An intelligent cloud network scheduling method, storage medium, device and computer program product

Abstract

The application discloses an intelligent cloud network scheduling method, a storage medium, equipment and a computer program product, which comprises the following steps: finding a same type user group and a same region index network node in a cross-domain network according to log information, storing the network node into a node list, and marking the state of the network node; exploring the network node in the node list by using a path exploration algorithm, forming a plurality of links, and finding an optimal link; preferentially using the optimal link for data forwarding, and distributing a calculation task in the data to an interworking network node for calculation; and performing calculation along with transmission for the calculation task, so that network node congestion caused by the concentration of the calculation task is avoided.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of network intelligent terminal scheduling, and particularly relates to an intelligent cloud network scheduling method, a storage medium, equipment and a computer program product. BACKGROUND

[0002] With the rapid development of the information age, intelligent cloud networks also become increasingly complex, and as a control component in the cloud network, as the central brain of the scheduling in the entire cloud network, it will become crucial. At the same time, through network control intelligent discovery of idle resources in the network and utilization to improve network resource utilization.

[0003] In the process of improving network resource utilization by using intelligent network scheduling in a complex network, the process of exploring and evaluating multiple possible network data transmission nodes and finding the optimal process needs to explore the resources wasted by traversing each branch. At the same time, the centralized computation of data with computation tasks to the destination node in the data transmission process will cause the problem of network congestion due to excessive node pressure. SUMMARY

[0004] In view of the problems in the prior art, the application provides an intelligent cloud network scheduling method, a storage medium, equipment and a computer program product, which performs computation on the fly to avoid network node congestion caused by concentrated computation tasks.

[0005] To achieve the above technical purpose, the application adopts the following technical scheme: an intelligent cloud network scheduling method, specifically comprising the following steps:

[0006] Step S1, finding network nodes of the same type of user group and the same regional index in a cross-domain network according to log information, storing the network nodes into a node list, and marking the state of the network nodes;

[0007] Step S2, exploring the network nodes in the node list using a path exploration algorithm to form a plurality of links and find an optimal link;

[0008] Step S3, preferentially using the optimal link for data forwarding, and distributing the computation tasks in the data to the interworking network nodes for computation.

[0009] Further, executing a ping command on the network nodes in the node list to confirm the connectivity to the destination network node, if connected, marking the network node state as available, otherwise, marking the network node state as unavailable.

[0010] Further, the specific process of exploring the network nodes in the node list using the path exploration algorithm to form a plurality of links in step S2 is as follows:

[0011] i. Determine the starting point and the destination network node of the cloud network scheduling;

[0012] ii. Find the network node interworking with the starting point of the cloud network scheduling from the network nodes available in the node list;

[0013] iii. Continue to find the network node interworking with the network node found in step ii;

[0014] iv. Repeat step iii until interworking with the destination network node, and obtain a plurality of links.

[0015] Further, in the process of exploration by using the path exploration algorithm, the number of interworking between each network node found and the neighbor node is recorded, and the total number of interworking between all network nodes and the neighbor nodes in each link is counted, and the link with the maximum total number is taken as the optimal link.

[0016] Further, the specific process of step S3 of distributing the calculation task to the interworking network node for calculation is: determining the idle computing power on the interworking network node, dividing the calculation task into sub-computation tasks according to the idle computing power, distributing the sub-computation tasks to the corresponding interworking network node for calculation, and transmitting the calculation result from the interworking network node to the destination network node.

[0017] Further, the calculation process of the idle computing power on the interworking network node is:

[0018]

[0019] Wherein, C br is the idle computing power, n is the number of logical operation chips, i is the index of n, f(a i ) is the mapping function of logical operation, alpha i is the mapping proportion coefficient of the i th logical operation chip, q1(TOPS) is the redundant computing power of logical operation; m is the number of parallel computing chips, j is the index of m, f(b j ) is the mapping function of parallel computing, beta j is the mapping proportion coefficient of the j th parallel computing chip, q2(FLOPS) is the redundant computing power of parallel computing; p is the number of neural network acceleration chips, k is the index of p, f(c k ) is the mapping function of neural network acceleration, gamma k is the mapping proportion coefficient of the k th neural network acceleration chip, q3(FLOPS) is the redundant computing power of neural network acceleration.

[0020] Further, the present application also provides a computer readable storage medium storing a computer program, wherein the computer program enables a computer to execute the intelligent cloud network scheduling method.

[0021] Further, the present application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the intelligent cloud network scheduling method when executing the computer program.

[0022] Further, the present application also provides a computer program product, comprising a computer program, wherein the computer program implements the intelligent cloud network scheduling method when executed by a processor.

[0023] Compared with the prior art, the present application has the following beneficial effects: the intelligent cloud network scheduling method of the present application can optimize the path and speed of data transmission by finding available network nodes and constructing a link therefrom, which helps to reduce network delay, improve bandwidth utilization, and reduce transmission error, and the link with the largest number of interworking nodes is taken as the optimal link, so that the computing task is allocated to the interworking network nodes for calculation, improving the operation efficiency and relieving the pressure caused by centralized operation of the nodes when the operation task reaches the destination node, avoiding network node downtime, and improving user experience. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is a flowchart of the intelligent cloud network scheduling method of the present application;

[0025] Figure 2 is a path exploration diagram for the path exploration algorithm in the present application. DETAILED DESCRIPTION

[0026] The technical solutions of the present application will be further explained in combination with the accompanying drawings.

[0027] As Figure 1 is a flowchart of the intelligent cloud network scheduling method of the present application, which specifically comprises the following steps:

[0028] Step S1: In the cross-domain network, find network nodes of the same type of user group and the same regional index according to log information, store them in a node list, and mark the state of the network nodes, specifically, execute a ping command for the network nodes in the node list to confirm the connectivity to the destination network node, if connected, mark the network node state as available, otherwise, mark the network node state as unavailable. This avoids traversing each branch of the multi-path exploration, only traversing the nodes with available states, reducing resource waste and lowering the algorithm complexity.

[0029] Step S2, the network nodes in the node list are explored by using a path exploration algorithm to form several links, and an optimal link is found; when the optimal path is selected to transmit data of a computing task, since the passing nodes have the most neighbor intercommunication nodes, the passing nodes can undertake a part of the computing task, thereby reducing the computing pressure of the passing nodes and the pressure of the centralized computing of the destination node.

[0030] In the present application, the network nodes in the node list are explored by using a path exploration algorithm to form several links, and the specific process is as follows:

[0031] i. determining the starting point and the destination network node of cloud network scheduling;

[0032] ii. searching for the network nodes intercommunicating with the starting point of cloud network scheduling from the network nodes available in the node list;

[0033] iii. continuing to search for the network nodes intercommunicating with the searched network nodes;

[0034] iv. repeating step iii until the destination network node is reached, and several links are obtained.

[0035] In the process of exploration by using the path exploration algorithm, the number of neighbor nodes intercommunicating with each searched network node is recorded, and the total number of neighbor nodes intercommunicating with all network nodes in each link is counted, and the link with the largest total number is taken as the optimal link. In order to maximize the resource utilization, the intercommunication nodes in the optimal link need to undertake part of the data computing task, and the more the number of neighbor nodes of the passing nodes, the greater the possibility of sharing the data computing task.

[0036] Step S3, preferentially using the optimal link for data forwarding, and for the computing task in the data, the computing task is transmitted to the intercommunication network nodes for calculation, and the specific process is as follows: determining the idle computing power of the intercommunication network nodes, dividing the computing task into sub-computing tasks according to the idle computing power, distributing the sub-computing tasks to the corresponding intercommunication network nodes for calculation, and transmitting the calculation results from the intercommunication network nodes to the destination network nodes. By sharing the data computing task by the neighbor nodes, the computing pressure of the passing nodes can be directly reduced, and the pressure of the centralized computing of the destination node can be indirectly reduced.

[0037] The calculation process of the idle computing power of the intercommunication network nodes is as follows:

[0038]

[0039] Wherein, C br is the idle computing power, n is the number of logical operation chips, i is the index of n, f(a i ) is a mapping function of logical operation, and a iis the mapping proportion coefficient of the i-th logical operation chip, q1(TOPS) is the redundant computing power of the logical operation; m is the number of parallel computing chips, j is the index of m, f(b j ) is the mapping function of parallel computing, β j is the mapping proportion coefficient of the j-th parallel computing chip, q2(FLOPS) is the redundant computing power of the parallel computing; p is the number of neural network acceleration chips, k is the index of p, f(c k ) is the mapping function of neural network acceleration, γ k is the mapping proportion coefficient of the k-th neural network acceleration chip, q3(FLOPS) is the redundant computing power of the neural network acceleration.

[0040] In one of the technical solutions of the present application, a computer readable storage medium is also provided, which stores a computer program, and the computer program causes a computer to execute the intelligent cloud network scheduling method.

[0041] In one of the technical solutions of the present application, an electronic device is also provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the intelligent cloud network scheduling method is implemented.

[0042] In one of the technical solutions of the present application, a computer program product is also provided, which includes a computer program, and when the computer program is executed by a processor, the intelligent cloud network scheduling method is implemented.

[0043] In the embodiments disclosed in the present application, the computer storage medium can be a tangible medium, which can contain or store programs for use by or in conjunction with an instruction execution system, device or apparatus. The computer storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or apparatus, or any suitable combination of the above. More specific examples of computer storage media can include one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0044] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present application can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software manner depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0045] The above is only the preferred embodiment of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solution falling within the concept of the present application shall fall within the protection scope of the present application. It should be noted that, for ordinary skilled persons in the art, some improvements and refinements without departing from the principles of the present application shall be considered as the protection scope of the present application.

Claims

1. A method for intelligent cloud network scheduling, characterized in that, Specifically, the steps include the following: Step S1: In the cross-domain network, find network nodes with the same type of user group and the same regional indicators based on log information, store them in the node list, and mark the status of the network nodes; execute the ping command on the network nodes in the node list to confirm the connectivity to the destination network node. If the connectivity is successful, mark the network node status as available; otherwise, mark the network node status as unavailable. Step S2: Explore the network nodes in the node list using a path exploration algorithm to form several links, and find the optimal link; specifically: i. Determine the starting point and destination network node for cloud network scheduling; ii. Find the network node that is interconnected with the starting point of cloud network scheduling from the network nodes in the node list that are available in the current state; iii. Continue searching for network nodes that can communicate with the previously found network nodes; iv. Repeat step iii until communication is established with the target network node, resulting in several links; v. Record the number of times each network node communicates with its neighboring nodes, and count the total number of times all network nodes in each link communicate with their neighboring nodes. Select the link with the largest total number of communication as the optimal link. Step S3: Prioritize using the optimal link for data forwarding. For computational tasks in the data, assign the computational tasks to interconnected network nodes for computation.

2. The intelligent cloud network scheduling method according to claim 1, characterized in that, Step S3, which assigns computing tasks to interconnected network nodes, involves determining the available computing power on the interconnected network nodes, dividing the computing tasks into sub-computing tasks based on the available computing power, assigning the sub-computing tasks to the corresponding interconnected network nodes for computation, and transmitting the computation results from the interconnected network nodes to the destination network node.

3. The intelligent cloud network scheduling method according to claim 2, characterized in that, The calculation process for idle computing power on interconnected network nodes is as follows: in, For idle computing power, n The number of logic operation chips. i for n index, f ( a i ) is a mapping function for logical operations. α i For the first i The mapping ratio of each logic operation chip. q 1 (TOPS) represents the redundant computing power for logical operations; m To increase the number of chips used for parallel computing, j for m index, f ( b j ) is a mapping function for parallel computation. β j For the first j The mapping ratio of each parallel computing chip. q 2 (FLOPS) represents the redundant computing power for parallel computing; p The number of neural network acceleration chips, k for p index, f ( c k ) is a mapping function for accelerating neural networks. γ k For the first k The mapping ratio of a neural network acceleration chip. q 3 (FLOPS) represents redundant computing power for accelerating neural networks.

4. A computer-readable storage medium storing a computer program, characterized in that, The computer program causes the computer to execute the intelligent cloud network scheduling method as described in any one of claims 1-3.

5. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the intelligent cloud network scheduling method as described in any one of claims 1-3.

6. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent cloud network scheduling method according to any one of claims 1-3.

Images