A time-sensitive network tester port bandwidth expansion test device and method

Abstract

The application provides a time-sensitive network tester port bandwidth expansion testing device and method, which can connect a low-speed tester and a bandwidth expansion adapter under the premise that the low-speed tester is not modified, and then connect a high-speed device under test for testing. The adapter can ensure that the time delay generated in the rate conversion process is fixed, or the bandwidth of the received traffic is recovered after the bandwidth expansion of the transmitted traffic, so as to meet the testing requirements of the high-speed device under test. According to whether bandwidth expansion is needed, whether time delay needs to be ensured and whether traffic scheduling is needed, all network testing scenes are divided into three categories, the adapter adopts different traffic processing strategies in different testing scenes, ensures that the key traffic characteristics can still be exhibited after the traffic passes through the rate conversion, so that the testing results can be analyzed directly through the statistical parameters of the low-speed tester, additional statistical information does not need to be provided by the adapter, and the system complexity and use difficulty of the adapter are simplified.

Description

Technical Field

[0001] The present invention belongs to the technical field of communications, and in particular relates to a device and method for testing port bandwidth expansion of a time-sensitive network tester. Background Art

[0002] Traditional Layer 2 and 3 network testers typically test the forwarding and scheduling performance of the device under test (DUT) by sending high-bandwidth traffic. This means that network testers can typically only test DUTs with port speeds equal to or lower than the tester's. Furthermore, due to the greater system complexity and design challenges of high-speed network testers, network tester manufacturers tend to develop lower-speed devices faster and bring them to market sooner when designing test solutions and products for new network protocols. Once these high-speed testers are released, their prices are typically significantly higher than those of lower-speed devices, significantly increasing the development time and cost of network equipment.

[0003] For TSN (Time Sensitive Network), the testing needs for high-speed TSN devices are equally urgent. For the testing of some network protocols, a low-speed tester can meet the testing needs by simply connecting to the device under test through a simple port rate conversion. However, in the time-sensitive network protocol suite, the testing of some protocols is very sensitive to the forwarding delay of the test frame. For example, in the test of the time gating mechanism defined in the IEEE802.1Qbv protocol, the gate opening time of the output port of the device under test needs to be determined by the receiving time. If the additional delay of the rate converter is large or cannot be determined, the test cannot proceed normally; some protocols need to send large bandwidth traffic to the device under test to test the bandwidth reservation function of the device under test. Therefore, simple port rate conversion cannot meet the testing needs of time-sensitive networks. For this reason, it is urgent to research and implement a new switching device that can assist low-speed network testers in testing high-speed time-sensitive network devices to solve the high-speed testing problems that are becoming increasingly prominent with the development and maturity of time-sensitive network technology. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a device and method for testing the port bandwidth expansion of a time-sensitive network tester. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0005] The present invention provides a time-sensitive network tester port bandwidth expansion test device comprising: a converter with a high-speed port and a low-speed port; the converter is connected to a low-speed tester via the low-speed port and is connected to a high-speed device under test via the high-speed port;

[0006] The tester sends test frames to the adapter via a low-speed port. The adapter, based on its configuration, performs bandwidth expansion and latency control on the traffic using different control modes. This allows the test frames to undergo rate conversion and then enter the device under test via a high-speed port, completing the test of the device under test. After the device under test forwards the test frames back to the adapter, the adapter samples the test frames using different strategies based on its configuration. Finally, a test frame that demonstrates the forwarding traffic characteristics of the device under test is sent back to the tester via a low-speed port. The tester then performs calculations and analysis within the tester, presenting the test results.

[0007] Beneficial effects of the present invention:

[0008] In response to the need to use a low-speed TSN tester to test a high-speed device under test, the present invention proposes a time-sensitive network tester port bandwidth expansion test device and method. Without modifying the low-speed tester, the low-speed tester can be connected to a bandwidth expansion adapter, and then connected to the high-speed device under test for testing. The adapter can ensure that the delay generated during the rate conversion process is fixed, or restore the bandwidth of the received traffic after the bandwidth of the sent traffic is expanded, thereby meeting the testing requirements of the high-speed device under test. The present invention divides all network test scenarios into three categories based on whether bandwidth expansion is required, whether delay needs to be guaranteed, and whether traffic scheduling is required. The adapter will adopt different traffic processing strategies in different test scenarios to ensure that the traffic can still show key traffic characteristics after rate conversion, so that the test results can be directly analyzed through the statistical parameters of the low-speed tester, without the need for the adapter to provide additional statistical information, simplifying the system complexity and difficulty of use of the adapter.

[0009] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a schematic diagram of a time-sensitive network tester port bandwidth expansion test device provided by the present invention;

[0011] Figure 2 This is a functional module diagram of the adapter provided by the present invention;

[0012] Figure 3 This is a schematic diagram of a sending end according to requirement 1 provided by the present invention;

[0013] Figure 4 This is a schematic diagram of a receiving end according to requirement 1 provided by the present invention;

[0014] Figure 5 This is a schematic diagram of the sending end of requirement 2 provided by the present invention;

[0015] Figure 6 This is a schematic diagram of the receiving end of requirement 2 provided by the present invention;

[0016] Figure 7 This is a schematic diagram of the sending end of requirement three provided by the present invention;

[0017] Figure 8 This is a schematic diagram of the receiving end of requirement three provided by the present invention;

[0018] Figure 9 This is a schematic diagram of the Qav burst rectification test principle provided by the present invention;

[0019] Figure 10 This is a schematic diagram of the basic function test principle of Qbv gating provided by the present invention;

[0020] Figure 11a This is a schematic diagram of the Qbv door opening time test provided by the present invention;

[0021] Figure 11b This is a schematic diagram of the Qbv door opening time test provided by the present invention;

[0022] Figure 12a This is a system block diagram of the adapter transmitting end provided by the present invention;

[0023] Figure 12b This is a system block diagram of the adapter receiving end provided by the present invention. DETAILED DESCRIPTION

[0024] The present invention will be further described in detail below with reference to specific examples, but the embodiments of the present invention are not limited thereto.

[0025] Example 1

[0026] refer to Figure 1 The present invention provides a time-sensitive network tester port bandwidth expansion test device including: a converter with a high-speed port and a low-speed port; the converter is connected to the tester through the low-speed port and connected to the device under test through the high-speed port.

[0027] The adapter receives test frames sent by the tester via the low-speed port, then performs bandwidth expansion and latency control on the traffic according to different control modes based on the configuration, and finally sends the test frames to the device under test via the high-speed port. The adapter receives test frames forwarded by the device under test, then samples the test frames using different strategies based on the configuration, and finally sends test frames that can demonstrate the traffic characteristics of the device under test back to the tester via the low-speed port. The tester then performs calculations and analysis within the tester and displays the test results.

[0028] Example 2

[0029] In order to ensure the delay or scheduling mode of the test frame while performing bandwidth conversion under different strategies, the functional modules of the test system proposed in the present invention are as shown in the attached figure. Figure 2 As shown:

[0030] The converter of the present invention comprises: a test mode configuration unit, a low-speed port test unit, a high-speed port test unit, a sending end queue and counting storage unit, a sending end test frame delay control and scheduling unit, a receiving end test frame delay control and sampling unit, and a receiving end queue and storage unit;

[0031] The test mode configuration unit is used to configure a bandwidth conversion strategy; for example, whether bandwidth expansion is required, whether the delay of the test frame needs to be guaranteed, whether traffic scheduling is required, etc.

[0032] The low-speed port test unit is the interface for connecting to the tester, and the high-speed port test unit is the interface for the device under test. The number of low-speed and high-speed ports is the same and they correspond one to one. Each port can be used as a sending or receiving port.

[0033] The sending end queue and counting storage unit is responsible for storing the test frames sent by the tester according to the configuration information; wherein, the number, length and storage method of the queue are different according to different configuration information;

[0034] The transmitting end test frame delay control and scheduling unit is responsible for controlling the sending time and sending mode of the test frames in the queue;

[0035] After the test frame sent by the tester passes through the device under test, its delay and sequence will be changed by the relevant functions of the device under test, and then returned to the receiving end queue and storage unit through the high-speed port unit; the receiving end test frame delay control and sampling unit will complete the sampling of different modes of the test frame according to the configuration information, and finally send it back to the tester through the low-speed port.

[0036] Example 3

[0037] The adapter performs three different modes of bandwidth expansion and delay control on traffic according to the configuration information, namely: a first control mode that does not require bandwidth expansion but requires delay guarantee; a second control mode that requires bandwidth expansion and delay guarantee but does not require transmission scheduling; or a third control mode that requires bandwidth expansion and transmission scheduling but does not require delay guarantee;

[0038] In three different control modes, the converter has test processes corresponding to the three control modes.

[0039] Based on the requirements of TSN testing and Ethernet performance testing scenarios, the adapter scenarios can be divided into three categories based on whether bandwidth expansion is required, whether latency needs to be guaranteed, and whether transmission scheduling needs to be guaranteed. These three corresponding control modes are: 1. No bandwidth expansion is required, but latency needs to be guaranteed; 2. Bandwidth expansion and latency need to be guaranteed, but transmission scheduling is not required; 3. Bandwidth expansion and transmission scheduling are required, but latency does not need to be guaranteed. The following will introduce the transmission and reception principles of the adapter in the three scenarios:

[0040] The first control mode corresponds to scenario 1:

[0041] In the first control mode, it is necessary to ensure that the forwarding delay of the test frame in the adapter is fixed. In this way, after the test frame is sent back to the tester, the forwarding delay of the DUT can be calculated by subtracting the fixed delay of the adapter from the total delay.

[0042] As the sending end, after the adapter receives the test frame sent by the tester through the low-speed port, it waits for the maximum frame length to be sent at the low-speed port rate from the beginning of reception, which is recorded as T l Then the current test frame is sent from the high-speed port, so as to ensure that the delay of the test frame sent in the adapter is T l , as attached Figure 3 As shown;

[0043] As the receiving end, after the adapter receives the test frame forwarded by the device under test through the high-speed port, it waits for the maximum frame length to be sent at the high-speed port rate from the beginning of reception, which is recorded as T h Then, the current test frame is sent from the low-speed port to ensure that the delay of the test frame received in the adapter is T h , as attached Figure 4 shown.

[0044] The second control mode corresponds to scenario 2:

[0045] In the second control mode, the adapter does not schedule test frames, and the DUT does not change the order of received frames. Because bandwidth recovery at the receiving end occurs after receiving the last replicated frame before sending a frame to the tester, the latency of the last replicated test frame at the transmitting end must be constant.

[0046] As the sending end, after receiving the test frame sent by the tester, the adapter copies the received test frame to N times and waits 2 seconds from the start of receiving the current test frame. l Then send the last frame after copying, as shown in the attached Figure 5 As shown, the delay of the last frame in the converter is 2 l ;

[0047] As the receiving end, the adapter sends a current test frame to the tester after receiving N identical test frames from the device under test. If it does not receive N identical test frames, it is considered that the device under test has lost frames and will not send the current test frame back to the tester. In addition, the adapter needs to wait for T h Then send this frame to the tester, as shown in the attached Figure 6 As shown, to ensure that the delay of the frame in the converter is T h ;

[0048] In order to avoid the conflict of sending time when the adapter sends two adjacent frames, which will cause additional random waiting delay for the subsequent frames, the interval between the two frames at the tester sending end is Δ l The transmission time of the two adjacent frames at low rate is t A , the next frame is T B If the DUT rate is N times that of the tester, the test transmitter can be limited to

[0049]

[0050] Or only send test frames of equal length to avoid transmission conflicts.

[0051] When packet loss occurs in the DUT, it can be divided into two cases: fixed-format frame loss and uniform random frame loss. Fixed-format frame loss refers to the DUT dropping all fixed-format test frames; uniform random frame loss refers to the DUT dropping irregular frames due to the throughput being less than the data received. In the case of fixed-format frame loss, the DUT's frame loss behavior can be analyzed based on the characteristics of the dropped test frames; in the case of random frame loss, since the adapter will drop all current test frames if it cannot receive all N replicated frames, the frame loss rate obtained in the tester will be greater than the actual frame loss rate of the DUT. Let the DUT frame loss rate be y and the frame loss rate measured by the tester be x. The relationship between the two is:

[0052]

[0053] After random frame loss occurs, the actual frame loss rate of the DUT can be obtained through the above formula.

[0054] The third control mode corresponds to scenario three:

[0055] As attached Figure 7 As shown in the figure, in the third control mode, as the sending end, the adapter will copy the received test frame N times and then schedule it among multiple queues before sending the test frame; as the receiving end, each queue of the forwarder will send out the Nth frame as soon as it receives N frames. Under this requirement, after copying the received test frame N times, it still needs to be scheduled among queues before it can be sent out. Figure 7 The figure shows the test frame sent out after CBS scheduling.

[0056] At the receiving end, the frame loss rate of each priority obtained by the tester needs to be equal to the actual frame loss rate of the DUT. Therefore, the queue receiving strategy at the receiving end is changed to send the Nth frame as soon as N frames are received in each queue, as shown in the attached figure. Figure 8 shown.

[0057] The present invention provides a time-sensitive network tester port bandwidth extension test method, using a time-sensitive network tester port bandwidth extension test device, the time-sensitive network tester port bandwidth extension test method comprising:

[0058] Determine the converter's control mode for bandwidth expansion and latency control based on TSN test and Ethernet performance test scenario requirements;

[0059] The converter performs three different modes of bandwidth expansion and delay control on traffic: the first control mode does not require bandwidth expansion but requires delay guarantee; the second control mode requires bandwidth expansion and delay guarantee but does not require transmission scheduling; or the third control mode requires bandwidth expansion and transmission scheduling but does not require delay guarantee;

[0060] As the sending end, the converter receives the test frame sent by the tester and sends the test frame to the device under test according to the corresponding control mode;

[0061] As the receiving end, after receiving the test frame of the device under test, the converter returns the test frame to the tester according to the corresponding control mode.

[0062] Based on the above principles and test methods, the present invention is connected to a low-speed tester to test the high-speed TSN device under test as follows. The tester referred to in the present invention needs to have the general functions of general data link layer protocol testing and TSN protocol testing functions, and on this basis, the adapter device proposed by the present invention is connected. The basic test connection relationship remains the same as in the previous appendix. Figure 1 shown.

[0063] IEEE802.1Qav:

[0064] This protocol primarily uses a credit-based shaping (CBS) mechanism to reserve bandwidth and perform burst shaping for high-priority traffic. This protocol primarily involves two test points: bandwidth reservation and burst shaping. The following describes the test methods for these two test points in detail.

[0065] Test point 1: The bandwidth reservation test mainly uses the tester to send specific priority traffic with different bandwidths to the device under test, and determines whether the reserved bandwidth is accurate based on whether frame loss occurs in the traffic sent back by the DUT and the frame loss ratio. In the test, the DUT is first sent to each priority traffic with the same bandwidth as the DUT's configured bandwidth. In subsequent tests, traffic that exceeds the configured bandwidth of each priority is sent in turn. Statistics are then made to see whether the bandwidth of each priority traffic at the receiving end of the tester is equal to the configured bandwidth. If it is equal, the DUT bandwidth reservation function is correct. If it is less or greater, the DUT bandwidth reservation is inaccurate. In the protocol, the reserved bandwidth of the two highest priority traffic is required to be 50% and 25% of the port rate respectively. The bandwidth of the low-speed tester traffic cannot meet the test requirements. After connecting the adapter, the tester needs to configure the bandwidth of each priority traffic on the sending end to 1 / N of the DUT's configured bandwidth. After the adapter expands the bandwidth according to Mode 3 on the sending end, the test bandwidth requirement can be met. On the receiving end, the adapter needs to configure the frame drop mode to the frame drop mode in Mode 3. In this way, the receiving bandwidth of each priority frame received by the tester is 1 / N of the bandwidth sent by the DUT, and the test analysis can be completed.

[0066] The difference between the above method and direct testing with a high-speed tester is that because the adapter transmitter needs to schedule the copied test frames, the delay data in the tester will be much larger than the actual forwarding delay of the DUT. However, since this test does not refer to the delay data, it will not affect the test results.

[0067] Test point 2: The burst shaping test uses the tester to send burst traffic to the device under test under the premise that the bandwidth ratio meets the configuration. After forwarding by the DUT, the shaping effect of the DUT is judged by whether the received traffic is still sent back to the tester in a burst manner. If only a simple rate conversion is performed on the low-speed burst data stream, the low-speed burst data stream will produce a large frame interval at a high rate, making the test unfeasible. After connecting the adapter, the sending end of the adapter needs to be configured as the non-scheduling mode in mode 3, so that the data stream after bandwidth expansion is still sent to the DUT in a burst manner, as shown in the attached figure. Figure 9 As shown in the first and second time axes in the figure; the adapter receiver needs to be configured in mode 3. If the DUT can shape the traffic, the interval between the 10 consecutive frames received will also be increased, and the low-speed test frames sent by the adapter to the tester will also have an equal frame interval as shown in the attached figure. Figure 9 As shown in the third and fourth time axes in the figure, the test analysis can be completed based on the frame interval information within the test.

[0068] The difference between the above method and direct testing with a high-speed tester is that under the same frame length, the interval between the test frames sent by the adapter to the tester is the sum of the intervals of 10 frames continuously sent by the DUT. Therefore, the frame interval of the low-speed data stream received by the tester is a proportional magnification of the frame interval of the high-speed data stream sent by the DUT. Therefore, the analysis conclusion of this test is the same as that of direct testing with a high-speed tester.

[0069] IEEE802.1Qbv:

[0070] This protocol uses a time-aware gating mechanism (TAS) to send traffic of a specific priority level during specific time periods. This protocol primarily tests the correctness of the time gating mechanism and whether the parameters of the gating table meet the specifications.

[0071] Test point 1: Send frames of different priorities at equal intervals and observe whether the receiving end receives frames of different priorities in bursts according to the gating order, as shown in the attached figure. Figure 10 If yes, then the DUT has basic gating functionality. In this test, only a small number of test frames need to be sent, so the adapter needs to be configured as per requirement 1.

[0072] The difference between this method and direct testing with a high-speed tester is that the test frames sent by the DUT to the tester experience a fixed delay within the adapter, which must be subtracted when calculating the DUT's door-opening time. Subsequent bursts of test frames of the same priority level, other than the first frame, received by the tester are queued within the adapter, resulting in inaccurate latency. However, since only the reception time of the first frame is important, this does not affect the accuracy of the test results.

[0073] Test Point 2: The tester sends a burst of data streams corresponding to the length of a gating cycle to the DUT every cycle. The tester then determines the accuracy of the gate-open duration based on the length of the received data burst. If so, a second test is conducted, in which a burst of data streams exceeding the length of a gating cycle is sent every cycle to observe whether the received traffic burst length matches the gate-open duration. If so, the gate-open duration is accurate; if not, the gate-open duration is excessive. In this test, the test frames must fill the DUT's gate-open duration, so the adapter requires bandwidth extension. The transmitter is configured in Mode 3, and the receiver is configured in Mode 2 to account for frame loss.

[0074] Test result analysis: Assuming that the DUT can pass through m test frames of a certain length within the door opening time, then, ① if the tester can receive all m back-to-back test frames in each gated cycle, it means that the DUT door opening time has not shortened; otherwise, it means that the DUT door opening time has shortened. ② After the adapter receives 10 identical test frames, it sends 1 frame to the tester. If the DUT door opening time is too long, the adapter will receive several more test frames that should have been sent in the next cycle within one door opening cycle. After several cycles, it will receive the last several frames of the previous test frame and all the duplicate frames of the next two test frames, causing the adapter to send (m+1) frames to the tester in this cycle, as shown in the attached figure. Figure 10 Therefore, if the tester can receive m test frames per cycle, it indicates that the DUT's door opening time is correct. If the tester receives more than m test frames in some cycles under the premise that ① passes, it indicates that the DUT's door opening time is longer than the configured time.

[0075] The difference between the above method and direct testing with a high-speed tester is that the high-speed tester can precisely count the number of test frames forwarded by the DUT in each cycle, thereby more accurately reflecting the DUT's door opening time in each cycle; when testing through low-speed and adapters, the error of each cycle is accumulated to the length of a test frame and reflected in the current cycle. The final test conclusion is the same, but the granularity is reduced.

[0076] In other examples testing gating table parameters, only a small number of test frames were sent throughout the test, and only frame latency was of concern. Therefore, all configurations were performed according to requirement 1, and the results were identical to those obtained by directly testing with a high-speed tester.

[0077] According to the device and test method of the present invention, an embodiment of the present invention is as follows. The basic test connection relationship is still as shown in the attached figure. Figure 1 The TSN tester referred to in the present invention needs to have the general functions of general data link layer protocols and TSN testing, and on this basis, connect an adapter to perform higher rate testing.

[0078] In order to meet the above requirements, taking a 1G tester and a 10G device under test as an example, the adapter transmitting end system of the present invention is as follows: Figure 12a As shown:

[0079] The transmitting end of the adapter is mainly composed of a frame classification module, a queue module, and a scheduler module. The frame classification module has two modes. In the mode where classification is required, the test frames are divided into 8 categories according to the Priority field in the test frame VLAN tag and sent to the corresponding queues respectively; in the mode where classification is not required, the test frames are directly sent to queue 0. The queue module has a total of 8 queues. When frame replication is required, the queue needs to send 10 frames to the DUT for each frame received. Therefore, each queue needs to maintain a counter to record the number of frames that need to be sent after the first frame in the current queue; when frame replication is not required, only one queue is required, and after receiving the frame, it can wait for scheduling and sending. At the same time, the queue module also needs to complete the rate conversion of the test frame, so the queue input rate is the tester port rate of 1G, and the output rate is the DUT port rate of 10G. The scheduler has two main scheduling modes: delayed sending and credit shaping scheduling. The delayed sending function calculates the delay time based on the test frame length and the configured calculation mode, and enables queue sending after the counting ends; credit shaping scheduling is used synchronously with the tester in a mode that requires bandwidth expansion. It is necessary to maintain credit values ​​for queue 3 and queue 2 respectively according to the CBS scheduling mechanism defined in the IEEE802.1Qav protocol, and then schedule sending based on two factors: whether the queue is empty and whether the credit value is greater than 0.

[0080] At the receiving end, in order to be able to express the key characteristics of high-speed data streams through low-speed data streams in different modes, the adapter receiving end module of the present invention is as follows: Figure 12b As shown:

[0081] The receiving end of the adapter primarily consists of a frame classification module, a queue module, and a delay control module. The frame classification module, similar to the transmitting end, assigns test frames to different queues based on their priority in different modes. The queue module consists of eight queues. When bandwidth expansion is not required, or only expansion without classification is required, only queue 0 is operational. When both expansion and classification are required, each of the eight queues stores test frames of the corresponding priority. Each queue maintains a counter to record the number of identical frames received. Furthermore, the queue module compares the content of newly input frames with existing frames in the queue. During the frame input process, the content of existing frames in the queue is read and compared after input. If the two frames are identical, the counter is incremented and the new frame is deleted to prevent duplicate storage. If the two frames are different, the old frame is deleted and the counter is reset to 1, indicating the current number of new frames. Furthermore, the queue module's input rate is 10G, the DUT port rate, and its output rate is 1G, the tester port rate. The delay control module calculates the required waiting time based on the length of the test frame that meets the transmission requirements and transmits the test frame after counting and timing.

[0082] Before starting the test, you must first configure the stream processing mode of the adapter's sending and receiving ends. On the sending end, it is necessary to configure the sending mode based on whether the test focuses on latency, whether bandwidth expansion is required, and whether different priority test traffic needs to be shaped before sending. The adapter's traffic scheduling and sending method for each priority level should be the same as that of the tester; on the receiving end, it is necessary to configure the receiving mode based on whether the test focuses on latency, whether bandwidth reduction is required, and whether frame loss needs to be reflected proportionally. After the configuration is complete, you can connect the low-speed tester and the high-speed device under test for testing. In this embodiment, 1G tester and 10G device under test are taken as examples. Tests of other different rates are based on the same principles as this embodiment and are also within the scope of protection of this patent. The following will introduce the embodiments of the three scenarios described in the principle part.

[0083] Scenario 1:

[0084] In the test where the test requirements meet scenario 1, after the transmitter receives the test frame, the frame classification module directly sends it to queue 0. The queue module stores it at 1G rate and waits for the scheduler enable signal to send it out at 10G rate. The scheduler starts timing when the frame classification module receives the first bit and waits for T l Then send an enable signal to the queue.

[0085] After receiving the test frame forwarded by the DUT, the frame classification module directly places it into queue 0. The queue module stores the frame at 10G and waits for the scheduler's enable signal before sending it out at 1G. The scheduler starts counting when the frame classification module receives the first bit and waits for Th to send the queue enable signal.

[0086] The delay time between the sending and receiving ends of the adapter is calculated as follows:

[0087] The adapter delay needs to take into account the physical port occupancy time when the test frame is transmitted. Therefore, the transmission delay time needs to take into account the fixed overhead of the Ethernet test frame, including the Ethernet minimum interframe interval, preamble and frame delimiter, a total of 20B. The maximum Ethernet frame length is L max , the fixed overhead length of the frame is L inte , then the delay at the sending end of the adapter is T send for:

[0088]

[0089] Receiver delay T recv for:

[0090]

[0091] Therefore, the DUT forwarding delay T DUT Subtracting the switch delay from the total delay is:

[0092] T DUT =T total -T send -T recv =T total -13.5344us

[0093] Scenario 2:

[0094] In the test that meets the requirements of scenario 2, after the sender receives the test frame, the frame classification module directly sends it to queue 0. After the queue stores it at a rate of 1G, it sets the counter to 10 and waits for the enable signal to be received before sending 10 frames back to back. The time when the scheduler receives the first bit is taken as the starting point, the current frame length is recorded as L, and the port rates are P 1G and P 10G , in order to ensure that the last frame after copying is sent for 2T l , we can get the sending time of the first frame as When the timer reaches this point, 10 frames are sent back to back.

[0095] After the receiving end receives the test frame forwarded back by the DUT, the frame classification module directly sends it to queue 0. During the receiving process, the queue module reads the contents of the existing frames in the queue and compares them with the current frame. After the reception is completed, if the two are the same, the counter is increased by one and the new frame is deleted to avoid duplicate storage; if the two are different, the old frame is deleted and the counter is set to 1, which is the number of new frames at this time. When the counter is 9, the scheduling module timer will start timing when the first bit of the next frame is received. If the frame is the same as the frame already in the queue, it means that all 10 copied frames have been received, and the timer is set to T h Then enable the queue to send the frame.

[0096] According to the above principle analysis, in order to ensure that the delay in the adapter is fixed, the tester transmitter needs to ensure Or send test frames of equal length. Similar to the analysis in scenario 1, the delay T of the last frame after copying at the sending end of the adapter is send for:

[0097]

[0098] Receiver delay T recv for:

[0099]

[0100] Therefore, the DUT forwarding delay T DUT Subtract the switch delay from the total delay:

[0101] T DUT =T total -Tsend -T recv =T total -25.8384us

[0102] At the same time, if the adapter receives 10 frames and then sends one, and the DUT experiences random frame loss, the relationship between the DUT's actual frame loss rate y and the tester's measured frame loss rate x is:

[0103]

[0104] The approximate calculation can be performed by taking the low-order terms after Taylor expansion:

[0105]

[0106] Scenario 3:

[0107] In the test scenario 3, after the transmitter receives the test frame, the frame classification module places it into the corresponding queue based on its priority. Upon receiving the test frame, the queue module maintains a counter with an initial value of 10 for the frame. Each time a frame is sent from the queue, the counter is decremented until it reaches 0, at which point the frame is removed from the queue. The scheduler module then schedules according to the CBS rule, maintaining credit values ​​for queues 3 and 2, respectively. It then schedules based on whether the queue is empty and whether the queue credit value is greater than 0.

[0108] After receiving the test frame forwarded back by the DUT, the frame classification module at the receiving end sends it to the corresponding queue based on the priority information. In this mode, since the queue will send the 10th frame after receiving 10 frames without determining whether the contents of the 10 frames are identical, each queue maintains a counter that increments by 1 with each received frame. When the counter value is between 0 and 8, the queue directly discards the currently received test frame. When the counter value reaches 9, the queue stores the next incoming test frame and sends it at a 10G rate.

[0109] In this test scenario, there is no need to consider the test frame forwarding delay. At the same time, the frame loss rate measured by the tester is the same as the actual frame loss rate of the DUT, and no conversion is required.

[0110] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. Moreover, the above-mentioned concepts such as "frame", "frame loss", "frame loss rate", etc. are fully applicable to the relevant descriptions of other encapsulation units used for network data transmission, such as "packet", "packet loss" and "packet loss rate".

[0111] Although the application has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations using the principles of the application. For example, "comprising" as used throughout the specification and in each claim is not meant to exclude other components or steps. The terms "a" or "an" as used herein mean "one or more."

[0112] The above description is further to specific preferred embodiments of the present application and is not to be construed in any way as limiting the present application. Those skilled in the art who possess the benefit of this disclosure will appreciate that many other variations from the foregoing description and drawings of specific embodiments of the present application can be made without departing from the scope of the present application. Accordingly, it is intended that all such alternatives, modifications and variations be included within the scope of the present application.

Claims

1. A time-sensitive network tester port bandwidth expansion test device, characterized in that: include: A converter with a high-speed port and a low-speed port; The converter is connected to a low-speed tester via a low-speed port and is connected to a high-speed device under test via a high-speed port; The tester sends a test frame to the adapter via a low-speed port. The adapter performs bandwidth expansion and delay control on the traffic in different control modes according to the configuration, so that the test frame, after rate conversion by the converter, enters the device under test through the high-speed port to complete the test of the device under test. After the device under test forwards the test frame back to the adapter, the adapter samples the test frame using different strategies according to the configuration, and finally sends the test frame that can reflect the forwarding traffic characteristics of the device under test back to the tester via the low-speed port. Finally, the test result is displayed after calculation and analysis in the tester. The adapter performs three modes of bandwidth expansion and delay control on the traffic according to the configuration information, namely: a first control mode that does not require bandwidth expansion but requires delay guarantee; a second control mode that requires bandwidth expansion and delay guarantee but does not require transmission scheduling; or a third control mode that requires bandwidth expansion and transmission scheduling but does not require delay guarantee; Under the three control modes, the converter has test processes corresponding to the three control modes; In the second control mode, As the sending end, after receiving the test frame sent by the tester, the adapter will copy the received test frame N times and wait for the test frame to be received from the beginning. The last frame after copying is sent to ensure that the delay of the last frame in the converter is ; As the receiving end, the adapter sends a current test frame to the tester after receiving N identical test frames from the device under test. If it does not receive N identical test frames, it is considered that the device under test has lost data frames and will not send the current test frame back to the tester. The adapter needs to wait when receiving the Nth frame Then send this frame to the tester to ensure that the delay of the frame in the converter is ; Among them, the test frame interval sent by the tester to the adapter is , the sending time of the two adjacent frames at the low-speed port is and , subscript A is the previous frame, B is the next frame, the high-speed port rate of the device under test is N times the low-speed port rate of the tester, Needs to be satisfied , or only send equal-length test frames for testing to avoid conflicts between adapters; When data frame loss occurs in the device under test, it is divided into fixed frame format frame loss and uniform random frame loss; Fixed frame format loss refers to the device under test losing all test frames of the fixed format; uniform random frame loss refers to the device under test losing frames at random due to the throughput being less than the data received. The tester performs corresponding actual frame loss rate analysis based on two situations: fixed frame format frame loss and uniform random frame loss.

2. The time-sensitive network tester port bandwidth extension test device according to claim 1, characterized in that: The converter includes: a test mode configuration unit, a low-speed port test unit, a high-speed port test unit, a sending end queue and count storage unit, a sending end test frame delay control and scheduling unit, a receiving end test frame delay control and sampling unit, and a receiving end queue and storage unit; The test mode configuration unit is used to configure a bandwidth conversion strategy; The low-speed port test unit is the interface for connecting to the tester, and the high-speed port test unit is the interface for the device under test. The number of low-speed and high-speed ports is the same and they correspond one to one. Each port can be used as a sending or receiving port. The sending end queue and counting storage unit is responsible for storing the test frames sent by the tester according to the configuration information; wherein, the number, length and storage method of the queue are different according to different configuration information; The transmitting end test frame delay control and scheduling unit is responsible for controlling the sending time and sending mode of the test frames in the queue; After the test frame sent by the tester is forwarded by the device under test, its delay and sequence will be changed by the relevant functions of the device under test, and then returned to the receiving end queue and storage unit through the high-speed port unit; the receiving end test frame delay control and sampling unit will complete the sampling of different modes of the test frame according to the configuration information, and finally send it back to the tester through the low-speed port.

3. The time-sensitive network tester port bandwidth extension test device according to claim 2, characterized in that: In the first control mode, As the sending end, after the adapter receives the test frame sent by the tester through the low-speed port, it waits for the maximum frame length to be sent at the low-speed port rate from the beginning of reception, which is recorded as , then send the current test frame from the high-speed port, so as to ensure that the delay of the test frame sent in the adapter is ; As the receiving end, after the adapter receives the test frame forwarded by the device under test through the high-speed port, it waits for the maximum frame length to be sent at the high-speed port rate from the start of reception, which is recorded as Then the current test frame is sent out from the low-rate port to ensure that the delay of the test frame received in the adapter is .

4. The time-sensitive network tester port bandwidth extension test device according to claim 2, characterized in that: In the case of uniform random frame loss, the actual frame loss rate is: Among them, the frame loss rate measured by the tester is .

5. The time-sensitive network tester port bandwidth extension test device according to claim 2, characterized in that: In the third control mode, As the sending end, the adapter copies the received test frame N times and then schedules it in multiple queues before sending the test frame. As the receiving end, each queue of the forwarder sends out the Nth frame as soon as it receives N frames. The frame loss rate obtained by the tester is equal to the actual frame loss rate of the device under test.

6. A method for testing port bandwidth expansion of a time-sensitive network tester, characterized in that: Using the time-sensitive network tester port bandwidth extension test device according to any one of claims 1 to 5, the time-sensitive network tester port bandwidth extension test method comprises: Determine the converter's control mode for bandwidth expansion and latency control based on TSN test and Ethernet performance test scenario requirements. The converter performs three different modes of bandwidth expansion and delay control on traffic: the first control mode does not require bandwidth expansion but requires delay guarantee; the second control mode requires bandwidth expansion and delay guarantee but does not require transmission scheduling; or the third control mode requires bandwidth expansion and transmission scheduling but does not require delay guarantee; As the sending end, the converter receives the test frame sent by the tester and sends the test frame to the device under test according to the corresponding control mode; As the receiving end, after receiving the test frame of the device under test, the converter returns the test frame to the tester according to the corresponding control mode.

Images