Search

CN-122019454-A - Data transmission method, system, equipment and computer readable storage medium

CN122019454ACN 122019454 ACN122019454 ACN 122019454ACN-122019454-A

Abstract

The application discloses a data transmission method, a system, equipment and a computer readable storage medium, which relate to the technical field of network-on-chip, are applied to a routing node of the network-on-chip, determine a source node for transmitting target data and a destination node for receiving the target data, determine a node routing direction according to the position relationship between the source node and the destination node, screen a next node according to the node routing direction according to the position relationship between the routing node and the destination node, transmit the target data to the next node through a conventional channel of the routing node in response to the fact that the same value exists between the coordinates of the next node and the coordinates of the routing node, and transmit the target data to the next node through a virtual channel of the routing node in response to the fact that the same value does not exist between the coordinates of the next node and the coordinates of the routing node. On the premise of not changing the network-on-chip architecture system, the routing algorithm is modified through the software layer to reduce the transmission congestion, and the data transmission efficiency of the network-on-chip is improved.

Inventors

  • TAN MAOJU

Assignees

  • 山东云海国创云计算装备产业创新中心有限公司

Dates

Publication Date
20260512
Application Date
20260129

Claims (10)

  1. 1. A method of data transmission, characterized in that it is applied to a routing node of a network on chip, comprising: determining a source node for transmitting target data and a destination node for receiving the target data; determining a node routing direction according to the position relationship between the source node and the destination node; Screening the next node according to the node routing direction according to the position relation between the routing node and the destination node; transmitting the target data to the next node through a conventional channel of the routing node in response to the existence of the same value between the coordinates of the next node and the coordinates of the routing node; transmitting the target data to the next node through a virtual channel of the routing node in response to the fact that the same value does not exist between the coordinates of the next node and the coordinates of the routing node; The topology structure of the network on chip is a two-dimensional network structure, the virtual channel is used for transmitting data in four directions of northeast, southeast, northwest and southwest, and the conventional channel is used for transmitting data in four directions of east, southwest, west and northwest.
  2. 2. The method of claim 1, wherein determining the node routing direction based on the positional relationship between the source node and the destination node comprises: detecting whether the same value exists between the coordinates of the source node and the coordinates of the destination node; determining a target area formed by the source node and the destination node in response to the fact that the same value does not exist between the coordinates of the source node and the coordinates of the destination node, and generating a node routing direction in the target area by taking the destination node as a reference; responding to the fact that the ordinate of the source node is identical to the ordinate of the destination node, and in a first setting range, carrying out fluctuation on the ordinate of the destination node to generate a node routing direction; And responding to the fact that the abscissa of the source node is the same as the abscissa of the destination node, and in a second setting range, fluctuating the abscissa of the destination node to generate the node routing direction.
  3. 3. The method of claim 1, wherein the step of screening the next node according to the node routing direction based on the positional relationship between the routing node and the destination node comprises: Responding to the difference value between the abscissa of the routing node and the abscissa of the destination node being 1, and the difference value between the ordinate of the routing node and the ordinate of the destination node being 1, determining the destination node as the next node; Responding to that the abscissa of the routing node is the same as the abscissa of the destination node, and the difference value between the ordinate of the routing node and the ordinate of the destination node is 1, and determining the destination node as the next node; and in response to the fact that the ordinate of the routing node is the same as the ordinate of the destination node, and the difference value between the abscissa of the routing node and the abscissa of the destination node is 1, determining the destination node as the next node.
  4. 4. The method of claim 1, wherein the step of screening the next node according to the node routing direction based on the positional relationship between the routing node and the destination node comprises: Responding to the difference value between the abscissa of the routing node and the abscissa of the destination node is less than or equal to 1, and the difference value between the ordinate of the routing node and the ordinate of the destination node is greater than 1, and taking the adjacent node which is positioned on the vertical line of the routing node and is close to the destination node as a first candidate node according to the routing direction of the node; detecting whether the first candidate node is congested; In response to the first candidate node being normal, determining the first candidate node as the next node; and responding to the congestion of the first candidate node, and taking the adjacent node which is positioned on the diagonal line of the routing node and is close to the destination node as the next node according to the routing direction of the node.
  5. 5. The method of claim 1, wherein the step of screening the next node according to the node routing direction based on the positional relationship between the routing node and the destination node comprises: Responding to the difference value between the ordinate of the routing node and the ordinate of the destination node is less than or equal to 1, and the difference value between the abscissa of the routing node and the abscissa of the destination node is greater than 1, and taking the adjacent node which is positioned on the horizontal line of the routing node and is close to the destination node as a second candidate node according to the routing direction of the node; detecting whether the second candidate node is congested; In response to the second candidate node being normal, determining the second candidate node as the next node; and responding to the congestion of the second candidate node, and taking the adjacent node which is positioned on the diagonal line of the routing node and is close to the destination node as the next node according to the routing direction of the node.
  6. 6. The method of claim 1, wherein the step of screening the next node according to the node routing direction based on the positional relationship between the routing node and the destination node comprises: Responding to the fact that the difference value between the abscissa of the routing node and the abscissa of the destination node is greater than 1, and the difference value between the ordinate of the routing node and the ordinate of the destination node is greater than 1, acquiring a direction zone bit of the routing node, wherein the direction zone bit is used for representing the direction of preferentially detecting congestion; determining a first detection direction corresponding to the direction marker bit; According to the node routing direction, taking an adjacent node which is positioned in the first detection direction of the routing node and is close to the destination node as a third candidate node; detecting whether the third candidate node is congested; in response to the third candidate node being normal, the third candidate node is determined to be the next node.
  7. 7. The method of claim 6, wherein after detecting whether the third candidate node is congested, further comprising: Responding to congestion of the third candidate node, and taking a direction perpendicular to the first detection direction as a second detection direction; According to the node routing direction, taking an adjacent node which is positioned in the second detection direction of the routing node and is close to the destination node as a fourth candidate node; detecting whether the fourth candidate node is congested; In response to the fourth candidate node being normal, determining the fourth candidate node as the next node; And responding to the congestion of the fourth candidate node, and taking the adjacent node which is positioned on the diagonal line of the routing node and is close to the destination node as the next node according to the routing direction of the node.
  8. 8. A data transmission system, characterized by a routing node for a network on chip, comprising: The node determining module is used for determining a source node for transmitting target data and a destination node for receiving the target data; The routing direction determining module is used for determining the routing direction of the node according to the position relationship between the source node and the destination node; the next node screening module is used for screening the next node according to the position relation between the routing node and the destination node and the routing direction of the node; The conventional channel transmission module is used for transmitting the target data to the next node through a conventional channel of the routing node in response to the fact that the same value exists between the coordinates of the next node and the coordinates of the routing node; The virtual channel transmission module is used for transmitting the target data to the next node through the virtual channel of the routing node in response to the fact that the same value does not exist between the coordinates of the next node and the coordinates of the routing node; The topology structure of the network on chip is a two-dimensional network structure, the virtual channel is used for transmitting data in four directions of northeast, southeast, northwest and southwest, and the conventional channel is used for transmitting data in four directions of east, southwest, west and northwest.
  9. 9. An electronic device, comprising: A memory for storing a computer program; Processor for implementing the steps of the data transmission method according to any one of claims 1 to 7 when executing said computer program.
  10. 10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the data transmission method according to any one of claims 1 to 7.

Description

Data transmission method, system, equipment and computer readable storage medium Technical Field The present application relates to the field of network on chip technology, and more particularly, to a data transmission method, system, device, and computer readable storage medium. Background Currently, central processors (Central Processing Unit, CPUs) may employ multi-core processor architectures, and as the number of cores increases, the need for inter-core scalable, low-latency, high-bandwidth communication structures increases. Based on this, network-on-Chip (NoC) as a communication architecture for aiding in realizing design isolation is becoming a technology for solving the performance bottleneck of the bus structure, which realizes the communication of each IP plug and play through a standard interface. Among existing network-on-chip routing algorithms, XY routing algorithms are widely used for their simplicity and efficiency. However, these algorithms generally only support data transmission in four directions, namely southeast, northwest and northwest, and lack of awareness of node congestion status, which easily causes congestion problems of some key nodes due to excessive traffic, and further causes a decrease in data transmission efficiency of the network on chip. In summary, how to improve the data transmission efficiency of the network on chip is a problem to be solved by those skilled in the art. Disclosure of Invention The application aims to provide a data transmission method which can solve the technical problem of how to improve the data transmission efficiency of a network on chip to a certain extent. The application also provides a data transmission system, electronic equipment and a computer readable storage medium. In order to achieve the above object, the present application provides the following technical solutions: A data transmission method is applied to a routing node of a network on chip, and comprises the following steps: determining a source node for transmitting target data and a destination node for receiving the target data; determining a node routing direction according to the position relationship between the source node and the destination node; Screening the next node according to the node routing direction according to the position relation between the routing node and the destination node; transmitting the target data to the next node through a conventional channel of the routing node in response to the existence of the same value between the coordinates of the next node and the coordinates of the routing node; transmitting the target data to the next node through a virtual channel of the routing node in response to the fact that the same value does not exist between the coordinates of the next node and the coordinates of the routing node; The topology structure of the network on chip is a two-dimensional network structure, the virtual channel is used for transmitting data in four directions of northeast, southeast, northwest and southwest, and the conventional channel is used for transmitting data in four directions of east, southwest, west and northwest. In an exemplary embodiment, the determining the node routing direction according to the position relationship between the source node and the destination node includes: detecting whether the same value exists between the coordinates of the source node and the coordinates of the destination node; determining a target area formed by the source node and the destination node in response to the fact that the same value does not exist between the coordinates of the source node and the coordinates of the destination node, and generating a node routing direction in the target area by taking the destination node as a reference; responding to the fact that the ordinate of the source node is identical to the ordinate of the destination node, and in a first setting range, carrying out fluctuation on the ordinate of the destination node to generate a node routing direction; And responding to the fact that the abscissa of the source node is the same as the abscissa of the destination node, and in a second setting range, fluctuating the abscissa of the destination node to generate the node routing direction. In an exemplary embodiment, the screening the next node according to the node routing direction according to the position relationship between the routing node and the destination node includes: Responding to the difference value between the abscissa of the routing node and the abscissa of the destination node being 1, and the difference value between the ordinate of the routing node and the ordinate of the destination node being 1, determining the destination node as the next node; Responding to that the abscissa of the routing node is the same as the abscissa of the destination node, and the difference value between the ordinate of the routing node and the ordinate of the destination node is 1, and determining the destinatio