive computing network. This system, through the synergistic action of multiple functional modules, transforms the biological intelligence of fungal hyphal networks into concrete technological implementations. First, the core of the system is the topology management module, which simulates the growth and fusion behavior of fungal hyphae in nature. Biologically, hyphal tips explore their surrounding environment, sensing the presence of nutrients, water, and other hyphal networks, thereby determining their growth direction and fusing with other hyphae to form a vast, interconnected network. In this system, this biological process is abstracted as the technological behavior of generating dynamic connection signals. This module continuously evaluates the current state of the computing network and proactively makes decisions, instructing when new connections need to be established between computing nodes, when the bandwidth or priority of existing connections needs to be increased, or, if necessary, when to dismantle certain non-critical or inefficient connections. This dynamic adjustment is not random but follows a biomimetic optimization algorithm, aiming to ensure that the entire network's topology always evolves in a more efficient and robust direction. This process is completely decentralized, with each node running a local instance of the module. Global coordination and consistency are achieved through the exchange of local information, thus avoiding the single point of failure and performance bottleneck problems of the central controller in traditional systems.

[0028] Furthermore, the system's operation relies on a signal diffusion module. This module receives the current network path information defined by the dynamic connection signals from the topology management module and is responsible for the transmission and communication of all information within the system. Its operation closely mimics the mechanism of information and material transmission through chemical signals and cytoplasmic flow in a mycelial network. In a mycelial network, signals do not propagate synchronously along a single, predetermined path, but rather diffusely, multi-path, and asynchronously throughout the network, ensuring that information ultimately reaches its destination even if a local network is damaged. In this system, this principle is manifested in the generation of routing diffusion signals. When a computational task or data packet needs to be transmitted, this module does not simply send the entire packet through an optimal path. Instead, it intelligently breaks down the data or task load to be distributed into multiple smaller, independently transmittable sub-signal units. This breakdown process is analogous to how a mycelial network encapsulates nutrients in vesicles for transport. Each sub-signal unit contains sufficient information, such as its source node identifier, its sequence number within the entire data packet, and the final destination node information.

[0029] In one embodiment of the invention, these sub-signal units are injected into multiple available network paths based on the current dynamic connection state of the network. They act like autonomous agents, transmitting independently and asynchronously forward within the network. Each intermediate node, upon receiving a sub-signal unit, forwards it to the next suitable neighbor node based on its locally maintained routing information (which itself is constantly updated from the dynamic connection signal). This transmission mechanism offers significant advantages: it completely avoids the congestion risks inherent in traditional networks due to reliance on a single path. Even if a path becomes slow or interrupted due to network congestion or node failure, sub-signal units on other paths can continue to transmit, ensuring the continuity of the overall transmission process. Ultimately, all these sub-signal units arrive at the target node via different paths. The signal propagation module instance on the target node, based on the sequence information carried by each sub-signal unit, like a sequence of... Figure 1 This mechanism reassembles them into complete original data or tasks. It greatly improves the efficiency of large-scale parallel data transmission and the robustness of the entire system in the face of network fluctuations or local failures, enabling the system to provide stable and reliable services in an unstable and unreliable network environment.

[0030] like Figure 1As shown, the system also includes a resource scheduling module. This module receives routing diffusion signals containing recombined task information from the signal diffusion module and determines where the computational task should be executed. Its decision-making mechanism directly mimics the nutrient transport behavior of fungal networks. In fungal networks, nutrients flow along concentration gradients from enriched to scarce regions, or respond to network chemical signals and are directed to the most resource-intensive areas. In this system, the resource scheduling module continuously receives resource status information from each computing node, such as the current utilization rate of the CPU, available memory capacity, and remaining network bandwidth. This information collectively constitutes the "resource concentration field" within the network. When a computational task arrives, the module analyzes this "resource concentration field" and generates a corresponding task scheduling signal. This signal is essentially an instruction that guides the task to the node with the most abundant computing resources and best suited for handling that type of task. For example, for a task requiring a large amount of memory, the scheduling signal will direct it to a cluster of nodes with ample free memory; for a task requiring high-speed computation, the signal will direct it to a node with low CPU utilization. This process is dynamic and real-time. As the task load and resource status in the network change continuously, the scheduling strategy will also be adjusted accordingly, thereby achieving automatic load balancing and optimal resource utilization, as if the entire computing network has a collective intelligence that seeks to benefit and avoid harm.

[0031] like Figure 1 As shown, to accomplish complex computational tasks, the system also incorporates a node collaboration module. This module receives task scheduling signals from the resource scheduling module and activates when a task is too large for a single node to complete independently. It mimics the ability of mycelial networks to solve complex environmental problems through collective collaboration (e.g., collaboratively decomposing large lignocellulose fibers). The module intelligently decomposes a large computational task into multiple logically related but computationally relatively independent subtask blocks. Subsequently, it generates collaborative computation signals containing a blueprint for the task decomposition, the execution code for each subtask block, and the necessary communication and synchronization protocols between the subtasks. These signals are distributed to temporary collaborative groups composed of multiple computing nodes. Each node is responsible for executing one or more subtasks and, during execution, exchanges intermediate computation results with other nodes through the system's built-in communication mechanism (supported by the signal propagation module). Finally, a designated node or a dedicated integration node is responsible for collecting all intermediate results, performing integrated computation, and producing the final output. This mechanism achieves an efficient and flexible decentralized collaborative computing model.

[0032] Furthermore, the entire system forms a sophisticated closed-loop feedback control system. The topology management module doesn't rely solely on initial information for decision-making; it continuously receives node status feedback signals from the node coordination module. These signals contain real-time performance data generated by nodes during task processing, such as task execution time, computational accuracy, and encountered errors. The topology management module incorporates this real-time feedback information into its decision-making algorithm, adjusting the generation strategy for subsequent dynamic connection signals accordingly. For example, if a node frequently reports computational errors or excessive latency, the topology management module might weaken its connection in subsequent decisions, or even guide the network to bypass that node. Conversely, if a node performs exceptionally well, the topology management module might strengthen its connection, making it a core hub in the network. This closed-loop feedback mechanism allows the entire system to learn from operational experience, continuously optimizing and repairing itself, ultimately exhibiting extremely strong adaptability and resilience, perfectly replicating the superior performance of biological systems in the face of changing environments.

[0033] like Figure 1 As shown, this scheme focuses on the resource scheduling module, elevating its function beyond simple task allocation to become an intelligent entity capable of proactively sensing risks and taking avoidance actions. Its core lies in simulating the stress avoidance response of a mycelial network when encountering physical obstacles or toxic substances in a natural environment. Technically, this module continuously monitors deep performance indicator signals fed back from each computing node. These signals include not only basic resource utilization but, more importantly, the node's computational error rate and task response latency. These indicators, like pain receptors in a living organism, can keenly indicate the "health" status of nodes. The system internally uses a dynamically adjusted threshold model to determine the node's state. When the resource scheduling module analyzes these signals and finds that the error rate of a specific node is continuously rising and exceeding its current tolerance threshold, it does not passively wait for tasks to fail on that node. Instead, it proactively identifies that node and its closely related neighboring nodes in the network topology as an "unhealthy network area" or "high-risk zone." Immediately afterwards, the module generates a special task avoidance signal. This signal carries a warning message, functioning similarly to dangerous chemical signals emitted by organisms. During scheduling, new computational tasks prioritize receiving and responding to this avoidance signal, automatically adjusting their expected execution path to bypass this perceived unstable region. More importantly, this process doesn't occur in isolation; the resource scheduling module collaborates with the topology management module. It triggers the topology management module to reassess its connections to the region and may initiate a "growth algorithm" to open new, safer alternative paths around the risky area, directing computational traffic to these new paths. This mechanism greatly ensures the continuity and ultimate success rate of computational tasks, achieving proactive fault tolerance for the system.

[0034] Specifically, how does the node collaboration module implement complex distributed computing tasks? It simulates the collective intelligent decision-making behavior exhibited by mycelial networks and even broader biological systems (such as ant colonies and bee colonies). When a large, complex computing task enters the system, this module does not attempt to find a supernode to handle it independently, as this is often inefficient and impractical. Instead, it first acts as an "architect," intelligently decomposing the task into a set of interconnected but logically independent subtask blocks. These subtask blocks are designed to execute in parallel on different computing nodes. Subsequently, the module generates an information-rich collaborative computing signal. This signal not only contains the subtask blocks themselves but also encodes the execution blueprint of the entire task: including dependencies between subtasks, data flow direction, necessary synchronization points, and communication protocols for inter-node interaction. This signal is distributed to a pre-selected or dynamically organized group of computing nodes. After receiving its portion of the signal, each node begins to execute its subtask independently. During execution, the nodes do not remain isolated from each other. They collaborate and share data by exchanging intermediate computation result signals according to the protocol in the collaborative computation signal. This communication ensures that all nodes work in parallel towards the same goal. Ultimately, one designated node (or a node dedicated to aggregation) is responsible for receiving all the final or intermediate outputs from the other nodes and performing the final integration computation to generate the final result of the entire task. This mechanism perfectly realizes a highly decentralized collaborative problem-solving model that does not rely on any master node for micromanagement. Each node, relying on a shared blueprint and simple interaction rules, collectively emerges the ability to solve complex problems.

[0035] This invention relates to a distributed computing system based on fungal hyphal network topology. By simulating the core biological mechanisms of fungal hyphal networks, such as growth, signal transmission, and resource allocation, the system abstracts and transforms these mechanisms into a complete technical solution. The system utilizes a topology management module for dynamic self-organizing network connections, a signal diffusion module for efficient asynchronous communication, and a resource scheduling module for biomimetic guidance of task flow. Ultimately, it forms a decentralized, environmentally adaptable, and highly fault-tolerant and self-healing distributed computing system.

[0036] Therefore, the distributed computing system based on fungal hyphal network topology of the present invention can solve the problems of distributed systems lacking adaptability, high robustness and low energy consumption consensus.

[0037] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A distributed computing system based on fungal hyphal network topology, characterized in that, include: The topology management module generates dynamic connection signals by simulating the biological behavior of fungal hyphae growth and fusion. These dynamic connection signals are used to indicate the establishment, enhancement, or dissolution of connections between computing nodes. A signal diffusion module receives the dynamic connection signal and generates a routing diffusion signal, which asynchronously propagates the status and instructions between nodes by simulating the chemical signal transmission in a mycelial network. The resource scheduling module receives the routing diffusion signal and generates a task scheduling signal, which directs the computing task to a resource-rich node by simulating a mycelial nutrient transport mechanism. A node collaboration module receives the task scheduling signal and generates a collaborative computing signal, which is used to coordinate multiple computing nodes to jointly execute a single distributed task. The topology management module also receives node status feedback signals from the node collaboration module and adjusts the generation strategy of the dynamic connection signals according to the signals, thereby forming a closed-loop adaptive network topology control system.

2. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The specific method by which the topology management module generates dynamic connection signals includes: simulating the vegetative growth behavior of mycelia, continuously receiving resource status broadcast signals from each computing node in the system, wherein the resource status broadcast signals include at least the CPU utilization, memory idle rate, and network bandwidth redundancy; the topology management module calculates the resource enrichment direction based on the resource status broadcast signals, and preferentially generates connection establishment signals pointing to high resource nodes, while generating connection weakening or interruption signals for resource overloaded nodes, thereby realizing the biomimetic growth and expansion of the computing network topology towards resource-rich areas.

3. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The specific method by which the signal diffusion module generates the routing diffusion signal includes: simulating the multipath transmission mechanism of cytoplasmic flow and vesicle transport in mycelial networks, splitting the computing tasks or data packets to be distributed into multiple sub-signal units; each sub-signal unit is independently transmitted asynchronously along the network path defined by the dynamic connection signal, carrying the source node identifier and sequence information during transmission; finally, all sub-signal units are reassembled and restored at the target node according to the sequence information they carry. This method avoids the blocking risk of a single path in traditional networks and significantly improves the parallel efficiency and overall robustness of data transmission.

4. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The resource scheduling module generates task scheduling signals in the following ways: simulating the stress avoidance response of a mycelial network when encountering obstacles or toxins, and continuously monitoring the error rate and response delay signals of each computing node; when the error rate of a specific node exceeds a preset threshold, the resource scheduling module determines that the area is an unhealthy network area and generates a task avoidance signal; the task avoidance signal will guide new computing tasks to bypass the unhealthy node and its closely related neighboring nodes, and through the topology management module, trigger the process of opening alternative paths in the network, thereby ensuring the continuity and success rate of computing tasks.

5. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The specific method by which the node collaboration module generates collaborative computing signals includes: simulating the swarm intelligence decision-making behavior of mycelial networks, decomposing a large computing task into multiple interconnected sub-task blocks; the collaborative computing signals are distributed to a group of computing nodes, and the signals encode the execution logic of each sub-task block and the necessary interaction protocols between nodes; each node collaborates by exchanging intermediate computing result signals during the execution of sub-tasks, and finally one node or a dedicated node is responsible for integrating all intermediate results and outputting the final result, thus realizing decentralized collaborative problem solving.

6. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The system also includes a network memory learning module, which connects the topology management module and the node collaboration module. This module receives and stores historical dynamic connection signals and collaborative computing signals. The network memory learning module trains its internal decision-making model by analyzing the success and failure patterns of historical signals and generates network optimization prediction signals, which are then fed back to the topology management module. The topology management module uses these prediction signals to pre-build and optimize potential future connection relationships, thereby enabling the entire system to learn from historical experience and continuously improve itself.

7. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The sub-signal unit transmission mechanism of the signal diffusion module also includes a priority marking function; each sub-signal unit is assigned a priority mark when it is created, which is generated based on the urgency of its computing task and the importance of the data. During transmission, when network path congestion occurs, the sub-signal unit holding the high priority tag will gain priority forwarding rights and occupy network bandwidth resources for transmission, thereby ensuring that critical tasks can be delivered and processed with low latency and achieving differentiated service quality assurance.

8. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The task avoidance mechanism of the resource scheduling module also includes a node recovery monitoring function. After generating a task avoidance signal and guiding the task to detour, the resource scheduling module will not permanently abandon the unhealthy node, but will continuously send lightweight probe task signals to the node at a low frequency and monitor its response. Once the error rate and response latency of the node are detected to return to normal levels, the resource scheduling module will generate a node recovery signal and gradually reschedule new computing tasks to the node, thereby achieving full and flexible utilization of network resources.

9. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The dynamic connection strategy of the topology management module includes a periodic connection optimization function. This module evaluates the current topology connection relationship of the entire network at fixed time intervals, and the evaluation is based on connection utilization signals and path delay signals. For connections that are in a low utilization state for a long time, the topology management module generates a connection reclamation signal to release the connection and save resources. At the same time, for critical path areas with high load and high latency, a connection enhancement signal is generated to incentivize the establishment of new redundant connections in their vicinity to share the load, thereby realizing the periodic self-optimization of the network.

10. The distributed computing system based on fungal hyphal network topology according to claim 1, characterized in that, The collaborative computing signal of the node collaboration module also integrates a result verification mechanism; when the node responsible for integrating the final result completes the result calculation, it will broadcast a result summary verification request signal to other nodes that participated in the collaborative computing. After receiving a verification request, participating nodes use the intermediate computation results stored locally to perform consistency verification on the final result and generate a verification result signal through a voting mechanism. Only when the verification result signal indicates that consistency has been achieved will the final result be confirmed and output. Otherwise, the task recalculation process will be triggered to ensure the correctness and reliability of the distributed computation results.

Images