Network switch testing device and method for testing network switch
The proposed test apparatus and method for network switches address inefficiencies and costs in current testing methods by using internal loopback and flooding within a virtual LAN to efficiently and accurately test high-speed network switches, reducing test time and cost while enhancing accuracy and range.
Patent Information
- Application Number
- JP2024060134
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-11-07
- Filing Date
- 2024-04-03
- Publication Date
- 2025-05-19
- Estimated Expiration
- 2044-04-03
AI Technical Summary
Current network switch testing methods are inefficient and costly, particularly when testing high-speed network switches, as they often require expensive external test equipment and cannot accurately measure network speed with small packets.
A test apparatus and method for network switches that utilize a central processing unit and multiple ports to generate packets and set acceleration and forwarding ports within the same virtual LAN, allowing for internal loopback and flooding to reach a predetermined line rate, thereby testing network switches efficiently and accurately.
This solution significantly reduces test time and cost while improving test accuracy and range, enabling the simultaneous testing of multiple network switches and meeting industry requirements for high-speed network switch testing.
Smart Images

Figure 2025077952000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a test apparatus and a test method, and particularly to a test apparatus and a test method for a network switch.
Background Art
[0002] In the manufacturing process of a network switch, it is necessary to test the packet transfer function. As the requirement for the transfer speed of network data increases significantly, the test of the network switch also needs to support high-speed data transfer speeds. Currently, the industry includes loopback test, saturation test, line rate test, line rate unicast test, snake test, and external test.
[0003] In the loopback test, the central processing unit (CPU) of the network switch under test performs the transmission and reception of a specific packet format through its own transfer port, and the CPU checks the packet content. The loopback test examines the packet transfer path while also checking whether the direct memory access (DMA) transfer from the media access control (MAC) to the CPU is normal. However, its drawback is that since the CPU generates and checks the packet content, the transfer speed of the packet does not improve.
[0004] The saturation test arranges all ports of the network switch under test in the same virtual local area network (VLAN), and the network switch under test receives, transmits, and forwards packets of different sizes through broadcasts. This reception, transmission, and forwarding are carried out until the network switch under test approaches the full load state of the line. Then, it starts to inspect the packets, waits for the scheduled inspection time, and checks whether incorrect packets are received during the scheduled inspection period. The saturation test has a high possibility of detecting problems that may occur during high-speed transfer. However, since the line is filled with packets of different sizes, it is impossible to accurately calculate the network speed.
[0005] The line rate test sets two transfer ports of the network switch under test in the same VLAN, and through the VLAN mechanism, it sends, receives, and forwards fixed-length packets through broadcasts within the network switch under test until the network switch under test reaches full load. Then, it calculates the overall transfer speed using the number and fixed length of the sent and received packets. However, since the memory management unit (MMU) inside the new system on a chip (SoC) has usage rate limitations for unknown unicast, multicast, and broadcast in the distribution of limited resources in packet transfer, when the transferred packets are below a specific length, a phenomenon occurs where the network transfer speed does not reach the standard. Therefore, the line rate test cannot perform a network speed test using small packets.
[0006] The Line rate unicast test configures two transfer ports of the network switch under test to belong to two VLANs and their respective Port-Based VLANs. Through external connection lines, changes to the L2 table, and the VLAN mechanism, fixed-length unicast packages are transmitted, received, and transferred within the network switch under test until the packages reach full load within the network switch under test. Then, the overall transfer speed is calculated using the number and fixed length of the transmitted and received packages. The Line rate unicast test can solve the problem that the network transfer speed standard is not reached when the line rate test transfers multicast packages below a specific length, but it does not meet the requirements for using external network performance test equipment in the industry.
[0007] The Snake test connects all transfer ports of the network switch under test in series so that packages are transferred from the first transfer port to the last transfer port. Since the Snake test also requires packages to be generated and inspected by the CPU of the network switch under test, due to CPU limitations, it is not possible to generate a sufficient number of packages to reach the maximum speed of the line.
[0008] An external test performs packet transmission and reception through an external network performance test device under the same settings as a snake test, and calculates the overall transfer speed. The external test can reach the maximum speed of the line and verify the functions of VLANs. However, network performance test devices available on the market (such as Spirent Test Center, IXIA, etc.) are very expensive and costly to maintain and repair later, which does not meet the production cost-effectiveness. Furthermore, commercially available network performance test devices have a small number of transfer ports and cannot test multiple network switches to be tested simultaneously, resulting in a longer test time and increased costs.
Summary of the Invention
Problems to be Solved by the Invention
[0009] Therefore, the main object of the present invention is to develop a method that improves the accuracy of the network switch to be tested, reduces the test time and cost of high-speed network switches, and meets the industry's requirements for using external network performance test devices. The present invention has been made in view of the above, and its object is to provide a test device and a test method for network switches.
Means for Solving the Problems
[0010] To achieve the above object, the present invention provides a test apparatus for a network switch, comprising a central processing unit used to generate at least one packet, a first acceleration port that clears packet statistics, stops media access control address learning, and starts internal loopback, a second acceleration port that clears packet statistics, stops media access control address learning, and starts internal loopback, and at least one forwarding port that clears packet statistics and a forwarding list, stops media access control address learning, and is connected to at least one test target forwarding port of at least one test target network switch. The first acceleration port, the second acceleration port, and the forwarding port are set in the same virtual LAN. The central processing unit transmits a packet to the first acceleration port. The first acceleration port and the second acceleration port transmit the packet to the forwarding port through internal loopback and the flooding mechanism of the virtual LAN to reach a predetermined line rate. The forwarding port transmits the packet that has reached the predetermined line rate to the test target forwarding port of the test target network switch. At the end of the test time, the central processing unit stops packet transmission from the forwarding port.
[0011] In one embodiment, the test apparatus is a network switch inspected to reach the above predetermined line rate.
[0012] In one embodiment, the packet can specify a source address, a destination address, an Ethernet type, packet data, a packet size, and a tagged or untagged packet.
[0013] In one embodiment, the number of the transfer ports is plural, and the number of the network switches to be tested is also plural. Different ones of the transfer ports can transfer the packets to the test target transfer ports of different ones of the network switches to be tested simultaneously at the same or different predetermined line rates.
[0014] To achieve the above object, the present invention provides a network switch testing method. The network switch testing method includes providing a test device and at least one network switch to be tested, where the test device includes a central processing unit and a plurality of ports, and the network switch to be tested includes at least one transfer port to be tested; the central processing unit deletes the virtual LANs of all ports of the test device and sets a first acceleration port, a second acceleration port, and at least one transfer port among the plurality of ports; the central processing unit clears the packet statistics of the first acceleration port, the second acceleration port, and the transfer port, and clears the forwarding list of the transfer port; the central processing unit stops the media access control address learning (MAC Address Learning) of the first acceleration port, the second acceleration port, and the transfer port; the central processing unit starts the internal loopback of the first acceleration port and the second acceleration port; the central processing unit sets the first acceleration port, the second acceleration port, and the transfer port to the same virtual LAN; the central processing unit transmits at least one packet to the first acceleration port; the first acceleration port and the second acceleration port cause the packet to reach a predetermined line rate by means of internal loopback and the flooding mechanism of the virtual LAN and transmit it to the transfer port; the transfer port transmits the packet that has reached the predetermined line rate to the transfer port to be tested of the network switch to be tested, and the transfer port or ports receive the packet returned from the transfer port to be tested; when the test time has elapsed, the central processing unit stops the packet transmission of the transfer port; and verifying the packet transmission quality of the network switch to be tested based on the number of packets transmitted and received by the test device.
[0015] In one embodiment, the packet can specify a source address, a destination address, an Ethernet type, packet data, a packet size, and a tagged or untagged packet.
[0016] In one embodiment, the number of the transfer ports is plural, and the number of the test target network switches is plural. Different said transfer ports may be set to transfer the packet to the test target transfer ports of different said test target network switches simultaneously at the same or different said predetermined line rates.
[0017] In one embodiment, before the central processing unit transmits the packet to the first acceleration port, a timer is used to set the test time.
[0018] In one embodiment, the central processing unit stops the transfer port from transmitting the packet by clearing the transfer list of the first acceleration port.
[0019] In the network switch test apparatus and test method of the present invention, the first acceleration port, the second acceleration port, and the transfer port of the test apparatus are set as traffic generators through software. The user can adjust the packet according to actual needs. For example, by specifying different packet data, packet sizes, tagged or untagged packets, and further sending packets at different predetermined line rates from different transfer ports, a plurality of test target network switches can be tested simultaneously. This meets the inspection requirements of the industry, improves the test range and accuracy, greatly reduces the test time and cost, and as a result, enhances the production capacity of the network switch.
Brief Description of the Drawings
[0020]
Figure 1
Figure 2A
Figure 2B
Figure 3
Embodiments for Carrying Out the Invention
[0021] Hereinafter, in order for those skilled in the art to be able to implement the present invention after reading this specification, embodiments of the present invention will be described in more detail with reference to the drawings and element symbols.
[0022] FIG. 1 is a block diagram of a test apparatus for a network switch of the present invention. As shown in FIG. 1, the test apparatus 1 for a network switch includes a central processing unit 11, a first acceleration port 12, a second acceleration port 13, and a plurality of transfer ports 14. The central processing unit 11, the first acceleration port 12, the second acceleration port 13, and the plurality of transfer ports 14 are connected by circuits within the apparatus, and the plurality of transfer ports 14 can be connected to a plurality of test target transfer ports (not shown) of at least one test target network switch through an external circuit.
[0023] Specifically, the network switch to be tested is, for example, an Ethernet switch, and the test apparatus 1 for the network switch can select and use an Ethernet switch that conforms to a predetermined line rate (for example, 100 Gb / s, 200 Gb / s, 400 Gb / s, 800 Gb / s or higher) and specifications. The test apparatus 1 for the network switch includes a central processing unit 11 and a plurality of ports 10, and stores the software of the Ethernet Traffic Generator (ETG) of the present invention. Before testing the network switch, the central processing unit 11 executes the following steps. Delete the virtual LANs of all ports 10, set a first acceleration port 12 and a second acceleration port 13 among the plurality of ports 10, and set a part of the other ports 10 as transfer ports 14. Set the first acceleration port 12, the second acceleration port 13, and the transfer port 14 in the same virtual LAN V1. Clear the packet statistics of the first acceleration port 12, stop the media access control address learning of the first acceleration port 12, and start the internal loopback of the first acceleration port 12. Clear the packet statistics of the second acceleration port 13, stop the media access control address learning of the second acceleration port 13, and start the internal loopback of the second acceleration port 13. Clear the packet statistics and transfer list of the transfer port 14, and stop the media access control address learning of the transfer port 14. By the ETG software, the central processing unit can set the first acceleration port 12, the second acceleration port 13, and the transfer port 14 as traffic generators.
[0024] When performing a test network switch, the central processing unit 11 generates at least one packet and transmits it to the first acceleration port 12. The first acceleration port 12 and the second acceleration port 13 transmit the packet to the transfer port 14 after reaching a predetermined line rate through the internal loopback and the flooding mechanism of the virtual LAN V1. The predetermined line rate is determined according to the line full speed of the test target network switch. The transfer port 14 transmits the packet that has reached the predetermined line rate to the test target transfer port of the test target network switch and receives the packet returned from the test target transfer port. The central processing unit 11 stops the packet transmission of the transfer port 14 when the test time ends. The packet transmission quality of the test target network switch can be verified by the number of packets transmitted and received by the test device 1. This includes whether the number of packets is correct, whether there are no incorrect packets, and whether the packet size is as expected.
[0025] Figure 2A is a schematic diagram of a bi-directional packet transmission test of a test device for a network switch according to an embodiment of the present invention. As shown in Figure 2A, the test device 2 for a network switch includes a central processing unit (not shown), a first acceleration port 22, a second acceleration port 23, and a plurality of transfer ports 24. The central processing unit, the first acceleration port 22, the second acceleration port 23, and the plurality of transfer ports 24 are connected by circuits within the device. Six of the plurality of transfer ports 24 are respectively connected to two test target transfer ports 251 of the test target network switch 25, two test target transfer ports 261 of the test target network switch 26, and two test target transfer ports 271 of the test target network switch 27 through the external circuits 201, 202, 203. The setting methods of the first acceleration port 22, the second acceleration port 23, and the transfer port 24 are as follows.
[0026] The test device 2 of the network switch includes a central processing unit and a plurality of ports, and stores the software of the Ethernet traffic generator (ETG) of the present invention. Before testing the test target network switches 25, 26, and 27, the central processing unit executes the following steps based on the ETG software. Delete the virtual LAN from all ports of the test device 2 in the network switch. Set the first acceleration port 22 and the second acceleration port 23 among the plurality of ports, and set all or part of the remaining ports as transfer ports 24. Set the six transfer ports 24 connected to the first acceleration port 22, the second acceleration port 23, and the test target transfer ports 251, 261, 271 to the same virtual LAN V2. Clear the packet statistics of the six transfer ports 24 connected to the first acceleration port 22, the second acceleration port 23, and the test target transfer ports 251, 261, 271, stop the media access control address learning of the six transfer ports 24 connected to the first acceleration port 22, the second acceleration port 23, and the test target transfer ports 251, 261, 271, start the internal loopback of the first acceleration port 22 and the second acceleration port 23, and clear the transfer list of the six transfer ports 24 connected to the test target transfer ports 251, 261, 271.
[0027] In this embodiment, the user can generate the same or different packets through the central processing unit. Also, a predetermined line rate of each transfer port 24 can be set through the central processing unit, and the three test target network switches 25, 26, 27 can be tested bidirectionally at the same time. The packets can be adjusted according to actual requirements. For example, the source address, destination address, Ethernet type, packet data, packet size, tagged or untagged packets, etc. can be specified. The predetermined line rate for transferring packets at each transfer port 24 may be the same or different.
[0028] When testing the test target network switches 25, 26, and 27, the central processing unit generates two different types of packets with multiple quantities (e.g., one digit, two digits, three digits), and sends these two types of packets to the first acceleration port 22. The two types of packets are looped back within the first acceleration port 22. Since the first acceleration port 22 and the second acceleration port 23 are set to the same virtual LAN V2, after the two types of packets are looped back within the first acceleration port 22, they are flooded to the second acceleration port 23, and after being looped back within the second acceleration port 23, they are flooded to the first acceleration port 22. The two types of packets circulate between the first acceleration port 22 and the second acceleration port 23 for a certain period of time and then reach a predetermined line rate. The predetermined line rate is set based on the line full speed of each test target network switch 25, 26, 27. For example, by setting the transfer port 24 connected to the test target network switches 25, 26 to 400G and the transfer port 24 connected to the test target network switch 27 to 100G, it is possible to verify the test target network switches 25, 26 with a line speed of 400G and the test target network switch 27 with a line speed of 100G simultaneously. Since the six transfer ports 24 connected to the test target transfer ports 251, 261, 271 are also set to the virtual LAN V2, the two types of packets in the first acceleration port 22 and the second acceleration port 23 are flooded to the six transfer ports 24 connected to the test target transfer ports 251, 261, 271, whereby the six transfer ports 24 transfer the two types of packets that have reached the predetermined line rate to the test target transfer ports 251, 261, 271 respectively. After receiving the packets, the test target network switches 25, 26, 27 return the packets from the test target transfer ports 251, 261, 271 to the transfer port 24 respectively.
[0029] Figure 2B is a schematic diagram of a unidirectional packet transmission test of a network switch test apparatus according to an embodiment of the present invention. As shown in Figure 2B, the network switch test apparatus 2 includes a central processing unit (not shown), a first acceleration port 22, a second acceleration port 23, a transfer port 24, and a port 20. The central processing unit, the first acceleration port 22, the second acceleration port 23, the transfer port 24, and the port 20 are connected by circuits within the apparatus. The transfer port 24 and the port 20 are connected to two test target transfer ports 251 of a test target network switch 25, two test target transfer ports 261 of a test target network switch 26, and two test target transfer ports 271 of a test target network switch 27 through external circuits 201, 202, 203, respectively. The setting methods for the first acceleration port 22, the second acceleration port 23, and the transfer port 24 are as follows.
[0030] The network switch test apparatus 2 includes a central processing unit and a plurality of ports, and stores the software of the Ethernet traffic generator (ETG) of the present invention. Before testing the test target network switches 25, 26, 27, the central processing unit executes the following steps. Remove the virtual LAN from all ports in the network switch test apparatus 2. Set the first acceleration port 22 and the second acceleration port 23 among the plurality of ports, and set a part of the remaining ports as the transfer port 24. Set the first acceleration port 22, the second acceleration port 23, and the transfer port 24 to the same virtual LAN V2. The unconfigured port 20 is not included in the virtual LAN V2, and packet transfer to other ports is not performed. Clear the packet statistics of the first acceleration port 22, the second acceleration port 23, and the transfer port 24, stop the media access control address learning of the first acceleration port 22, the second acceleration port 23, and the transfer port 24, start the internal loopback of the first acceleration port 22 and the second acceleration port 23, and clear the transfer list of the transfer port 24.
[0031] When testing the test target network switches 25, 26, and 27, the central processing unit generates packets in multiple quantities (e.g., one digit, two digits, three digits) and sends these packets to the first acceleration port 22. The packets are looped back within the first acceleration port 22. Since the first acceleration port 22 and the second acceleration port 23 are set to the same virtual LAN V2, after being looped back within the first acceleration port 22, they are flooded to the second acceleration port 23, and after being looped back within the second acceleration port 23, they are flooded to the first acceleration port 22. The packets circulate between the first acceleration port 22 and the second acceleration port 23 for a certain period of time and then reach a predetermined line rate. The predetermined line rate is set based on the line full speed of each test target network switch 25, 26, 27. For example, by setting the transfer port 24 connected to the test target network switches 25, 26 to 400G and the transfer port 24 connected to the test target network switch 27 to 100G, it is possible to simultaneously verify the test target network switches 25, 26 with a line speed of 400G and the test target network switch 27 with a line speed of 100G. Since the transfer ports 24 connected to the test target transfer ports 251, 261, 271 are also set to the virtual LAN V2, the packets in the first acceleration port 22 and the second acceleration port 23 are flooded to the transfer ports 24 connected to the test target transfer ports 251, 261, 271, whereby the transfer port 24 transfers the packets that have reached the predetermined line rate to the test target transfer ports 251, 261, 271 respectively. After receiving the packets, the three test target network switches 25, 26, 27 return the packets from the respective other test target transfer ports 251, 261, 271 to the port 20.
[0032] Regardless of whether it is a two-way or one-way packet transmission test, the user can either preset the test time or determine it in real time. For example, when transmitting packets for 5 minutes, the user can preset the test time to 5 minutes or stop packet transmission through a command when 5 minutes have elapsed. When the test time elapses, the central processing unit clears the transfer list of the first acceleration port 22 and stops packet flooding. As a result, the transfer port 24 stops packet transmission. After a certain period of time (e.g., 0.2 seconds), the transfer list of the first acceleration port 22 is restored and it waits for the next test. The user connects to the central processing unit of the test device 2 using an external computer device and can inspect the line speed and quality (e.g., packet loss rate, error rate, etc.) of the test target network switches 25, 26, 27 based on the number of two packets transmitted and received by the transfer port 24 and the test time.
[0033] The test device for the network switch of the present invention can transmit specific packets at a predetermined line rate, and the user can adjust the packets according to actual requirements. For example, by specifying different packet data, packet sizes, tagged or untagged packets, it is possible to improve the scope and accuracy of the test. Current network switches have up to 32 ports or 48 ports. By using software settings to use a network switch that conforms to the product specifications as a test device, multiple test target network switches can be tested simultaneously (up to (number of ports N - 2) / 2 units at most). This can significantly reduce the test time and cost and improve the production capacity of the network switch.
[0034] Figure 3 is a flowchart of the network switch testing method of the present invention. As shown in Figure 3, the network switch testing method of the present invention includes the following steps: Step S30: Provide a test device and at least one network switch to be tested. Here, the test device includes a central processing unit and a plurality of ports, and the network switch to be tested includes a plurality of transfer ports to be tested. Step S31: The central processing unit removes the virtual LANs of all ports and sets a first acceleration port, a second acceleration port, and at least one transfer port among the plurality of ports. Step S32: The central processing unit clears the packet statistics of the first acceleration port, the second acceleration port, and the transfer port, and clears the transfer list of the transfer port. Step S33: The central processing unit stops the media access control address learning of the first acceleration port, the second acceleration port, and the transfer port. Step S34: The central processing unit starts the internal loopback of the first acceleration port and the second acceleration port. Step S35: The central processing unit sets the first acceleration port, the second acceleration port, and the transfer port to the same virtual LAN. Step S36: The central processing unit sends a packet to the first acceleration port. Step S37: The first acceleration port and the second acceleration port cause the packet to reach a predetermined line rate through the internal loopback and the flooding mechanism of the virtual LAN and send it to the transfer port. Step S38: The transfer port sends the packet that has reached the predetermined line rate to the transfer port to be tested of the network switch to be tested, and receives the packet returned from the transfer port or the port from the transfer port to be tested. Step S381: Check whether the test time has elapsed. If not, return to Step S38. If it has elapsed, proceed to Step S39. Step S39: The central processing unit stops the packet transmission of the transfer port when the test time has elapsed, and the test device inspects the packet transmission quality of the network switch to be tested based on the number of packets transmitted and received.
[0035] Specifically, the network switch to be tested is, for example, an Ethernet switch, and the test apparatus can selectively use an Ethernet switch that meets a predetermined line rate (for example, 100 Gb / s, 200 Gb / s, 400 Gb / s, 800 Gb / s or higher) and specifications. The packets can be single or different and can be adjusted according to actual requirements. For example, the source address, destination address, Ethernet type, packet data, packet size, tagged or untagged packets can be specified. The predetermined line rate is set according to the actual requirements of each network switch to be tested, and packets with different predetermined line rates can be sent from different transfer ports to test multiple network switches to be tested simultaneously.
[0036] It should be noted that there is no specific order for steps S32 to S35. Before performing step S36, the user can preset the test time using the timer of the central processing unit. By providing the preset test time, the user can determine the start and stop times and duration of the test, and improve the flexibility of the test. In step S39, the central processing unit can stop packet transmission of the transfer port by clearing the transfer list of the first acceleration port. After a certain time (for example, 0.2 seconds) has elapsed after the completion of step S39, the central processing unit can restore the transfer list of the first acceleration port and wait for the next test.
[0037] Generally, in the network switch test apparatus and test method of the present invention, software is used to set the first acceleration port, the second acceleration port, and the transfer port of the test apparatus as traffic generators. The user can adjust the packets according to actual requirements. For example, different packet data, packet sizes, and tagged / untagged packets can be specified. Also, packets with different predetermined line rates can be sent from different transfer ports, and multiple network switches to be tested can be tested simultaneously. Thereby, not only can the test requirements of the industry be met, but also the test range and accuracy can be improved, the test time and cost can be significantly reduced, and the production capacity of the network switch can be improved.
[0038] The above-described embodiments are exemplary explanations showing the principles and effects of the present invention, and do not limit the present invention. Those skilled in the art can make modifications, combinations, and changes to the above embodiments within the scope not contrary to the spirit and scope of the present invention. Therefore, all equivalent modifications, combinations, or changes made by those with expertise in this technical field without departing from the spirit and technical principles disclosed by the present invention should also be included in the scope of the claims of the present invention.
Explanation of Signs
[0039] 1,2 Test apparatus for network switch 10,20 Ports 11 Central processing unit 12,22 First acceleration port 13,23 Second acceleration port 14,24 Transfer port 25,26,27 Network switches to be tested 201,202,203 External circuit 251,261,271 Transfer ports to be tested V1,V2 Virtual LAN S30,S31,S32,S33,S34,S35,S36,S37,S38,S381,S39 Steps
Claims
1. a central processing unit adapted to generate at least one packet; a first acceleration port that clears packet statistics, stops media access control address learning, and starts an internal loopback; a second acceleration port that clears packet statistics, stops media access control address learning, and starts an internal loopback; at least one forwarding port that clears packet statistics and forwarding lists, stops media access control address learning, and connects to at least one forwarding port under test of at least one network switch under test; Equipped with a first acceleration port, a second acceleration port, and a forwarding port, which are set to a same virtual LAN; a central processing unit transmits the packet to the first acceleration port; the first acceleration port and the second acceleration port cause the packet to reach a predetermined line rate through an internal loopback and a flooding mechanism of the virtual LAN and transmit the packet to the forwarding port; the forwarding port transmits the packet, which has reached the predetermined line rate, to the test target forwarding port of the test target network switch; and the central processing unit stops transmitting the packet by the forwarding port when a test time has elapsed.
2. 2. The network switch test apparatus according to claim 1, wherein the network switch is adapted to meet the predetermined line rate.
3. 2. The network switch test apparatus of claim 1, wherein the packets can specify a source address, a destination address, an Ethernet type, packet data, a packet size, and whether the packet is tagged or untagged.
4. 2. The network switch testing device of claim 1, wherein the number of forwarding ports is multiple, the number of network switches to be tested is also multiple, different forwarding ports are set to the same or different predetermined line rates, and the packets are simultaneously transmitted to the test target forwarding ports of the different network switches to be tested.
5. 2. The network switch testing apparatus of claim 1, wherein the central processing unit stops the packet transmission of the first acceleration ports by clearing the forwarding list of those forwarding ports.
6. A method for testing network switches. providing a test device including a central processing unit and a plurality of ports, and at least one network switch under test including at least one forwarding port under test; the central processing unit removes a virtual LAN for the plurality of ports of the test device and configures a first acceleration port, a second acceleration port, and at least one forwarding port among the plurality of ports; the central processing unit clearing packet statistics of the first acceleration port, the second acceleration port and the forwarding port, and clearing a forwarding list of the forwarding port; the central processing unit stopping media access control address learning of the first acceleration port, the second acceleration port and the forwarding port; the central processing unit initiating an internal loopback of the first acceleration port and the second acceleration port; the central processing unit sets the first acceleration port, the second acceleration port and the forwarding port to the same virtual LAN; the central processing unit transmitting at least one packet to the first acceleration port; the first acceleration port and the second acceleration port accelerate the packets to a predetermined line rate through an internal loopback and a flooding mechanism of the virtual LAN, and transmit the packets to the forwarding port; the forwarding port transmits the packets at the constant line rate to the forwarding port under test of the network switch under test, and the forwarding port or ports receive the packets returned from the forwarding port under test; the central processing unit stopping transmission of the packet by the forwarding port when a test time has elapsed; and checking a quality of the network switch under test transmitting the packets based on the number of the packets transmitted and received by the test equipment.
7. 7. The method of claim 6, wherein the packet can specify a source address, a destination address, an Ethernet type, packet data, a packet size, and whether the packet is tagged or untagged.
8. 7. The method for testing a network switch according to claim 6, wherein the number of the forwarding ports is multiple, the number of the test target network switches is also multiple, the different forwarding ports are set to the same or different predetermined line rates, and the packets are simultaneously sent to the test target forwarding ports of the different test target network switches.
9. 7. The method of claim 6, further comprising: presetting the test time using a timer before the central processing unit transmits the packet to the first acceleration port.
10. 7. The method of claim 6, wherein the central processing unit stops the packet transmission of the first acceleration ports by clearing a forwarding list of those forwarding ports.
Citation Information
Patent Citations
Testing method of network transmission rate
CN105897494A
Communication apparatus
JP2006245849A
Transmission apparatus
JP2009225404A
Ethernet port inspection system, ethernet port inspection method, and communication device inspection program
JP2010118823A
Relay device, network system, and packet forwarding method
JP6171704B2