A network controller and a network control method

Abstract

This invention provides a network controller and network control method. The structure includes: a locking block, an insulating box, a control body, a power supply slot, a data connection structure, and a heat dissipation end. The locking block is fixed to the left and right sides of the insulating box, which covers the outside of the control body. The power supply slot and the data connection structure penetrate the insulating box and form an integral part with the control body. The heat dissipation end is located at the upper end of the control body. This invention improves the data connection structure by using a locking ring to firmly connect the flat plate to the surface of the insulating box. Then, a central restraining rod of a reinforcing frame is embedded in the center of the data cable spool connector for restraint. Simultaneously, the adjusting parts of the reinforcing frame are clamped symmetrically, ensuring a secure connection to the data cable spool connector in conjunction with the central restraining rod. This prevents automatic detachment during long-term use, facilitating disassembly and assembly, and improving the actual stability of the network controller.

Description

Technical Field

[0001] This invention relates to the field of network control technology, and more specifically to a network controller and a network control method. Background Technology

[0002] Network controllers are key components in computer networks, responsible for managing and coordinating data traffic, device communication, and policy enforcement. While their specific meaning varies across contexts, their core function is centralized control of network behavior, enabling networks to become more intelligent, flexible, and efficient. Whether in SDN, Wi-Fi, or 5G, controllers play a crucial role and are a key technology for building next-generation adaptive networks. In summary, the inventors have found that existing network controllers have the following main defects: Because the power and data cable spools of current network controllers are typically connected using a standard threading method, an assembly gap still exists between the spool and the controller after this method. This gap makes them prone to dust intrusion during long-term use and slow automatic disconnection due to various external factors (such as continuous vibration of the host device or minor geological disasters). Consequently, they cannot operate stably for extended periods, requiring regular maintenance and repair of the interfaces, which reduces the long-term stability of the current network controller. Summary of the Invention

[0003] The technical solution adopted by the present invention to achieve the technical objective is: a network controller and a network control method, the structure of which includes: a locking block, an insulating box, a control body, a power supply slot, a data connection structure, and a heat dissipation end. The locking block is fixed on the left and right sides of the insulating box and the insulating box covers the outside of the control body. The power supply slot and the data connection structure penetrate the insulating box and form an integral part with the control body. The heat dissipation end is located at the upper end of the control body.

[0004] As a further improvement of the present invention, the data connection structure is provided with a locking ring, which is located on the upper edge of the plate. The upper surface of the plate is equipped with a reinforcing frame and has a slot and a central restraining rod inside.

[0005] As a further improvement of the present invention, the control method of the network controller is as follows: Its setting: Objective: Host A (10.0.0.1) sends a data packet (e.g., a TCP SYN packet) to host B (10.0.0.2) for the first time. Network: A simple linear topology: Host A - Switch S1 - Switch S2 - Host B; Controller: A centralized SDN controller (such as Ryu or ONOS) that establishes OpenFlow connections with both S1 and S2; Initial state: All switches have empty flow tables; Its control and processing method will consist of the following stages: Phase 1: Network Discovery and Topology Building (Initialization, Continuous Operation); Phase Two: Initial processing of the data stream (the core of reaction control); Phase 3: High-speed forwarding of subsequent traffic.

[0006] As a further improvement of the present invention, the locking block is symmetrically arranged at the edge of the insulating box and the shape of the insulating box matches that of the control body. The power supply slot of the control body is larger than the size of the data connection structure, and the heat dissipation end is set at the upper end of the control body in a parallel orientation.

[0007] As a further improvement of the present invention, the locking ring is circular in shape and has an internal threaded groove, the slot of the reinforcing frame is set in a vertical position, and the central restraining rod is a silicone cylinder.

[0008] As a further improvement of the present invention, the reinforcing frame is provided with a welding layer, which is integral with the lower edge of the frame. An adjustment component is mounted on the surface of the frame edge, which passes through the edge of the frame and communicates with the power-conducting chip and clamping ring provided inside.

[0009] As a further improvement of the present invention, the frame is parallel to the plate through a welding layer, the adjusting parts of the frame are set in a symmetrical orientation, and the center position of the powered chip is determined by a clamping ring to determine the position of the central restraining rod.

[0010] As a further improvement of the present invention, the adjusting member is further provided with a control block, the lower end of the control block is connected to a screw, the lower end of the screw passes through the center of the positioning block and is connected to the inside of the rotating sleeve, and the lower end face of the rotating sleeve is fixed to the upper center position of the clamping block.

[0011] As a further improvement of the present invention, the control block is perpendicular to the screw, the screw is threadedly connected to the center of the positioning block through an external thread, the rotating sleeve has a restraining block and an internal thread groove to restrain the lower end of the screw without affecting the parallel displacement of the clamping block, and the clamping block is arc-shaped.

[0012] As a further improvement of the present invention, the clamping block is also provided with a rubber block, which is integrated with the lower layer of the block body. An edge block is connected to the upper end of the block body, and the edge block is integrated with the contact wall. A magnetic block is installed in the spacing between the edge blocks.

[0013] As a further improvement of the present invention, the rubber block is arc-shaped and integrated with the block body. The edge block of the block body is set in a vertical orientation and the rotating sleeve is vertically embedded through the contact wall. The magnetic block is embedded in the lower center of the rotating sleeve.

[0014] As a further improvement to the present invention, a detailed analysis of stage one: network discovery and topology construction is as follows: Link layer discovery: The controller sends LLDP (Link Layer Discovery Protocol) packets to all switches via a southbound interface (such as OpenFlow); Once a switch receives an LLDP packet, it will forward it out through all of its ports. When a switch receives an LLDP packet from a neighboring switch on a certain port, it will report this packet to the controller via a Packet-In message; Topology calculation: The controller analyzes the received LLDP packets and learns that "port 2 of switch S1 is connected to port 1 of switch S2"; The controller stores all such information in its network database and constructs a real-time network topology map; Host discovery: When a host sends an ARP request or a regular data packet for the first time, the switch reports it, and the controller learns that "host A's MAC / IP connection is on port 1 of S1".

[0015] As a further improvement of the present invention, the specific analysis of stage two: the initial processing of the data stream is as follows: Packet arrival failed to match flow table: The data packet arrives at port 1 of S1; S1 extracts key fields from data packets (source / destination MAC, IP, port, etc.) and searches them in its flow table; Since this is the first communication, the flow table is empty and there are no matching entries. Report to controller: According to the OpenFlow protocol, packets without a match are encapsulated in a Packet-In message and sent to the controller; The Packet-In message carries: packet header information (or the first N bytes) and ingress port (in_port=1); Controller decision: Analysis: The controller parses information such as the source address (A) and destination address (B) of the data packet from the Packet-In; Query: The controller queries its network information database: Host A is connected to S1:port1; Host B is connected to S2:port3; S1 is connected to port1 of S2 via port2; Path calculation: The controller runs a routing algorithm (such as the shortest path) to calculate the path from A to B, S1:port1 ->S1:port2 ->S2:port1 ->S2:port3; Generation strategy: The controller decides that "all flows from A to B should be forwarded from port 2 of S1; all flows from B to A should be forwarded from port 1 of S2". Flow table distribution: The controller sends specific flow table entries to the switches along the path via Flow-Mod messages; For S1: Match item = (source IP = A, destination IP = B), action = output from port 2; For S2: Match item = (source IP = A, destination IP = B), action = output from port 3; In addition, in order to establish bidirectional communication, the controller will usually also issue flow table entries for the reverse path (B->A). Process the first data packet: The controller instructs S1 to send the original data packet that was just reported out from port 2 via a Packet-Out message; When this data packet arrives at S2, S2's flow table may have already been updated (or it may have undergone a Packet-In process as well). Finally, the data packet is sent from port 3 of S2 and arrives at host B.

[0016] As a further improvement of the present invention, the specific analysis of stage three: high-speed forwarding of subsequent traffic is as follows: When the second and subsequent data packets sent by A to B arrive at S1, there are already matching entries in the flow table; S1 and S2 no longer report to the controller, but instead perform high-speed forwarding directly in the data plane based on the "action" of the flow table; This achieves an efficient "first-time control, subsequent forwarding" model.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention improves the data connection structure by using a locking ring to securely connect the flat plate to the surface of the insulating box. Then, the central restraint rod of the reinforcing frame is embedded in the center of the data cable spool connector for restraint. At the same time, the adjusting parts of the reinforcing frame are clamped symmetrically, so that the central restraint rod can achieve a secure connection of the data cable spool connector, preventing automatic detachment during long-term use. This facilitates disassembly and assembly, thereby improving the actual stability of the network controller.

[0018] 2. With the improvement of the adjusting component, the positioning block allows the control block to stably drive the screw to rotate. Then, the screw, in conjunction with the rotating sleeve, stably pushes the arc-shaped clamping block, forcing it to clamp the edge of the data cable spool connector inside the slot, achieving a firm connection. Furthermore, the screw's thread self-locking property improves the connection's firmness. Subsequently, the clamping block, in conjunction with the rubber block, protects the edge of the data connector, preventing friction damage after clamping. This further improves the overall performance of the components. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a network controller.

[0020] Figure 2 This is a three-dimensional structural diagram of an improved data connection structure.

[0021] Figure 3 This is a top-view structural diagram of an improved reinforced frame.

[0022] Figure 4 This is a cross-sectional structural diagram of an improved adjustment component.

[0023] Figure 5 This is a cross-sectional structural diagram of an improved clamping block.

[0024] Figure 6 This is a three-stage flowchart of a network control method.

[0025] Figure 7 This is a flowchart illustrating the initial data processing in the second stage of a network control method.

[0026] In the diagram: Locking block-1, Insulation box-2, Control body-3, Power supply slot-4, Data connection structure-5, Heat dissipation end-6; Locking ring-51, flat plate-52, reinforced frame-53, slot-54, center restraint rod-55; Welding layer-531, frame-532, adjusting component-533, power-conducting chip-534, clamping ring-535; Control block-5331, screw-5332, positioning block-5333, rotating sleeve-5334, clamping block-5335; Rubber block-3351, block-3352, edge block-3353, contact wall-3354, magnetic block-3355. Detailed Implementation

[0027] The present invention will be further described below with reference to the accompanying drawings: Example Figures 1 to 7 As shown: This invention provides a network controller and a network control method. Its structure includes: a locking block 1, an insulating box 2, a control body 3, a power supply slot 4, a data connection structure 5, and a heat dissipation end 6. The locking block 1 is fixed to the left and right sides of the insulating box 2, and the insulating box 2 covers the outside of the control body 3. The power supply slot 4 and the data connection structure 5 pass through the insulating box 2 and form an integral part with the control body 3. The heat dissipation end 6 is located at the upper end of the control body 3.

[0028] The data connection structure 5 is provided with a locking ring 51, which is located on the upper edge of the plate 52. The upper surface of the plate 52 is equipped with a reinforcing frame 53 and has a slot 54 and a central restraint rod 55 inside.

[0029] The control method of the network controller is as follows: Its setting: Objective: Host A (10.0.0.1) sends a data packet (e.g., a TCP SYN packet) to host B (10.0.0.2) for the first time. Network: A simple linear topology: Host A - Switch S1 - Switch S2 - Host B; Controller: A centralized SDN controller (such as Ryu or ONOS) that establishes OpenFlow connections with both S1 and S2; Initial state: All switches have empty flow tables; Its control and processing method will consist of the following stages: Phase 1: Network Discovery and Topology Building (Initialization, Continuous Operation); Phase Two: Initial processing of the data stream (the core of reaction control); Phase 3: High-speed forwarding of subsequent traffic.

[0030] The locking block 1 is symmetrically positioned at the edge of the insulating box 2, and the shape of the insulating box 2 matches that of the control body 3. The power supply slot 4 of the control body 3 is larger than the size of the data connection structure 5, and the heat dissipation end 6 is positioned parallel to the upper end of the control body 3.

[0031] The locking ring 51 is circular and has an internal threaded groove. The slot 54 of the reinforcing frame 53 is set in a vertical position. The central restraining rod 55 is a silicone cylinder.

[0032] The reinforcing frame 53 is provided with a welding layer 531, which is integrated with the lower edge of the frame 532. An adjustment component 533 is mounted on the edge surface of the frame 532. The adjustment component 533 passes through the edge of the frame 532 and communicates with the power-conducting chip 534 and clamping ring 535 provided inside.

[0033] The frame 532 is parallel to the plate 52 via a welding layer 531, the adjusting member 533 of the frame 532 is set in a symmetrical orientation, and the center position of the power-on chip 534 is determined by the clamping ring 535 to determine the position of the central restraint rod 55.

[0034] The adjusting component 533 is further provided with a control block 5331. The lower end of the control block 5331 is connected to a screw 5332. The lower end of the screw 5332 passes through the center of the positioning block 5333 and is connected to the inside of the rotating sleeve 5334. The lower end face of the rotating sleeve 5334 is fixed to the upper center position of the clamping block 5335.

[0035] The control block 5331 and the screw 5332 are perpendicular to each other. The screw 5332 is threadedly connected to the center of the positioning block 5333 through an external thread. The rotating sleeve 5334 has a restraining block and an internal thread groove to restrain the lower end of the screw 5332 without affecting the parallel displacement of the clamping block 5335. The clamping block 5335 is arc-shaped.

[0036] The clamping block 5335 is also provided with a rubber block 3351, which is integrated with the lower layer of the block body 3352. An edge block 3353 is connected to the upper end of the block body 3352. The edge block 3353 is integrated with the contact wall 3354. A magnetic block 3355 is installed in the spacing of the edge blocks 3353.

[0037] The rubber block 3351 is arc-shaped and integrated with the block 3352. The edge block 3353 of the block 3352 is set in a vertical orientation and the rotating sleeve 5334 is vertically embedded through the contact wall 3354. The magnetic block 3355 is embedded in the lower center of the rotating sleeve 5334.

[0038] Specifically, a detailed analysis of Phase One: Network Discovery and Topology Construction is as follows: Link layer discovery: The controller sends LLDP (Link Layer Discovery Protocol) packets to all switches via a southbound interface (such as OpenFlow); Once a switch receives an LLDP packet, it will forward it out through all of its ports. When a switch receives an LLDP packet from a neighboring switch on a certain port, it will report this packet to the controller via a Packet-In message; Topology calculation: The controller analyzes the received LLDP packets and learns that "port 2 of switch S1 is connected to port 1 of switch S2"; The controller stores all such information in its network database and constructs a real-time network topology map; Host discovery: When a host sends an ARP request or a regular data packet for the first time, the switch reports it, and the controller learns that "host A's MAC / IP connection is on port 1 of S1".

[0039] Specifically, a detailed analysis of Phase Two: the initial processing of the data stream is as follows: Packet arrival failed to match flow table: The data packet arrives at port 1 of S1; S1 extracts key fields from data packets (source / destination MAC, IP, port, etc.) and searches them in its flow table; Since this is the first communication, the flow table is empty and there are no matching entries. Report to controller: According to the OpenFlow protocol, packets without a match are encapsulated in a Packet-In message and sent to the controller; The Packet-In message carries: packet header information (or the first N bytes) and ingress port (in_port=1); Controller decision: Analysis: The controller parses information such as the source address (A) and destination address (B) of the data packet from the Packet-In; Query: The controller queries its network information database: Host A is connected to S1:port1; Host B is connected to S2:port3; S1 is connected to port1 of S2 via port2; Path calculation: The controller runs a routing algorithm (such as the shortest path) to calculate the path from A to B, S1:port1 ->S1:port2 ->S2:port1 ->S2:port3; Generation strategy: The controller decides that "all flows from A to B should be forwarded from port 2 of S1; all flows from B to A should be forwarded from port 1 of S2". Flow table distribution: The controller sends specific flow table entries to the switches along the path via Flow-Mod messages; For S1: Match item = (source IP = A, destination IP = B), action = output from port 2; For S2: Match item = (source IP = A, destination IP = B), action = output from port 3; In addition, in order to establish bidirectional communication, the controller will usually also issue flow table entries for the reverse path (B->A). Process the first data packet: The controller instructs S1 to send the original data packet that was just reported out from port 2 via a Packet-Out message; When this data packet arrives at S2, S2's flow table may have already been updated (or it may have undergone a Packet-In process as well). Finally, the data packet is sent from port 3 of S2 and arrives at host B.

[0040] Specifically, a detailed analysis of stage three: high-speed forwarding of subsequent traffic is as follows: When the second and subsequent data packets sent by A to B arrive at S1, there are already matching entries in the flow table; S1 and S2 no longer report to the controller, but instead perform high-speed forwarding directly in the data plane based on the "action" of the flow table; This achieves an efficient "first-time control, subsequent forwarding" model.

[0041] The specific functions and operation procedures of this embodiment are as follows: In this invention, A network controller and a network control method The control unit 3 of the network controller is first externally protected by the insulating box 2, then its position is determined by the locking block 1, and then it is connected to the main power supply through the power supply slot 4. Subsequently, it exchanges data through the data connection structure 5. During operation, its own temperature can be balanced through the heat dissipation end 6. Therefore, when the control unit 3 performs network control, it can: Phase 1: Network Discovery and Topology Building (Initialization, Continuous Operation); Phase Two: Initial processing of the data stream (the core of reaction control); Phase 3: High-speed forwarding of subsequent traffic; These three stages are controlled, and they will be used in a specific scenario (scenario setting: Objective: Host A (10.0.0.1) sends a data packet (e.g., a TCP SYN packet) to host B (10.0.0.2) for the first time. Network: A simple linear topology: Host A - Switch S1 - Switch S2 - Host B; Controller: A centralized SDN controller (such as Ryu or ONOS) that establishes OpenFlow connections with both S1 and S2; Initial state: Control operations are performed with all switches' flow tables empty. Then, a complete network control method is formed based on the detailed process of each of the following stages: Phase 1: Detailed Analysis of Network Discovery and Topology Construction Link layer discovery: The controller sends LLDP (Link Layer Discovery Protocol) packets to all switches via a southbound interface (such as OpenFlow); Once a switch receives an LLDP packet, it will forward it out through all of its ports. When a switch receives an LLDP packet from a neighboring switch on a certain port, it will report this packet to the controller via a Packet-In message; Topology calculation: The controller analyzes the received LLDP packets and learns that "port 2 of switch S1 is connected to port 1 of switch S2"; The controller stores all such information in its network database and constructs a real-time network topology map; Host discovery: When a host sends an ARP request or a regular data packet for the first time, the switch reports it, and the controller learns that "host A's MAC / IP connection is on port 1 of S1".

[0042] A detailed analysis of Phase Two: the initial processing of the data stream. Packet arrival failed to match flow table: The data packet arrives at port 1 of S1; S1 extracts key fields from data packets (source / destination MAC, IP, port, etc.) and searches them in its flow table; Since this is the first communication, the flow table is empty and there are no matching entries. Report to controller: According to the OpenFlow protocol, packets without a match are encapsulated in a Packet-In message and sent to the controller; The Packet-In message carries: packet header information (or the first N bytes) and ingress port (in_port=1); Controller decision: Analysis: The controller parses information such as the source address (A) and destination address (B) of the data packet from the Packet-In; Query: The controller queries its network information database: Host A is connected to S1:port1; Host B is connected to S2:port3; S1 is connected to port1 of S2 via port2; Path calculation: The controller runs a routing algorithm (such as the shortest path) to calculate the path from A to B, S1:port1 ->S1:port2 ->S2:port1 ->S2:port3; Generation strategy: The controller decides that "all flows from A to B should be forwarded from port 2 of S1; all flows from B to A should be forwarded from port 1 of S2". Flow table distribution: The controller sends specific flow table entries to the switches along the path via Flow-Mod messages; For S1: Match item = (source IP = A, destination IP = B), action = output from port 2; For S2: Match item = (source IP = A, destination IP = B), action = output from port 3; In addition, in order to establish bidirectional communication, the controller will usually also issue flow table entries for the reverse path (B->A). Process the first data packet: The controller instructs S1 to send the original data packet that was just reported out from port 2 via a Packet-Out message; When this data packet arrives at S2, S2's flow table may have already been updated (or it may have undergone a Packet-In process as well). Finally, the data packet is sent from port 3 of S2 and arrives at host B.

[0043] Phase Three: Detailed Analysis of High-Speed ​​Forwarding of Subsequent Traffic: When the second and subsequent data packets sent by A to B arrive at S1, there are already matching entries in the flow table; S1 and S2 no longer report to the controller, but instead perform high-speed forwarding directly in the data plane based on the "action" of the flow table; This achieves a highly efficient "first-time control, subsequent forwarding" model; This allows the control body 3 of the network controller to perform precise operations. The reinforcing frame 53 of the data connection structure 5 of the control body 3, combined with the central restraint rod 55, firmly secures the data cable spool connector, preventing automatic detachment of the data cable spool connector during long-term operation. The frame 532 of the reinforcing frame 53 is first fixed to the plate 52 via a welding layer 531. Then, the slot 54 inside the frame 532 allows the data cable spool connector to be embedded and connected to the power-conducting chip 534 for power supply. The silicone central restraint rod 55 of the clamping ring 535 is then inserted into the middle of the data connector, creating a central restraint effect. A secondary reinforcement is then achieved using the adjusting piece 533 on the edge of the frame 532, ensuring the connector's connection is secure and preventing detachment. The positioning block 5333 of the adjusting piece 533 is then fixed to the edge of the frame 532, allowing the control block 5331 to move the screw 5332 at the center. The screw 5332, through its threaded characteristics, rotates the sleeve 53. The rotating sleeve 5334 prevents the clamping block 5335 from rotating during rotation. This, combined with the internal thread of the positioning block 5333, allows the clamping block 5335 to be stably pushed and translated. The clamping block 5335, with its self-locking thread, securely clamps the edge of the data connector, preventing loosening and improving the stability of network power control. Finally, the clamping block 5335's body 3352 protects the edge of the data connector with rubber blocks 3351, preventing damage to the external coating during disassembly and assembly due to surface friction. The upper edge block 3353 of the body 3352 allows the rotating sleeve 5334 to be vertically inserted using the contact wall 3354, restraining its edge and preventing tilting. The magnetic block 3355 at the upper center of the body 3352 allows the lower end of the rotating sleeve 5334 to be embedded, creating a stable vertical constraint effect. This further ensures the vertical rotation of the screw 5332 and the accuracy of the translation of the clamping block 5335.

[0044] Any technical solution that achieves the above-mentioned technical effects by utilizing the technical solutions described in this invention, or by designing similar technical solutions by those skilled in the art under the inspiration of the technical solutions described in this invention, falls within the protection scope of this invention.

Claims

1. A network controller, the structure of which includes: The control body comprises a locking block (1), an insulating box (2), a control body (3), a power supply slot (4), a data connection structure (5), and a heat dissipation end (6). The locking block (1) is fixed to the left and right sides of the insulating box (2), and the insulating box (2) covers the outside of the control body (3). The power supply slot (4) and the data connection structure (5) penetrate the insulating box (2) and form an integral part with the control body (3). The heat dissipation end (6) is located at the upper end of the control body (3). The data connection structure (5) is provided with a locking ring (51), which is located on the upper edge of the plate (52). The upper surface of the plate (52) is equipped with a reinforcing frame (53) and has a slot (54) and a central restraint rod (55) inside. The control method of the network controller: Its setting: Objective: Host A (10.0.0.1) sends a data packet (e.g., a TCP SYN packet) to host B (10.0.0.2) for the first time. Network: A simple linear topology: Host A - Switch S1 - Switch S2 - Host B; Controller: A centralized SDN controller (such as Ryu or ONOS) that establishes OpenFlow connections with both S1 and S2; Initial state: All switches have empty flow tables; Its control and processing method will consist of the following stages: Phase 1: Network Discovery and Topology Building (Initialization, Continuous Operation); Phase Two: Initial processing of the data stream (the core of reaction control); Phase 3: High-speed forwarding of subsequent traffic.

2. A network controller according to claim 1, characterized in that: The locking block (1) is symmetrically positioned at the edge of the insulating box (2), and the shape of the insulating box (2) matches that of the control body (3). The power supply slot (4) of the control body (3) is larger than the size of the data connection structure (5), and the heat dissipation end (6) is set at the upper end of the control body (3) in a parallel orientation.

3. A network controller according to claim 1, characterized in that: The locking ring (51) is circular and has an internal threaded groove. The slot (54) of the reinforcing frame (53) is set in a vertical position. The central restraint rod (55) is a silicone cylinder.

4. A network controller according to claim 1, characterized in that: The reinforcing frame (53) is provided with a welding layer (531), which is integral with the lower edge of the frame (532). An adjustment component (533) is mounted on the edge surface of the frame (532). The adjustment component (533) passes through the edge of the frame (532) and communicates with the internally provided power chip (534) and clamping ring (535). The frame (532) is parallel to the plate (52) through the welding layer (531), the adjustment part (533) of the frame (532) is set in a symmetrical orientation, and the center position of the power-on chip (534) is determined by the clamp (535) to determine the position of the central restraint rod (55).

5. A network controller according to claim 4, characterized in that: The adjusting component (533) is also provided with a control block (5331). The lower end of the control block (5331) is connected to a screw (5332). The lower end of the screw (5332) passes through the center of the positioning block (5333) and is connected to the inside of the rotating sleeve (5334). The lower end face of the rotating sleeve (5334) is fixed to the upper center of the clamping block (5335). The control block (5331) is perpendicular to the screw (5332). The screw (5332) is threadedly connected to the center of the positioning block (5333) through an external thread. The rotating sleeve (5334) has a restraining block and an internal thread groove to restrain the lower end of the screw (5332) without affecting the parallel displacement of the clamping block (5335). The clamping block (5335) is arc-shaped.

6. A network controller according to claim 5, characterized in that: The clamping block (5335) is also provided with a rubber block (3351), which is integrated with the lower layer of the block body (3352). An edge block (3353) is connected to the upper end of the block body (3352). The edge block (3353) is integrated with the contact wall (3354). A magnetic block (3355) is installed in the spacing between the edge blocks (3353). The rubber block (3351) is arc-shaped and integrated with the block (3352). The edge block (3353) of the block (3352) is set in a vertical orientation and the rotating sleeve (5334) is vertically embedded through the contact wall (3354). The magnetic block (3355) is embedded in the lower center of the rotating sleeve (5334).

7. The network control method according to claim 1, characterized in that: A detailed analysis of Phase One: Network Discovery and Topology Construction: Link layer discovery: The controller sends LLDP (Link Layer Discovery Protocol) packets to all switches via a southbound interface (such as OpenFlow); Once a switch receives an LLDP packet, it will forward it out through all of its ports. When a switch receives an LLDP packet from a neighboring switch on a certain port, it will report this packet to the controller via a Packet-In message; Topology calculation: The controller analyzes the received LLDP packets and learns that "port 2 of switch S1 is connected to port 1 of switch S2". The controller stores all such information in its network database and constructs a real-time network topology map; Host discovery: When a host sends an ARP request or a regular data packet for the first time, the switch reports it, and the controller learns that "host A's MAC / IP connection is on port 1 of S1".

8. The network control method according to claim 1, characterized in that: A detailed analysis of Phase Two: the initial processing of the data stream. Packet arrival failed to match flow table: The data packet arrives at port 1 of S1; S1 extracts key fields from data packets (source / destination MAC, IP, port, etc.) and searches them in its flow table; Since this is the first communication, the flow table is empty and there are no matching entries. Report to controller: According to the OpenFlow protocol, packets without a match are encapsulated in a Packet-In message and sent to the controller; The Packet-In message carries: packet header information (or the first N bytes) and ingress port (in_port=1); Controller decision: Analysis: The controller parses information such as the source address (A) and destination address (B) of the data packet from the Packet-In; Query: The controller queries its network information database: Host A is connected to S1:port1; Host B is connected to S2:port3; S1 is connected to port1 of S2 via port2; Path calculation: The controller runs a routing algorithm (such as the shortest path) to calculate the path from A to B, S1:port1 -> S1:port2 -> S2:port1 -> S2:port3; Generation strategy: The controller decides that "all flows from A to B should be forwarded from port 2 of S1; all flows from B to A should be forwarded from port 1 of S2". Flow table distribution: The controller sends specific flow table entries to the switches along the path via Flow-Mod messages; For S1: Match item = (source IP = A, destination IP = B), action = output from port 2; For S2: Match item = (source IP = A, destination IP = B), action = output from port 3; In addition, in order to establish bidirectional communication, the controller will usually also issue flow table entries for the reverse path (B->A). Process the first data packet: The controller instructs S1 to send the original data packet that was just reported out from port 2 via a Packet-Out message; When this data packet arrives at S2, S2's flow table may have already been updated (or it may have undergone a Packet-In process as well). Finally, the data packet is sent from port 3 of S2 and arrives at host B.

9. A network control method according to claim 1, characterized in that: Phase Three: Detailed Analysis of High-Speed ​​Forwarding of Subsequent Traffic: When the second and subsequent data packets sent by A to B arrive at S1, there are already matching entries in the flow table; S1 and S2 no longer report to the controller, but instead perform high-speed forwarding directly in the data plane based on the "action" of the flow table; This achieves the efficient model of "initial control, subsequent forwarding".

Images