An algorithm network scheduling method based on a soil intelligence algorithm platform, a storage medium, an equipment and a computer program product

Abstract

The application discloses a kind of based on Xiluocheng intelligent computing platform's algorithm network scheduling method, storage medium, equipment and computer program product, comprising: the operation data of each terminal equipment collected is transmitted to cloud through Xiluocheng intelligent computing platform, the network node that is passed through from each terminal equipment to cloud is obtained by tracking routing, the IP address of network node determines the intersection node of each terminal equipment in operation data transmission process;Calculate the idle computing power on intersection node, if the computing power required for operation data transmission to intersection node to carry out computing power calculation is less than the idle computing power of intersection node, after computing power calculation, continue to transmit to cloud;Otherwise, intersection node and its neighbor node are contracted to node, update intersection node, if the idle computing power of updated intersection node does not satisfy the computing power requirement, do not contract to node, intersection node introduces computing power space through star model, after computing power calculation to the operation data transmitted to intersection node, continue to transmit to cloud.

Description

Technical Field

[0001] This invention relates to the field of computing network scheduling technology, specifically to a computing network scheduling method, storage medium, device, and computer program product based on the Xirang intelligent computing platform. Background Technology

[0002] With the acceleration of global digital transformation and the rapid development of artificial intelligence technology, the demand for computing power has exploded, and the scale of computing power is also expanding. In the context of cloud computing, there is a tendency to build a centralized supercomputing pool to solve computing power problems. However, due to network latency and bandwidth limitations, the response speed of centralized processing will be slow. In addition, due to the huge amount of data involved in the task, a lot of energy will be consumed in the process of transmitting data to the supercomputing pool. Especially when transmitting over long distances, due to signal attenuation and transmission delay, more energy needs to be consumed to ensure the stability and reliability of the data.

[0003] To address the aforementioned issues, some studies have proposed computing network scheduling, which enables unified management and scheduling of computing and network resources, resolving the uneven distribution of computing resources and achieving efficient resource utilization and task processing. However, during computing network scheduling, the lack of management strategies for computing nodes leads to the abuse of computing nodes during data transmission, resulting in unnecessary resource waste and increased energy consumption. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a computing network scheduling method, storage medium, device, and computer program product based on the Xirang intelligent computing platform. By uniformly calculating computing power at the cross nodes of data transmission, it can realize the coordinated scheduling and flexible use of diverse computing resources, thereby reducing energy consumption.

[0005] To achieve the above technical objectives, the present invention adopts the following technical solution: a computing network scheduling method based on the Xirang intelligent computing platform, specifically including the following steps:

[0006] Step S1: The collected operating data of each terminal device is transmitted to the cloud through the Xirang Intelligent Computing Platform. The network nodes from each terminal device to the cloud are obtained by tracing the route, and the intersection nodes of each terminal device in the process of data transmission are determined according to the IP address of the network node.

[0007] Step S2: Calculate the idle computing power on the cross node. If the computing power required for the running data to be transmitted to the cross node for computing power calculation is less than the idle computing power of the cross node, the running data transmitted to the cross node will be processed for computing power calculation and then transmitted to the cloud.

[0008] Step S3: Otherwise, shrink the cross node and its neighboring nodes, update the cross node, and calculate whether the idle computing power of the updated cross node is sufficient to transmit the running data to the updated cross node for computing power calculation. If so, perform computing power calculation on the running data transmitted to the updated cross node and continue to transmit it to the cloud.

[0009] Step S4: If not, no node shrinkage is performed. The cross node introduces computing power space through the star model, performs computing power calculation on the running data transmitted to the cross node, and continues to transmit it to the cloud.

[0010] Furthermore, the calculation process for the idle computing power is as follows:

[0011]

[0012] Among them, C br Here, n represents the idle computing power, i is the number of logic operation chips, and f(a) represents the idle computing power. i ) is the mapping function for logical operations, α i Let be the mapping ratio coefficient of the i-th logic operation chip, q1(TOPS) be the redundant computing power of the logic operation; m be the number of parallel computing chips, j be the index of m, and f(b) be the mapping ratio coefficient of the i-th logic operation chip. j ) is the mapping function for parallel computation, β j Let be the mapping ratio coefficient of the j-th parallel computing chip, q2(FLOPS) be the redundant computing power of parallel computing; p be the number of neural network acceleration chips, k be the index of p, and f(c k ) is the mapping function for accelerating neural networks, γ k q3(FLOPS) is the mapping ratio coefficient of the k-th neural network acceleration chip, and q3(FLOPS) is the redundant computing power of neural network acceleration.

[0013] Furthermore, the specific process of shrinking the cross node and its neighboring nodes in step S3 is as follows: calculate the node similarity between the cross node and each neighboring node; if the node similarity exceeds 60%, merge the cross node with the corresponding neighboring node, update the cross node, and broadcast the updated cross node.

[0014] Furthermore, the calculation process for the node similarity is as follows:

[0015]

[0016] Among them, S ij T represents the similarity between the intersection node i and its neighbor node j. i T represents all neighboring nodes of the intersection node i. j This represents all neighboring nodes of neighboring node j.

[0017] Furthermore, the specific process of introducing computing power space into the cross node through the Starry Sky Big Model in step S4 is as follows: the IP address, idle computing power, and running data of the cross node are transmitted to the Starry Sky Big Model to input the computing power required for computing power calculation of the cross node. The Starry Sky Big Model finds the corresponding manufacturer of the cross node based on the input IP address, and searches for computing power space with a value greater than the computing power difference between the required computing power and the idle computing power in the same manufacturer based on the computing power difference. The Starry Sky Big Model then connects to the cross node through the API interface.

[0018] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program that enables a computer to execute the aforementioned network scheduling method based on the Xirang intelligent computing platform.

[0019] Furthermore, the present invention also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the aforementioned network scheduling method based on the Xirang intelligent computing platform.

[0020] Furthermore, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned network scheduling method based on the Xirang intelligent computing platform.

[0021] Compared with existing technologies, this invention has the following beneficial effects: Based on the computing network scheduling method of the Xirang Intelligent Computing Platform, this invention performs computing power calculations at cross nodes, enabling diversified collaborative scheduling and flexible use of computing resources, improving overall computing power utilization efficiency and business flexibility. Simultaneously, it reduces the time for data to travel to and from the cloud, thereby reducing network latency, alleviating the computing pressure on the cloud, and enabling the cloud to process other tasks more efficiently. Furthermore, to ensure that the idle computing power of cross nodes meets the computing power calculations for running data, this invention shrinks nodes through node similarity calculations, thereby merging similar nodes to reduce noise and errors, improving calculation accuracy and expanding the idle computing power of cross nodes. This allows for more effective utilization of computing resources in network nodes and reduces energy consumption. Further, this invention can also expand the computing power space of cross nodes through the Starry Sky Big Model, reducing dependence on hardware resources, thereby reducing hardware procurement and maintenance costs. Moreover, the Starry Sky Big Model can better protect the security of running data in cross nodes. Attached Figure Description

[0022] Figure 1 This is a flowchart of the computing network scheduling method based on the Xirang intelligent computing platform of the present invention;

[0023] Figure 2 This is a schematic diagram illustrating the introduction of computing power space at intersection nodes using the Starry Sky Model. Detailed Implementation

[0024] The technical solution of the present invention will be further explained and described below with reference to the accompanying drawings.

[0025] like Figure 1 This is a flowchart of the computing network scheduling method based on the Xirang intelligent computing platform of the present invention. The computing network scheduling method specifically includes the following steps:

[0026] Step S1: The collected operating data of each terminal device is transmitted to the cloud through the Xirang Intelligent Computing Platform. The network nodes from each terminal device to the cloud are obtained by tracing the route, and the intersection nodes of each terminal device in the process of data transmission are determined according to the IP address of the network node.

[0027] Step S2: Calculate the idle computing power on the cross node. If the computing power required for the running data to be transmitted to the cross node for computing power calculation is less than the idle computing power of the cross node, the running data transmitted to the cross node will be processed for computing power calculation before being transmitted to the cloud. Computing power calculation at the cross node can realize diversified collaborative scheduling and flexible use of computing power resources, improve the overall computing power utilization efficiency and business flexibility, and at the same time, reduce the time for running data to travel to and from the cloud, thereby reducing network latency, alleviating the computing pressure on the cloud, and enabling the cloud to process other tasks more efficiently.

[0028] The calculation process for idle computing power in this invention is as follows:

[0029]

[0030] Among them, C br Here, n represents the idle computing power, i is the number of logic operation chips, and f(a) represents the idle computing power. i ) is the mapping function for logical operations, α i Let be the mapping ratio coefficient of the i-th logic operation chip, q1(TOPS) be the redundant computing power of the logic operation; m be the number of parallel computing chips, j be the index of m, and f(b) be the mapping ratio coefficient of the i-th logic operation chip. j ) is the mapping function for parallel computation, β j Let be the mapping ratio coefficient of the j-th parallel computing chip, q2(FLOPS) be the redundant computing power of parallel computing; p be the number of neural network acceleration chips, k be the index of p, and f(c k ) is the mapping function for accelerating neural networks, γ k q3(FLOPS) is the mapping ratio coefficient of the k-th neural network acceleration chip, and q3(FLOPS) is the redundant computing power of neural network acceleration.

[0031] Step S3: Otherwise, shrink the cross node and its neighboring nodes, and update the cross node. Specifically, calculate the node similarity between the cross node and each neighboring node. If the node similarity exceeds 60%, merge the cross node with the corresponding neighboring node, update the cross node, and broadcast the updated cross node. Calculate whether the idle computing power of the updated cross node is sufficient to transmit the running data to the updated cross node for computing power calculation. If so, perform computing power calculation on the running data transmitted to the updated cross node and continue transmitting it to the cloud. By calculating node similarity and shrinking nodes, similar nodes are merged to reduce noise and errors, thereby improving calculation accuracy and expanding the idle computing power of cross nodes. This allows for more efficient utilization of computing resources in network nodes and reduces energy consumption.

[0032] In this invention, node similarity is determined by calculating the degree of overlap between the neighboring nodes of an intersecting node and the neighboring nodes of its neighboring nodes. The specific calculation process is as follows:

[0033]

[0034] Among them, S ij T represents the similarity between the intersection node i and its neighbor node j. i T represents all neighboring nodes of the intersection node i. j This represents all neighboring nodes of neighboring node j.

[0035] Step S4: If not, do not perform node shrinkage. The intersecting nodes are introduced into the computing space through the Starry Sky large model, such as... Figure 2 The system transmits the IP address, idle computing power, and running data of the cross-nodes to the Starry Sky Big Data Model for computing power calculation. The Starry Sky Big Data Model identifies the corresponding vendor for each cross-node based on the input IP address. Then, based on the difference between the required computing power and the idle computing power, it searches for computing power space greater than the difference among vendors. This space is accessed through an API interface to the cross-node. After computing power calculation is performed on the running data transmitted to the cross-node, it is transmitted to the cloud. Expanding the computing power space of cross-nodes through the Starry Sky Big Data Model enables intelligent identification of computing power space from the same vendor, avoiding resource waste, improving overall resource utilization, and reducing dependence on hardware resources, thereby reducing hardware procurement and maintenance costs. Furthermore, the Starry Sky Big Data Model better protects the security of the running data within the cross-nodes.

[0036] The computing network scheduling method based on the Xirang intelligent computing platform of this invention performs computing power calculations on the cross nodes of the transmission network by collecting and transmitting the operation data to the cloud. This enables diversified collaborative scheduling and flexible use of computing power resources, improves the overall computing power utilization efficiency and business flexibility, and reduces the time for data to travel to and from the cloud, thereby reducing network latency, alleviating the computing pressure on the cloud, and enabling the cloud to process other tasks more efficiently.

[0037] In one technical solution of the present invention, a computer-readable storage medium is also provided, which stores a computer program that enables a computer to execute the network scheduling method based on the Xirang intelligent computing platform.

[0038] In one technical solution of the present invention, an electronic device is also provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the computing network scheduling method based on the Xirang intelligent computing platform.

[0039] In one technical solution of the present invention, a computer program product is also provided, including a computer program, which, when executed by a processor, implements the computing network scheduling method based on the Xirang intelligent computing platform.

[0040] In the embodiments disclosed in this application, a computer storage medium may be a tangible medium that may contain or store programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer storage media include electrical connections based on 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, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0041] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0042] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A computing network scheduling method based on the Xirang intelligent computing platform, characterized in that, Specifically, the steps include the following: Step S1: The collected operating data of each terminal device is transmitted to the cloud through the Xirang Intelligent Computing Platform. The network nodes from each terminal device to the cloud are obtained by tracing the route, and the intersection nodes of each terminal device in the process of data transmission are determined according to the IP address of the network node. Step S2: Calculate the idle computing power on the cross node. If the computing power required for the running data to be transmitted to the cross node for computing power calculation is less than the idle computing power of the cross node, the running data transmitted to the cross node will be processed for computing power calculation and then transmitted to the cloud. Step S3: Otherwise, shrink the cross node and its neighboring nodes, update the cross node, and calculate whether the idle computing power of the updated cross node is sufficient to transmit the running data to the updated cross node for computing power calculation. If so, perform computing power calculation on the running data transmitted to the updated cross node and continue to transmit it to the cloud. Step S4: If not, no node shrinkage is performed. The cross node is introduced into the computing space through the star model. After the running data transmitted to the cross node is calculated, it continues to be transmitted to the cloud. The specific process of introducing computing power space into the cross node through the Starry Sky Big Model is as follows: the IP address, idle computing power, and running data of the cross node are transmitted to the Starry Sky Big Model to input the computing power required for computing power calculation of the cross node. The Starry Sky Big Model finds the corresponding manufacturer of the cross node based on the input IP address, and searches for computing power space with a value greater than the computing power difference between the required computing power and the idle computing power in the same manufacturer based on the computing power difference. The cross node is then connected through the API interface.

2. The computing network scheduling method based on the Xirang intelligent computing platform according to claim 1, characterized in that, The calculation process for the idle computing power 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.

3. The computing network scheduling method based on the Xirang intelligent computing platform according to claim 1, characterized in that, The specific process of shrinking the intersection node and its neighboring nodes in step S3 is as follows: calculate the node similarity between the intersection node and each neighboring node. If the node similarity exceeds 60%, merge the intersection node with the corresponding neighboring node, update the intersection node, and broadcast the updated intersection node.

4. The computing network scheduling method based on the Xirang intelligent computing platform according to claim 3, characterized in that, The calculation process for the node similarity is as follows: in, Indicates the intersection node i with neighboring nodes j similarity, Indicates the intersection node i All neighboring nodes, Representing neighboring nodes j All neighboring nodes.

5. A computer-readable storage medium storing a computer program, characterized in that, The computer program causes the computer to execute the network scheduling method based on the Xirang intelligent computing platform as described in any one of claims 1-4.

6. 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 when the processor executes the computer program, it implements the network scheduling method based on the Xirang intelligent computing platform as described in any one of claims 1-4.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the network scheduling method based on the Xirang intelligent computing platform as described in any one of claims 1-4.

Images