CN112698944A - Distributed cloud computing system and method based on human brain simulation - Google Patents

Distributed cloud computing system and method based on human brain simulation Download PDF

Info

Publication number
CN112698944A
CN112698944A CN202011603114.7A CN202011603114A CN112698944A CN 112698944 A CN112698944 A CN 112698944A CN 202011603114 A CN202011603114 A CN 202011603114A CN 112698944 A CN112698944 A CN 112698944A
Authority
CN
China
Prior art keywords
node
cloud
computing
nodes
distributed
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202011603114.7A
Other languages
Chinese (zh)
Inventor
桑洪升
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Leling Auman Electronic Technology Co ltd
Original Assignee
Leling Auman Electronic Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Leling Auman Electronic Technology Co ltd filed Critical Leling Auman Electronic Technology Co ltd
Priority to CN202011603114.7A priority Critical patent/CN112698944A/en
Publication of CN112698944A publication Critical patent/CN112698944A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5072—Grid computing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computer And Data Communications (AREA)

Abstract

The invention belongs to the technical field of cloud computing, and particularly relates to a distributed cloud computing system and method based on human brain simulation. The system comprises: the distributed computing cloud cluster comprises a plurality of cloud nodes; each cloud node has different computing characteristics; according to different computing characteristics, the cloud nodes at least comprise the following three types: the system comprises an image computing cloud node, a digital computing cloud node and a storage computing cloud node; the cloud nodes each include: a collection node and a number of distributed computing nodes. The process of task classification and neural information transmission of the human brain is simulated, so that efficient distributed computation is realized, and the efficiency of the distributed computation is improved; meanwhile, in the process of classifying and distributing the calculation tasks, the distribution is carried out according to the resource occupancy rate and the calculation task types, and the distribution accuracy is high.

Description

Distributed cloud computing system and method based on human brain simulation
Technical Field
The invention belongs to the technical field of cloud computing, and particularly relates to a distributed cloud computing system and method based on human brain simulation.
Background
In computer science, Distributed computing (english) is translated into Distributed computing. This field of research is mainly concerned with how Distributed systems (Distributed systems) perform calculations. A distributed system is a system of computers interconnected via a network to communicate messages and communications and coordinate their activities. The components interact with each other to achieve a common goal. The engineering data which needs to be calculated in large quantity is divided into small blocks, a plurality of computers calculate the small blocks respectively, and after the calculation results are uploaded, the results are unified and combined to obtain the science of data conclusion. Examples of distributed systems come from different service-oriented architectures, massively multiplayer online games, peer-to-peer network applications.
Currently common distributed computing projects typically use the idle computing power of tens of millions of volunteer computers around the world for data transmission (volunteer computing) over the internet. For example, the Folding @ home project for analyzing and calculating the internal structure of a protein and related drugs is huge in structure, requires a tremendous amount of calculation, and cannot be completed by one computer. Although supercomputers with powerful computing power are available, these devices are expensive, some scientific research institutions have very limited expenses, and the goal can be achieved with less cost by means of distributed computing.
In the prior art, patent No. CN201310345015.7A is directed to a method and system for facilitating service provisioning and coordination for a distributed computing system, which provides a distributed node service management system that supports the primary data processing functions of the distributed computing system using a plurality of existing processor nodes of the distributed computing system. The distributed node service management system coordinates and manages service functions on behalf of the processor nodes of the distributed computing system. Other features and aspects may be implemented depending on the particular application. Mainly aiming at the distributed computing of processing, but aiming at the distributed computing of cloud computing, no corresponding solution is provided yet.
Patent No. CN201280061203.0A discloses a distributed computing method and a distributed storage module for data tasks; the method is started by the following operations: a Distributed Storage (DS) processing module selects a set of Distributed Storage and Task (DST) execution units for executing a task and determines a distributed storage error encoding parameter for data. The method continues with the following operations: the DS processing module performs distributed storage error coding on the data according to the parameters to generate a plurality of coded data blocks, and groups the plurality of coded data blocks into a plurality of coded data block groups; the method continues with the following operations: the DS processing module partitions the task into a set of partial tasks, outputs at least some of the plurality of encoded data block groupings to a set of DST execution units, and outputs the set of partial tasks to the set of DST execution units to execute the set of partial tasks on the at least some of the plurality of encoded data block groupings. The emphasis is on the storage of distributed computing, and meanwhile, the resource allocation is not combined with the specific situation of computing tasks, and the problem of the efficiency of distributed computing is still not solved.
Disclosure of Invention
In view of this, the main objective of the present invention is to provide a distributed cloud computing system and method based on human brain simulation, which implement efficient distributed computing and improve the efficiency of distributed computing by simulating the processes of task classification and neural information transmission of human brain; meanwhile, in the process of classifying and distributing the calculation tasks, the distribution is carried out according to the resource occupancy rate and the calculation task types, and the distribution accuracy is high.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
a distributed cloud computing system based on human brain simulation, the system comprising: the method comprises the following steps: the distributed computing cloud cluster comprises a plurality of cloud nodes; each cloud node has different computing characteristics; according to different computing characteristics, the cloud nodes at least comprise the following three types: the system comprises an image computing cloud node, a digital computing cloud node and a storage computing cloud node; the cloud nodes each include: the system comprises a collection node and a plurality of distributed computing nodes; the collection node is configured to distribute the received computing tasks to the distributed computing nodes based on the resource occupation condition of each current distributed computing node; the distributed computing nodes are configured to compute the received computing tasks, and return the computing results to the sink node after the computing results are obtained; the system further comprises: the transit node is configured and used for analyzing the received computing task, judging the type of the computing task, and sending the computing task to the corresponding cloud node after path planning based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster; meanwhile, the calculation results of the collection nodes of each cloud node are collected in real time, and after being converged, the calculation results are sent to a sender of the calculation request to complete the calculation.
Further, the transit node includes: the task analysis node is configured to analyze the received computing task and judge the type of the computing task; and the routing node is configured to perform path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then send the computing task to the corresponding cloud node.
Further, the method for sending the route searching node to the corresponding cloud node after performing path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster executes the following steps: randomly selecting N cloud nodes in the distributed computing cloud cluster; determining a central node from the selected N nodes, wherein the coordinate of the central node is (0, 0); defining a transmission min chain, the transmission min chain comprising: transmitting a minimum chain root node and transmitting a minimum chain sub-node; other nodes to which sub-nodes in the transmission minimum chain can be connected; considering resource occupancy L of each cloud nodemnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: m and n represent nodes; data transmission is carried out according to the established path, the child nodes in the minimum chain are transmitted to the father node, the father node carries out aggregation operation on the collected data and the data sent by the child nodes, and then the data obtained by aggregation are sent to the father node and the center node; after the operation of the set time period, the self-adaptive maintenance and updating of the tree are carried out, and the data transmission is continued.
Further, the resource occupancy rate L of each cloud node is consideredmnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: the method of m and n representing nodes performs the following steps: the original path planning formula is:
Figure BDA0002869409550000031
Figure BDA0002869409550000041
eta inmnAnd updating by using a new operator, wherein the updated formula is as follows:
Figure BDA0002869409550000042
wherein: t is represented by time, τmnAnd the correction coefficient representing the resource occupancy rate of each cloud node has a value range as follows: 0.1 to 0.3, ηmnRepresenting an isolation operator between the m node and the node n, alpha and beta respectively represent the resource occupancy rate of each cloud node and the weight of the isolation operator, namely factors starting in the path selection process, and RS and BA represent that L is equal to Lmn、PmnAnd BmnParameters quantized to the same dimension.
Further, the transit node further includes a fault tolerant node configured to: removing cloud nodes that always respond slowly or do not respond within a predetermined time from the distributed computing cloud cluster; and detecting the removed cloud node, and moving the cloud node back to the distributed computing cloud cluster under the condition that the detection result shows that the cloud node is available.
A distributed cloud computing method based on human brain simulation, the method performing the steps of:
step 1: the transfer node receives the computing request, analyzes the received computing task, judges the type of the computing task, performs path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then sends the computing task to the corresponding cloud node;
step 2: the method comprises the steps that collection nodes in the distributed computing cloud cluster distribute received computing tasks to distributed computing nodes based on the resource occupation condition of each current distributed computing node;
and step 3: and the distributed computing nodes in the distributed computing cloud cluster compute the received computing tasks, and return the computing results to the collection node after obtaining the computing results.
And 4, step 4: and the transit node collects the calculation results of the collection node of each cloud node in real time, converges the calculation results and sends the calculation results to a sender of the calculation request to finish the calculation.
Further, the transit node includes: the task analysis node is configured to analyze the received computing task and judge the type of the computing task; and the routing node is configured to perform path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then send the computing task to the corresponding cloud node.
Further, the method for sending the route searching node to the corresponding cloud node after performing path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster executes the following steps: randomly selecting N cloud nodes in the distributed computing cloud cluster; determining a central node from the selected N nodes, wherein the coordinate of the central node is (0, 0); defining a transmission min chain, the transmission min chain comprising: transmitting a minimum chain root node and transmitting a minimum chain sub-node; other nodes to which sub-nodes in the transmission minimum chain can be connected; considering resource occupancy L of each cloud nodemnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: m and n represent nodes; data transmission is carried out according to the established path, the child nodes in the minimum chain are transmitted to the father node, the father node carries out aggregation operation on the collected data and the data sent by the child nodes, and then the data obtained by aggregation are sent to the father node and the center node; after the operation of the set time period, the self-adaptive maintenance and updating of the tree are carried out, and the data transmission is continued.
Further, the resource occupancy rate L of each cloud node is consideredmnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: the method of m and n representing nodes performs the following steps: will be originalThe path planning formula of (c):
Figure BDA0002869409550000051
Figure BDA0002869409550000052
eta inmnAnd updating by using a new operator, wherein the updated formula is as follows:
Figure BDA0002869409550000053
wherein: t is represented by time, τmnAnd the correction coefficient representing the resource occupancy rate of each cloud node has a value range as follows: 0.1 to 0.3, ηmnRepresenting an isolation operator between the m node and the node n, alpha and beta respectively represent the resource occupancy rate of each cloud node and the weight of the isolation operator, namely factors starting in the path selection process, and RS and BA represent that L is equal to Lmn、PmnAnd BmnParameters quantized to the same dimension.
Further, the transit node further includes a fault tolerant node configured to: removing cloud nodes that always respond slowly or do not respond within a predetermined time from the distributed computing cloud cluster; and detecting the removed cloud node, and moving the cloud node back to the distributed computing cloud cluster under the condition that the detection result shows that the cloud node is available.
The distributed cloud computing system and method based on human brain simulation have the following beneficial effects: the process of task classification and neural information transmission of the human brain is simulated, so that efficient distributed computation is realized, and the efficiency of the distributed computation is improved; meanwhile, in the process of classifying and distributing the calculation tasks, the distribution is carried out according to the resource occupancy rate and the calculation task types, and the distribution accuracy is high. The method is mainly realized by the following steps: 1. analyzing and distributing the computing tasks, namely analyzing the received computing tasks, judging the types of the computing tasks, planning paths of the computing tasks based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then sending the computing tasks to the corresponding cloud nodes; thus can be specifiedThe type of computing task is distributed to the corresponding cloud node, and the computing efficiency of the cloud node is higher for the type of computing task; if the image computing task is distributed to the image computing cloud node, the computing efficiency is higher than that of the digital computing cloud node; 2. path planning, namely performing path planning in a plurality of cloud nodes aiming at a computing task, so that the efficiency of sending the computing task to a target computing cloud node is higher, and the computing efficiency of the whole computing task is higher; 3. determination of a path planning formula:
Figure BDA0002869409550000061
the formula is calculated based on the resource occupancy rate of the node, and compared with the traditional path planning algorithm, the path planning method is more suitable for the invention; 4. adding fault-tolerant nodes: according to the invention, through the fault-tolerant node, the problem node in the system can be found in real time, and then the problem node is removed, so that the situation of calculation failure is avoided.
Drawings
Fig. 1 is a schematic system structure diagram of a distributed cloud computing system based on human brain simulation according to an embodiment of the present invention;
fig. 2 is a schematic method flow diagram of a distributed cloud computing method based on human brain simulation according to an embodiment of the present invention.
Fig. 3 is a schematic network structure diagram of a distributed cloud computing system and method based on human brain simulation according to an embodiment of the present invention;
fig. 4 is a schematic diagram illustrating a principle of classifying computing tasks by a transit point of the distributed cloud computing system and method based on human brain simulation according to the embodiment of the present invention.
Fig. 5 is a comparison experiment effect schematic diagram of a calculation efficiency experiment effect schematic diagram of the distributed cloud computing system and method based on human brain simulation provided by the embodiment of the invention and a calculation efficiency experiment effect schematic diagram in the prior art.
1. The invention relates to an experimental curve, 2-the existing distributed calculation experimental curve and 3-the non-distributed calculation experimental curve.
Detailed Description
The method of the present invention will be described in further detail below with reference to the accompanying drawings and embodiments of the invention.
Example 1
As shown in fig. 1, a distributed cloud computing system based on human brain simulation, the system comprising: the method comprises the following steps: the distributed computing cloud cluster comprises a plurality of cloud nodes; each cloud node has different computing characteristics; according to different computing characteristics, the cloud nodes at least comprise the following three types: the system comprises an image computing cloud node, a digital computing cloud node and a storage computing cloud node; the cloud nodes each include: the system comprises a collection node and a plurality of distributed computing nodes; the collection node is configured to distribute the received computing tasks to the distributed computing nodes based on the resource occupation condition of each current distributed computing node; the distributed computing nodes are configured to compute the received computing tasks, and return the computing results to the sink node after the computing results are obtained; the system further comprises: the transit node is configured and used for analyzing the received computing task, judging the type of the computing task, and sending the computing task to the corresponding cloud node after path planning based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster; meanwhile, the calculation results of the collection nodes of each cloud node are collected in real time, and after being converged, the calculation results are sent to a sender of the calculation request to complete the calculation.
Particularly, the efficient distributed computation is realized and the efficiency of the distributed computation is improved by simulating the processes of task classification and neural information transmission of the human brain; meanwhile, in the process of classifying and distributing the calculation tasks, the distribution is carried out according to the resource occupancy rate and the calculation task types, and the distribution accuracy is high. The method is mainly realized by the following steps: 1. analyzing and distributing the computing tasks, namely analyzing the received computing tasks, judging the types of the computing tasks, planning paths of the computing tasks based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then sending the computing tasks to the corresponding cloud nodes; in this way, the specific type of computing task can be distributed to the corresponding cloud node, and the computing efficiency of the cloud node is higher for the type of computing task; if the image computing task is distributed to the image computing cloud node, the computing efficiency is higher than that of the digital computing cloud node; 2. path planning, namely performing path planning in a plurality of cloud nodes aiming at a computing task, so that the efficiency of sending the computing task to a target computing cloud node is higher, and the computing efficiency of the whole computing task is higher; 3. determination of a path planning formula:
Figure BDA0002869409550000081
the formula is calculated based on the resource occupancy rate of the node, and compared with the traditional path planning algorithm, the path planning method is more suitable for the invention; 4. adding fault-tolerant nodes: according to the invention, through the fault-tolerant node, the problem node in the system can be found in real time, and then the problem node is removed, so that the situation of calculation failure is avoided.
Example 2
On the basis of the above embodiment, the transit node includes: the task analysis node is configured to analyze the received computing task and judge the type of the computing task; and the routing node is configured to perform path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then send the computing task to the corresponding cloud node.
Specifically, in the existing distributed computing method for processing computing tasks in units of server clusters, one server cluster can only implement one computing task processing method, and when there are many computing task processing methods, a large number of server clusters must be deployed to meet the processing requirements of the computing tasks, so that the construction cost of the server clusters is greatly increased.
Example 3
On the basis of the previous embodiment, the method for sending the routing node to the corresponding cloud node after performing path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster performs the following steps: randomly selecting N cloud nodes in the distributed computing cloud cluster; determining a central node from the selected N nodes, wherein the coordinate of the central node is (0, 0); defining a transmission min chain, the transmission min chain comprising: transmitting a minimum chain root node and transmitting a minimum chain sub-node; other nodes to which sub-nodes in the transmission minimum chain can be connected; considering resource occupancy L of each cloud nodemnA transmission distance PmnChannel security BmmAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: m and n represent nodes; data transmission is carried out according to the established path, the child nodes in the minimum chain are transmitted to the father node, the father node carries out aggregation operation on the collected data and the data sent by the child nodes, and then the data obtained by aggregation are sent to the father node and the center node; after the operation of the set time period, the self-adaptive maintenance and updating of the tree are carried out, and the data transmission is continued.
In particular, clusters are often used in relatively large multi-user computing systems where high performance and reliability are a concern. For example, a cluster may be used to provide redundancy or fault tolerance such that, in the event of a failure of any node in the cluster, operations previously performed by that node are handed over to other nodes in the cluster. Clusters are also used to improve overall performance over single ー computers, since multiple nodes are typically capable of handling a large number of tasks in parallel, more than a single computer. In general, load balancing may be utilized to ensure that tasks are reasonably distributed among nodes to prevent individual nodes from being overloaded and thereby maximize overall system performance. For example, ー specific applications of a cluster are to provide multi-user access to a shared resource, such as a database or storage device, because multiple nodes are capable of handling a relatively large number of user access requests, and the shared resource is typically still available to users even when any node in the cluster fails.
Example 4
On the basis of the previous embodiment, the resource occupancy rate L of each cloud node is consideredmnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: the method of m and n representing nodes performs the following steps: the original path planning formula is:
Figure BDA0002869409550000101
eta inmnAnd updating by using a new operator, wherein the updated formula is as follows:
Figure BDA0002869409550000102
wherein: t is represented by time, τmnAnd the correction coefficient representing the resource occupancy rate of each cloud node has a value range as follows: 0.1 to 0.3, ηmnRepresenting an isolation operator between the m node and the node n, alpha and beta respectively represent the resource occupancy rate of each cloud node and the weight of the isolation operator, namely factors starting in the path selection process, and RS and BA represent that L is equal to Lmn、PmnAnd BmnParameters quantized to the same dimension.
Example 5
On the basis of the above embodiment, the transit node further includes a fault-tolerant node configured to: removing cloud nodes that always respond slowly or do not respond within a predetermined time from the distributed computing cloud cluster; and detecting the removed cloud node, and moving the cloud node back to the distributed computing cloud cluster under the condition that the detection result shows that the cloud node is available.
Specifically, as the demands and complexity of clusters and other distributed computing systems increase, the concerns about scalability and performance continue to increase. It is not unreasonable to expect distributed computing systems to serve potentially millions of clients, and it has been found that as the complexity of distributed computing systems serving these clients increases, the distribution of load among servers, nodes, or other computer components making up such systems becomes increasingly critical to the stability and performance of such systems. In a distributed computing system, routing client requests to the appropriate computers (e.g., servers) is critical to properly distributing the load.
Example 6
As shown in fig. 2, a distributed cloud computing method based on human brain simulation, the method performs the following steps:
step 1: the transfer node receives the computing request, analyzes the received computing task, judges the type of the computing task, performs path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then sends the computing task to the corresponding cloud node;
step 2: the method comprises the steps that collection nodes in the distributed computing cloud cluster distribute received computing tasks to distributed computing nodes based on the resource occupation condition of each current distributed computing node;
and step 3: and the distributed computing nodes in the distributed computing cloud cluster compute the received computing tasks, and return the computing results to the collection node after obtaining the computing results.
And 4, step 4: and the transit node collects the calculation results of the collection node of each cloud node in real time, converges the calculation results and sends the calculation results to a sender of the calculation request to finish the calculation.
Referring to fig. 3, each distributed computing node corresponds to one aggregation node and several cloud nodes.
Example 7
On the basis of the above embodiment, the transit node includes: the task analysis node is configured to analyze the received computing task and judge the type of the computing task; and the routing node is configured to perform path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then send the computing task to the corresponding cloud node.
Referring to fig. 4, in the process of classifying the transit nodes, the transit nodes first store the received computation tasks, and then classify the computation tasks when the computation tasks are accumulated to a certain extent. Because the calculation tasks of the same category often have similar characteristics, after one task is analyzed, the calculation tasks of the same analogy can be classified, and the classification efficiency is improved.
Example 8
On the basis of the previous embodiment, the method for sending the routing node to the corresponding cloud node after performing path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster performs the following steps: randomly selecting N cloud nodes in the distributed computing cloud cluster; determining a central node from the selected N nodes, wherein the coordinate of the central node is (0, 0); defining a transmission min chain, the transmission min chain comprising: transmitting a minimum chain root node and transmitting a minimum chain sub-node; other nodes to which sub-nodes in the transmission minimum chain can be connected; considering resource occupancy L of each cloud nodemnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: m and n represent nodes; data transmission is carried out according to the established path, the child nodes in the minimum chain are transmitted to the father node, the father node carries out aggregation operation on the collected data and the data sent by the child nodes, and then the data obtained by aggregation are sent to the father node and the center node; after the operation of the set time period, the self-adaptive maintenance and updating of the tree are carried out, and the data transmission is continued.
Example 9
On the basis of the previous embodiment, the resource occupancy rate L of each cloud node is consideredmnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: the method of m and n representing nodes performs the following steps: the original path planning formula is:
Figure BDA0002869409550000121
eta inmnAnd updating by using a new operator, wherein the updated formula is as follows:
Figure BDA0002869409550000122
wherein: t is represented by time, τmnAnd the correction coefficient representing the resource occupancy rate of each cloud node has a value range as follows: 0.1 to 0.3, ηmnRepresenting an isolation operator between the m node and the node n, alpha and beta respectively represent the resource occupancy rate of each cloud node and the weight of the isolation operator, namely factors starting in the path selection process, and RS and BA represent that L is equal to Lmn、PmnAnd BmnParameters quantized to the same dimension.
Example 10
On the basis of the above embodiment, the transit node further includes a fault-tolerant node configured to: removing cloud nodes that always respond slowly or do not respond within a predetermined time from the distributed computing cloud cluster; and detecting the removed cloud node, and moving the cloud node back to the distributed computing cloud cluster under the condition that the detection result shows that the cloud node is available.
Referring to fig. 5, the invention realizes the distributed computation of the college through the process of simulating the task classification and neural information transmission of the human brain, and improves the efficiency of the distributed computation; meanwhile, in the process of classifying and distributing the calculation tasks, the distribution is carried out according to the resource occupancy rate and the calculation task type, and the distribution accuracy is high
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process and related description of the system described above may refer to the corresponding process in the foregoing method embodiments, and will not be described herein again.
It should be noted that, the system provided in the foregoing embodiment is only illustrated by dividing the functional units, and in practical applications, the functions may be distributed by different functional units according to needs, that is, the units or steps in the embodiments of the present invention are further decomposed or combined, for example, the units in the foregoing embodiment may be combined into one unit, or may be further decomposed into multiple sub-units, so as to complete all or the functions of the units described above. The names of the units and steps involved in the embodiments of the present invention are only for distinguishing the units or steps, and are not to be construed as unduly limiting the present invention.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes and related descriptions of the storage device and the processing device described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
Those of skill in the art would appreciate that the various illustrative elements, method steps, described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that programs corresponding to the elements, method steps may be located in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. To clearly illustrate this interchangeability of electronic hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The terms "first," "second," and the like are used for distinguishing between similar elements and not necessarily for describing or implying a particular order or sequence.
The terms "comprises," "comprising," or any other similar term are intended to cover a non-exclusive inclusion, such that a process, method, article, or unit/apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or unit/apparatus.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.

Claims (10)

1. Distributed cloud computing system based on human brain simulation, characterized in that the system comprises: the method comprises the following steps: the distributed computing cloud cluster comprises a plurality of cloud nodes; each cloud node has different computing characteristics; according to different computing characteristics, the cloud nodes at least comprise the following three types: the system comprises an image computing cloud node, a digital computing cloud node and a storage computing cloud node; the cloud nodes each include: the system comprises a collection node and a plurality of distributed computing nodes; the collection node is configured to distribute the received computing tasks to the distributed computing nodes based on the resource occupation condition of each current distributed computing node; the distributed computing nodes are configured to compute the received computing tasks, and return the computing results to the sink node after the computing results are obtained; the system further comprises: the transit node is configured and used for analyzing the received computing task, judging the type of the computing task, and sending the computing task to the corresponding cloud node after path planning based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster; meanwhile, the calculation results of the collection nodes of each cloud node are collected in real time, and after being converged, the calculation results are sent to a sender of the calculation request to complete the calculation.
2. The system of claim 1, wherein the transit node comprises: the task analysis node is configured to analyze the received computing task and judge the type of the computing task; and the routing node is configured to perform path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then send the computing task to the corresponding cloud node.
3. The system of claim 2, wherein the routing node performs the following steps in a method for sending the computation task to the corresponding cloud node after performing path planning based on the resource occupation status of each cloud node and the location information of each cloud node in the distributed computing cloud cluster: randomly selecting N cloud nodes in the distributed computing cloud cluster; determining a central node from the selected N nodes, wherein the coordinate of the central node is (0, 0); defining a transmission min chain, the transmission min chain comprising: transmitting a minimum chain root node and transmitting a minimum chain sub-node; other nodes to which sub-nodes in the transmission minimum chain can be connected; considering resource occupancy L of each cloud nodemnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: m and n represent nodes; data transmission is carried out according to the established path, the child nodes in the minimum chain are transmitted to the father node, the father node carries out aggregation operation on the collected data and the data sent by the child nodes, and then the data obtained by aggregation are sent to the father node and the center node; after the operation of the set time period, the self-adaptive maintenance and updating of the tree are carried out, and the data transmission is continued.
4. The system of claim 3, wherein the consideration of resource occupancy L for each cloud nodemnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: the method of m and n representing nodes performs the following steps: planning the original pathThe formula:
Figure FDA0002869409540000021
eta inmnAnd updating by using a new operator, wherein the updated formula is as follows:
Figure FDA0002869409540000022
wherein: t is represented by time, τmnAnd the correction coefficient representing the resource occupancy rate of each cloud node has a value range as follows: 0.1 to 0.3, ηmnRepresenting an isolation operator between the m node and the node n, alpha and beta respectively represent the resource occupancy rate of each cloud node and the weight of the isolation operator, namely factors starting in the path selection process, and RS and BA represent that L is equal to Lmn、PmnAnd BmnParameters quantized to the same dimension.
5. The system of claim 4, wherein the transit node further comprises a fault tolerant node configured to: removing cloud nodes that always respond slowly or do not respond within a predetermined time from the distributed computing cloud cluster; and detecting the removed cloud node, and moving the cloud node back to the distributed computing cloud cluster under the condition that the detection result shows that the cloud node is available.
6. Human brain simulation based distributed cloud computing method based on the system according to one of the claims 1 to 5, characterized in that the method performs the following steps:
step 1: the transfer node receives the computing request, analyzes the received computing task, judges the type of the computing task, performs path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then sends the computing task to the corresponding cloud node;
step 2: the method comprises the steps that collection nodes in the distributed computing cloud cluster distribute received computing tasks to distributed computing nodes based on the resource occupation condition of each current distributed computing node;
and step 3: and the distributed computing nodes in the distributed computing cloud cluster compute the received computing tasks, and return the computing results to the collection node after obtaining the computing results.
And 4, step 4: and the transit node collects the calculation results of the collection node of each cloud node in real time, converges the calculation results and sends the calculation results to a sender of the calculation request to finish the calculation.
7. The method of claim 6, wherein the transit node comprises: the task analysis node is configured to analyze the received computing task and judge the type of the computing task; and the routing node is configured to perform path planning on the computing task based on the resource occupation condition of each cloud node and the position information of each cloud node in the distributed computing cloud cluster, and then send the computing task to the corresponding cloud node.
8. The method of claim 2, wherein the routing node performs the following steps in a method for sending the computation task to the corresponding cloud node after performing path planning based on the resource occupation status of each cloud node and the location information of each cloud node in the distributed computing cloud cluster: randomly selecting N cloud nodes in the distributed computing cloud cluster; determining a central node from the selected N nodes, wherein the coordinate of the central node is (0, 0); defining a transmission min chain, the transmission min chain comprising: transmitting a minimum chain root node and transmitting a minimum chain sub-node; other nodes to which sub-nodes in the transmission minimum chain can be connected; considering resource occupancy L of each cloud nodemnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: m and n represent nodes; data transmission is carried out according to the established path, the child nodes in the minimum chain are transmitted to the father node, the father node carries out aggregation operation on the collected data and the data sent by the child nodes, and then the data obtained by aggregation are sent to the father node and the center node; after running for a set period of timeThe tree is adaptively maintained and updated, and data transmission is continued.
9. The method of claim 8, wherein the considering resource occupancy L of each cloud nodemnA transmission distance PmnChannel security BmnAnd k value updating the algorithm formula of the path planning, wherein the k value is the number of the neighbor nodes: the method of m and n representing nodes performs the following steps: the original path planning formula is:
Figure FDA0002869409540000041
eta inmnAnd updating by using a new operator, wherein the updated formula is as follows:
Figure FDA0002869409540000042
wherein: t is represented by time, τmnAnd the correction coefficient representing the resource occupancy rate of each cloud node has a value range as follows: 0.1 to 0.3, ηmnRepresenting an isolation operator between the m node and the node n, alpha and beta respectively represent the resource occupancy rate of each cloud node and the weight of the isolation operator, namely factors starting in the path selection process, and RS and BA represent that L is equal to Lmn、PmnAnd BmnParameters quantized to the same dimension.
10. The method of claim 9, wherein the transit node further comprises a fault tolerant node configured to: removing cloud nodes that always respond slowly or do not respond within a predetermined time from the distributed computing cloud cluster; and detecting the removed cloud node, and moving the cloud node back to the distributed computing cloud cluster under the condition that the detection result shows that the cloud node is available.
CN202011603114.7A 2020-12-29 2020-12-29 Distributed cloud computing system and method based on human brain simulation Pending CN112698944A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011603114.7A CN112698944A (en) 2020-12-29 2020-12-29 Distributed cloud computing system and method based on human brain simulation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011603114.7A CN112698944A (en) 2020-12-29 2020-12-29 Distributed cloud computing system and method based on human brain simulation

Publications (1)

Publication Number Publication Date
CN112698944A true CN112698944A (en) 2021-04-23

Family

ID=75512227

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011603114.7A Pending CN112698944A (en) 2020-12-29 2020-12-29 Distributed cloud computing system and method based on human brain simulation

Country Status (1)

Country Link
CN (1) CN112698944A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113934545A (en) * 2021-12-17 2022-01-14 飞诺门阵(北京)科技有限公司 Video data scheduling method, system, electronic equipment and readable medium

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104320468A (en) * 2014-10-29 2015-01-28 北京恩信创业科技有限公司 Server cluster deploying method based on cloud computing
CN106293933A (en) * 2015-12-29 2017-01-04 北京典赞科技有限公司 A kind of cluster resource configuration supporting much data Computational frames and dispatching method
CN108595368A (en) * 2018-04-26 2018-09-28 北京计算机技术及应用研究所 Concurrent computational system and method based on production domesticization computer cluster
CN108829509A (en) * 2018-05-03 2018-11-16 山东汇贸电子口岸有限公司 Distributed container cluster framework resources management method based on domestic CPU and operating system
CN108924217A (en) * 2018-06-29 2018-11-30 中山大学 A kind of distribution cloud system Automation arranging method
CN109067859A (en) * 2018-07-20 2018-12-21 北京航空航天大学 A kind of double-deck cloud framework system and implementation method towards cross-domain cooperation with service
CN110057360A (en) * 2019-03-08 2019-07-26 江苏海事职业技术学院 A kind of paths planning method and its system based on Distributed Parallel Computing
CN110113406A (en) * 2019-04-29 2019-08-09 成都网阔信息技术股份有限公司 Based on distributed calculating service cluster frame
CN110875833A (en) * 2018-08-31 2020-03-10 阿里巴巴集团控股有限公司 Cluster hybrid cloud, job processing method and device and electronic equipment
CN111614785A (en) * 2020-06-03 2020-09-01 成都智视慧语科技有限公司 Edge AI (Artificial Intelligence) computing cluster based on micro-container cloud
CN111614567A (en) * 2020-04-03 2020-09-01 杭州鹿扬科技有限公司 Node path selection method, system and device in block chain network
CN112055402A (en) * 2020-08-14 2020-12-08 杭州伍暨通信设备有限公司 Wireless sensor network node awakening system and method based on way finding algorithm

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104320468A (en) * 2014-10-29 2015-01-28 北京恩信创业科技有限公司 Server cluster deploying method based on cloud computing
CN106293933A (en) * 2015-12-29 2017-01-04 北京典赞科技有限公司 A kind of cluster resource configuration supporting much data Computational frames and dispatching method
CN108595368A (en) * 2018-04-26 2018-09-28 北京计算机技术及应用研究所 Concurrent computational system and method based on production domesticization computer cluster
CN108829509A (en) * 2018-05-03 2018-11-16 山东汇贸电子口岸有限公司 Distributed container cluster framework resources management method based on domestic CPU and operating system
CN108924217A (en) * 2018-06-29 2018-11-30 中山大学 A kind of distribution cloud system Automation arranging method
CN109067859A (en) * 2018-07-20 2018-12-21 北京航空航天大学 A kind of double-deck cloud framework system and implementation method towards cross-domain cooperation with service
CN110875833A (en) * 2018-08-31 2020-03-10 阿里巴巴集团控股有限公司 Cluster hybrid cloud, job processing method and device and electronic equipment
CN110057360A (en) * 2019-03-08 2019-07-26 江苏海事职业技术学院 A kind of paths planning method and its system based on Distributed Parallel Computing
CN110113406A (en) * 2019-04-29 2019-08-09 成都网阔信息技术股份有限公司 Based on distributed calculating service cluster frame
CN111614567A (en) * 2020-04-03 2020-09-01 杭州鹿扬科技有限公司 Node path selection method, system and device in block chain network
CN111614785A (en) * 2020-06-03 2020-09-01 成都智视慧语科技有限公司 Edge AI (Artificial Intelligence) computing cluster based on micro-container cloud
CN112055402A (en) * 2020-08-14 2020-12-08 杭州伍暨通信设备有限公司 Wireless sensor network node awakening system and method based on way finding algorithm

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113934545A (en) * 2021-12-17 2022-01-14 飞诺门阵(北京)科技有限公司 Video data scheduling method, system, electronic equipment and readable medium

Similar Documents

Publication Publication Date Title
Joshi et al. On the delay-storage trade-off in content download from coded distributed storage systems
KR101120847B1 (en) Data overlay, self-organized metadata overlay, and application level multicasting
CN107181724B (en) Identification method and system of cooperative flow and server using method
Tirado et al. Predictive data grouping and placement for cloud-based elastic server infrastructures
US20180131516A1 (en) Stable data-processing in a distributed computing environment
CN110798517A (en) Decentralized cluster load balancing method and system, mobile terminal and storage medium
Ng et al. A survey of coded distributed computing
Filip et al. Data capsule: Representation of heterogeneous data in cloud-edge computing
Kanwal et al. A genetic based leader election algorithm for IoT cloud data processing
Li et al. Data analytics for fog computing by distributed online learning with asynchronous update
Zhu et al. DR2: Dynamic request routing for tolerating latency variability in online cloud applications
CN112698944A (en) Distributed cloud computing system and method based on human brain simulation
Pius et al. Survey on load balancing in cloud computing
CN112925964A (en) Big data acquisition method based on cloud computing service and big data acquisition service system
Subbiah et al. Energy efficient big data infrastructure management in geo-federated cloud data centers
Dharmapriya et al. Smart platform for cloud service providers
Kolosov et al. Theory vs. practice in modeling edge storage systems
CN112925831A (en) Big data mining method and big data mining service system based on cloud computing service
Jayakumar et al. QoS Analysis for Cloud‐Based IoT Data Using Multicriteria‐Based Optimization Approach
Li et al. Esmo: Joint frame scheduling and model caching for edge video analytics
CN117596122B (en) General calculation fusion network architecture and resource adaptation method
Yeh et al. Bi-Objective simplified swarm optimization for fog computing task scheduling
Bruno et al. freeCycles-efficient multi-cloud computing platform
Chatziliadis et al. Efficient Placement of Decomposable Aggregation Functions for Stream Processing over Large Geo-Distributed Topologies
Datiri et al. Load Balancing for Resource Optimization in Internet of Things (IOT) Systems

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210423