Switching efficiency measurement method and device, switch, equipment and medium

Abstract

The invention provides a switching efficiency measuring method, a switching efficiency measuring device, a switch, computing equipment and a computer readable storage medium. The method is applied to a switch, the switch realizes switching between a plurality of sending ends and a plurality of receiving ends, and the method comprises the following steps: measuring switch switching overhead time by sending a message sequence from each sending end to the plurality of receiving ends; measuring phase switching overhead time by sending a message sequence from a plurality of sending ends to each receiving end; and calculating the switching efficiency of the switch according to the switch switching overhead time and the phase switching overhead time. According to the application, the switch can be utilized to realize the measurement of the switching efficiency, the measurement accuracy is high, and the use is convenient.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method for measuring switching efficiency, a device for measuring switching efficiency, a switch, a computing device, and a computer-readable storage medium. Background Art

[0002] Optical switches are one of the fundamental devices in the field of optical communication. Techniques for measuring the switching efficiency of optical switches exist in this field. However, existing optical switch switching efficiency measurement techniques require dedicated equipment for implementation, and cannot measure switching efficiency using the switch itself. Furthermore, the measurement results and ease of use are unsatisfactory. Therefore, there is an urgent need in this field for a switching efficiency measurement method suitable for optical switches that can measure switching efficiency using the switch itself, without requiring dedicated measurement equipment, and offers high accuracy and ease of use. Summary of the Invention

[0003] Therefore, this application aims to provide a method, device, switch, computing device, and computer-readable storage medium for measuring switching efficiency, which can measure switching efficiency using the switch itself, with high accuracy and ease of use.

[0004] In one aspect, this application provides a method for measuring switching efficiency, characterized in that it is applied to a switch, which performs switching between multiple transmitters and multiple receivers. The method includes: measuring the switching overhead time by sending a message sequence from each transmitter to multiple receivers; measuring the phase switching overhead time by sending a message sequence from multiple transmitters to each receiver; and calculating the switching efficiency of the switch based on the switching overhead time and the phase switching overhead time.

[0005] According to this aspect, by sending message sequences between multiple sending ends and multiple receiving ends according to certain rules, the switching time overhead and phase switching overhead can be accurately calculated. This enables the technology of this application to measure switching efficiency through the switch's own equipment, avoiding the use of dedicated equipment and software for measuring switching efficiency. This reduces the cost and complexity of switching efficiency measurement and improves the accuracy and ease of use of switching efficiency measurement.

[0006] In a specific embodiment of this application, the switch has N transmitting ends and N receiving ends. The switching overhead time is measured by sending message sequences from each transmitting end to multiple receiving ends, including: sending message sequences from the first to the Nth transmitting end to each of the N receiving ends, respectively, to obtain N switching overhead times corresponding to each of the N transmitting ends. The phase switching overhead time is measured by sending message sequences from multiple transmitting ends to each receiving end, including: sending message sequences from the N transmitting ends to the first to the Nth receiving ends, respectively, to obtain N phase switching overhead times corresponding to each of the N receiving ends. The switching efficiency of the switch is calculated based on the switching overhead time and the phase switching overhead time, including: determining the maximum value among the N switching overhead times and the N phase switching overhead times; and calculating the switching efficiency of the switch based on the maximum value.

[0007] According to this embodiment, assuming the number of transmitters and receivers is equal and both are N, by sending a message sequence from one transmitter to N receivers, the switching overhead time corresponding to that transmitter can be measured. Similarly, by sending a message sequence from N transmitters to one receiver, the phase switching overhead time corresponding to that receiver can be measured. In other words, through "one-to-many" and "many-to-one" methods, the switching time overhead and phase time overhead can be accurately measured. The switching efficiency of the switch can then be determined based on their maximum values. This measurement method is simple and easy to implement, achieving accurate measurement of switching time overhead and efficiency without the need for specialized equipment, and offers excellent cost and economic benefits.

[0008] In a particular embodiment of this application, a message sequence is sent from the first to the Nth of N transmitters to N receivers to obtain N switch switching overhead times corresponding to the N transmitters, including: sending a message sequence from the first of the N transmitters to the N receivers to obtain one switch switching overhead time corresponding to the first transmitter. The process of sending a message sequence from the first of N sending ends to N receiving ends to obtain the switching overhead time corresponding to the first sending end includes: constructing a message sequence, which consists of multiple message periods of equal duration, each message period including a message part and a blank part, and the header of the message sequence having an adjustable-duration header blank part; sending the message sequence from the first sending end; opening the N switching switches corresponding to the N receiving ends sequentially at time intervals based on the duration of the message period, so as to allow the message sequence to reach the N receiving ends respectively; adjusting the duration of the header blank part to determine the maximum and minimum duration of the header blank part while ensuring that all N receiving ends receive the message part completely; and calculating the switching overhead time corresponding to the first sending end based on the maximum and minimum duration values.

[0009] According to this embodiment, by setting the structure of the message sequence and sending the constructed message from one sender to multiple receivers, the maximum and minimum durations for complete message transmission and reception are determined. This allows for the measurement of the time range within which the message can be transmitted completely. Based on this time range and the effective message length, the switching overhead can be calculated. This measurement method is simple and easy to implement; it only requires constructing a message and observing its integrity to accurately measure the switching time overhead. No dedicated equipment is needed for measurement, resulting in good measurement efficiency and cost-effectiveness.

[0010] In a particular embodiment of this application, the switching overhead time corresponding to the first transmitter is calculated based on the maximum and minimum duration values, including: calculating the difference between the maximum and minimum duration values; calculating the sum of the difference and the duration of the message portion; and calculating the difference between the duration of the message period and the sum, thereby determining the switching overhead time corresponding to the first transmitter.

[0011] According to this embodiment, the time range during which a message is allowed to pass and be received is calculated by the difference between the maximum and minimum duration values. This allows for the measurement of the switching overhead in the switching matrix. This calculation method is accurate and simple, does not consume too many computing resources, and has good computational efficiency and cost-effectiveness.

[0012] In a particular embodiment of this application, sending message sequences from N transmitters to the first to Nth receivers among N receivers to obtain N phase switching overhead times corresponding to the N receivers includes: sending message sequences from N transmitters to the first receiver among N receivers to obtain one phase switching overhead time corresponding to the first receiver. The process of sending message sequences from N transmitters to the first receiver among N receivers to obtain the phase switching overhead time corresponding to the first receiver includes: constructing message sequences, each consisting of multiple message periods of equal duration, each message period including a message portion and a blank portion; alternately sending N message sequences from the N transmitters; sequentially opening the N switching switches corresponding to the N transmitters at time intervals, allowing the N message sequences to arrive at the first receiver; sequentially switching phases at the first receiver at time intervals, receiving the N message sequences in sequence; adjusting the start time of phase switching at the first receiver to determine the earliest and latest times while ensuring that the first receiver receives the complete message portion of the N message sequences; and calculating the phase switching overhead time corresponding to the first receiver based on the earliest and latest times.

[0013] According to this embodiment, by setting the structure of the message sequence and sending the constructed messages from multiple senders to one receiver, the earliest and latest times when the messages can be completely sent and received are determined. This allows for the measurement of the time range within which the messages can be completely transmitted, and the phase switching overhead can be calculated based on this time range and the effective message length. This measurement method is simple and easy to implement; it only requires constructing messages and observing their integrity to accurately measure the phase switching time overhead. No dedicated equipment is required for measurement, resulting in good measurement efficiency and cost-effectiveness.

[0014] In a particular embodiment of this application, the phase switching overhead time corresponding to the first receiving end is calculated based on the earliest time and the latest time, including: calculating the difference between the earliest time and the latest time; calculating the sum of the difference and the duration of the message portion; and calculating the difference between the duration of the message period and the sum, thereby determining the phase switching overhead time corresponding to the first receiving end.

[0015] According to this embodiment, the time range during which a message is allowed to pass and be received is calculated by the difference between the earliest and latest times. This allows for the measurement of the switching overhead in the switching matrix. This calculation method is accurate and simple, does not consume too many computing resources, and has good computational efficiency and cost-effectiveness.

[0016] In a particular embodiment of this application, calculating the switching efficiency of a switch based on the maximum value includes: determining the maximum overhead time of the switch based on the maximum value; calculating the maximum message length that the switch is allowed to transmit based on the maximum overhead time; and calculating the switching efficiency of the switch based on the maximum message length.

[0017] According to this embodiment, the maximum overhead time of the switch is determined to determine the length of the message that the switch is allowed to transmit, and finally the switching efficiency of the switch is determined. This calculation method is simple and easy to implement, and can produce accurate calculation results, with high calculation efficiency and economic benefits.

[0018] On the other hand, this application provides a switching efficiency measurement device, which is applied to a switch that performs switching between multiple transmitters and multiple receivers. The device includes: a first measurement module for measuring the switching overhead time by sending a sequence of messages from each transmitter to multiple receivers; a second measurement module for measuring the phase switching overhead time by sending a sequence of messages from multiple transmitters to each receiver; and a calculation module for calculating the switching efficiency of the switch based on the switching overhead time and the phase switching overhead time.

[0019] In a particular embodiment of this application, the switch has N transmitting ends and N receiving ends. The first measurement module is further configured to: send message sequences from the 1st to the Nth transmitting end to the N receiving ends respectively, obtaining N switchover overhead times corresponding to the N transmitting ends. The second measurement module is further configured to: send message sequences from the N transmitting ends to the 1st to the Nth receiving ends respectively, obtaining N phase switching overhead times corresponding to the N receiving ends. The calculation module is further configured to: determine the maximum value among the N switchover overhead times and the N phase switching overhead times; and calculate the switching efficiency of the switch based on the maximum value.

[0020] In a particular embodiment of this application, the first measurement module is further configured to: send a message sequence from the first of N transmitters to N receivers to obtain the switching overhead time corresponding to the first transmitter. The first measurement module is further configured to: construct a message sequence comprising multiple message periods of equal duration, each message period including a message portion and a blank portion, the header of the message sequence having an adjustable-duration header blank portion; send the message sequence from the first transmitter; sequentially open the N switching switches corresponding to the N receivers at time intervals, using the duration of the message period as the time interval, to allow the message sequence to reach the N receivers respectively; adjust the duration of the header blank portion to determine the maximum and minimum duration of the header blank portion while ensuring that all N receivers fully receive the message portion; and calculate the switching overhead time corresponding to the first transmitter based on the maximum and minimum durations.

[0021] In a particular embodiment of this application, the first measurement module is further configured to: calculate the difference between the maximum and minimum duration; calculate the sum of the difference and the duration of the message portion; calculate the difference between the duration of the message period and the sum, thereby determining the switching overhead time corresponding to the first transmitter.

[0022] In a particular embodiment of this application, the second measurement module is further configured to: send a message sequence from N transmitters to the first receiver among N receivers, and obtain a phase switching overhead time corresponding to the first receiver. The second measurement module is further configured to: construct a message sequence, the message sequence including multiple message periods of equal duration, each message period including a message portion and a blank portion; alternately send N message sequences from the N transmitters; sequentially open the N switching switches corresponding to the N transmitters at time intervals based on the duration of the message periods, to allow the N message sequences to arrive at the first receiver respectively; sequentially switch the phase at the first receiver at time intervals based on the duration of the message periods, to sequentially receive the N message sequences; adjust the time at which the first receiver begins switching the phase, to determine the earliest and latest times while ensuring that the first receiver completely receives the message portions of the N message sequences; and calculate the phase switching overhead time corresponding to the first receiver based on the earliest and latest times.

[0023] In a particular embodiment of this application, the second measurement module is further configured to: calculate the difference between the earliest time and the latest time; calculate the sum of the difference and the duration of the message portion; calculate the difference between the duration of the message period and the sum, thereby determining the phase switching overhead time corresponding to the first receiving end.

[0024] In a particular embodiment of this application, the calculation module is further configured to: determine the maximum overhead time of the switch based on the maximum value; calculate the maximum message length allowed to be transmitted by the switch based on the maximum overhead time; and calculate the switching efficiency of the switch based on the maximum message length.

[0025] On the other hand, this application provides a switch, which includes multiple transmitters, multiple receivers, a main controller, and a memory. The switch enables switching between the multiple transmitters and multiple receivers. The main controller is used to execute a computer program stored in the memory to implement the above-mentioned switching efficiency measurement method.

[0026] On the other hand, this application provides a computing device including a processor and a memory, the processor being used to execute a computer program stored in the memory to implement the above-described method for measuring switching efficiency.

[0027] On the other hand, this application provides a computer-readable storage medium storing a computer program for performing the above-described method for measuring exchange efficiency.

[0028] On the other hand, this application provides a computer program product, including program code, which, when a computer runs the computer program product, causes the computer to implement the above-described method for measuring exchange efficiency.

[0029] Any of the switching efficiency measurement devices, switches, computing devices, computer-readable storage media, or computer program products provided above are used to perform the switching efficiency measurement methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects of the corresponding schemes in the corresponding methods provided above, and will not be repeated here. Attached Figure Description

[0030] The specific embodiments of this application are described in detail below with reference to the accompanying drawings, wherein:

[0031] Figure 1 This invention illustrates a schematic diagram of the architecture of an optical switching system according to an embodiment of the present application.

[0032] Figure 2 This invention illustrates a schematic diagram of the architecture of a 2×2 switching system according to an embodiment of the present application.

[0033] Figure 3 Showing according to Figure 2 A schematic diagram of the link overhead of the switching system in the embodiment;

[0034] Figure 4 This diagram illustrates the structure of an optical switching system according to an embodiment of the present application.

[0035] Figure 5 A schematic flowchart of a method for measuring switching efficiency according to an embodiment of this application is shown;

[0036] Figure 6 A flowchart illustrating a method for measuring the switch switching overhead time corresponding to the first transmitting end according to an embodiment of this application is shown.

[0037] Figure 7 Showing according to Figure 6 A schematic diagram of the hardware architecture of the measurement method in the embodiment;

[0038] Figure 8 Showing according to Figure 6 A schematic diagram of the message sequence corresponding to the maximum duration of the header blank portion in the measurement method of the embodiment;

[0039] Figure 9 Showing according to Figure 6 A schematic diagram of the message sequence corresponding to the minimum duration of the header blank portion in the measurement method of the embodiment;

[0040] Figure 10 A flowchart illustrating a method for measuring the phase switching overhead time corresponding to the first receiving end according to an embodiment of this application is shown.

[0041] Figure 11 Showing according to Figure 10 A schematic diagram of the hardware architecture of the measurement method in the embodiment;

[0042] Figure 12 Showing according to Figure 10 A schematic diagram of the message sequence of the measurement method in the embodiment;

[0043] Figure 13 This diagram shows a schematic representation of a switching efficiency measuring device according to an embodiment of the present application.

[0044] Figure 14 A schematic diagram of the structure of a computing device according to an embodiment of this application is shown. DETAILED DESCRIPTION

[0045] To enable those skilled in the art to more clearly understand the concepts and ideas of this application, the application is described in detail below with reference to specific embodiments. It should be understood that the embodiments given herein are only a part of all possible embodiments of this application. After reading the specification of this application, those skilled in the art are capable of making improvements, modifications, or substitutions to parts or the entirety of the following embodiments, and such improvements, modifications, or substitutions are also included within the scope of protection claimed in this application.

[0046] In this document, the terms "one," "an," and other similar words are not intended to indicate that only one of the described things exists, but rather that the description refers only to one of the described things, which may have one or more. In this document, the terms "comprising," "including," and other similar words are intended to indicate a logical relationship, not a spatial relationship. For example, "A includes B" means that logically B belongs to A, not that spatially B is located inside A. Furthermore, the meanings of the terms "comprising," "including," and other similar words should be considered open-ended, not closed-ended. For example, "A includes B" means that B belongs to A, but B does not necessarily constitute all of A; A may also include other elements such as C, D, and E.

[0047] In this document, the terms "first," "second," and other similar terms are not intended to imply any order, quantity, or importance, but are merely used to distinguish different elements. In this document, the terms "embodiment," "this embodiment," "an embodiment," or "an example" do not indicate that the description applies only to one specific embodiment, but rather that such description may also be applicable to one or more other embodiments. Those skilled in the art will understand that any description made herein with respect to one embodiment can be substituted, combined, or otherwise combined with the descriptions in one or more other embodiments, and the new embodiments resulting from such substitutions, combinations, or other combinations are readily conceived by those skilled in the art and fall within the scope of protection of this application.

[0048] In the various embodiments of this application, a switch can refer to a device that, according to the information transmission needs of both communication ends, sends the information to be transmitted to the corresponding route that meets the requirements, either manually or automatically. For example, a switch can refer to a network device that can connect multiple devices to a computer network and forward data to its destination through packet switching. An optical switch can refer to a device that can perform data switching of optical signals, multiplexing and demultiplexing data signals to ensure network reliability and provide a flexible signal routing platform. For example, optical switches can have four switching modes: space-division optical switching, time-division optical switching, wavelength-division optical switching, and composite optical switching.

[0049] In various embodiments of this application, switching efficiency can refer to the ability and efficiency of a switch to transmit packets, such as the maximum packet length that a switch can transmit in each time slice, thereby calculating the packet transmission capacity of the switch. Switching efficiency measurement can refer to determining the packet transmission capacity of a switch, such as measuring the maximum packet length that a switch can transmit in each time slice, thereby estimating the maximum switching efficiency of the switch.

[0050] Optical switching is a next-generation switching technology. Its system mainly consists of a main controller, a NIC (Network Interface Card), and an optical switching matrix. A typical optical switching system architecture is as follows: Figure 1 As shown. Figure 1 This diagram illustrates the architecture of an optical switching system according to an embodiment of this application. The NICs (including NIC0, NIC1, ..., NICn) are used to interface with user equipment (such as servers), guiding user traffic into the switching network. The traffic is then converted into optical signals and transmitted via optical fibers (including fiber 0, fiber 1, ..., fiber n) to the optical switching matrix for switching. The optical switching matrix is ​​an array of optical switches that, based on control signals input from the main controller, control the corresponding optical switches to open or close, mapping different optical paths to achieve N-input to N-output optical path switching. The main controller controls the timing of data transmission by the NICs, switching the optical links by controlling the optical switching matrix to switch according to slice (time-slice) periods.

[0051] Because it's difficult to guarantee consistent arrival times of optical network interface cards (NICs) at the optical matrix, and consistent switching times of different switches on the optical matrix, the allowed transmission time lengths for different links can vary within a single slice cycle. For the entire system, the maximum allowed transmission packet length must be less than the minimum allowed transmission time length across different links; at this point, the switching system reaches its maximum switching efficiency. In other words, the maximum allowed transmission packet length for the entire optical switching system must be determined based on the maximum switching efficiency.

[0052] Figure 2A schematic diagram of the architecture of a 2×2 switching system according to an embodiment of this application is shown. Figure 2 As shown, the 2×2 switching system includes a main controller, two transmitting network interface cards (NIC0 and NIC1 on the left), and two receiving network interface cards (NIC0 and NIC1 on the right). There are four links between them, and each link is controlled by a switch in the optical switching matrix. There are a total of four switches (S). 00 S 01 S 10 S 11 ).

[0053] Figure 3 Showing according to Figure 2 A schematic diagram of the link overhead of the switching system in this embodiment. Figure 3 The diagram illustrates the link costs between the sending network interface card (NIC0) and the receiving network interface card (NIC0), between the sending NIC0 and the receiving NIC1, between the sending NIC1 and the receiving NIC0, and between the sending NIC1 and the receiving NIC1. Of these four types of costs, the cost between the sending NIC0 and the receiving NIC1 is the highest; therefore, this cost is taken as the maximum overhead time (Toverhead) of the entire switching system. Based on this, the maximum packet length (Tpkt) can be calculated. The switching efficiency of the system at this point is Tpkt / Toverhead.

[0054] Optical switching technology is the core technology of optical switching systems, and optical switching efficiency is the core performance indicator of the entire optical switching system. Therefore, it is crucial to calibrate and test the maximum optical switching efficiency of an optical switching system.

[0055] In one technique, the switching efficiency of a switching system is estimated by testing the switching overhead and the clock and data recovery (CDR) time at the receiver in a single-point test of the optical switching matrix. Furthermore, dedicated equipment and software are used to measure the maximum overhead time of different switches in the optical switching matrix and the maximum overhead time of phase switching in the receiver's CDR circuit. However, the fiber optic delay measurement scheme based on dedicated measuring instruments has the following problems: 1) The maximum switching overhead and the maximum overhead time of CDR phase switching differ from the overhead time calculated by single-point measurement, and are also affected by the path delay of different switch configurations and the time difference in the arrival time of messages at the CDR from different paths; 2) The measurement efficiency is low, requiring advance measurement of switching overhead and CDR recovery time, and cannot be automated in real time; 3) Only the maximum system switching efficiency can be estimated, but the system still needs calibration to achieve the maximum switching efficiency.

[0056] In some embodiments of this application, a scanning-based method for measuring the switching efficiency (switching overhead and phase switching overhead) of an optical switching system is proposed. This method can achieve rapid and high-precision measurement and calibration of the switching efficiency of an optical switching system without the aid of special instruments and equipment, thereby guiding the determination of the maximum length of the system's switched messages and achieving the maximum switching efficiency.

[0057] In some embodiments of this application, since the optical switching efficiency testing methods in the art are based on single-point measurement of optical switching overhead time and CDR recovery time, and the maximum optical switching efficiency of the system is approximately calculated, they cannot meet the application requirements of large-scale optical switching scenarios, and the measurement accuracy is not high. Therefore, a method for measuring the switching efficiency of optical switching systems is proposed. This method accurately obtains the switching efficiency of the optical switching system through a "one-to-many" switch switching overhead measurement mechanism and a "many-to-one" phase switching overhead measurement mechanism.

[0058] In some embodiments of this application, a scanning-based method for measuring the switching efficiency of an optical switching system is proposed. This method can achieve rapid and high-precision measurement and calibration of the switching efficiency of the optical switching system without the aid of specialized instruments and equipment, thereby guiding the maximum length of the system's switched messages and achieving maximum switching efficiency. Furthermore, a method for measuring the switching overhead based on a "one-to-many" optical network interface card (NIC) is proposed, realizing automated measurement of switching overhead and calibration of the NIC's message transmission timing. Also, a method for measuring the phase switching overhead based on a "many-to-one" optical NIC is proposed, realizing automated measurement of phase switching overhead and calibration of the NIC's phase switching timing. Finally, a method for calculating the switching efficiency and maximum message length of an optical switching system based on switching overhead and phase switching overhead is proposed.

[0059] Figure 4 A schematic diagram of an optical switching system according to an embodiment of this application is shown. Figure 4 As shown, the data stream enters the switching network from the NIC. After being scheduled by the optical switching controller, the NIC converts the data stream into an optical signal and enters the optical switching matrix for switching. After the switching is completed, the optical signal enters the corresponding NIC and is converted into an electrical signal before being output to the switching network. In some embodiments of this application, the alignment of incoming packets to the optical switching matrix can be achieved, so that packets (in optical signals) passing through different NICs in each cycle arrive at the optical switching matrix simultaneously.

[0060] like Figure 4As shown, the optical switching system architecture mainly includes a main controller, a NIC, an optical switching matrix, and optical fibers. The main controller is responsible for achieving high-precision clock synchronization for the entire system, ensuring that all units within the system operate at the same pace. The main controller also controls the data transmission time from the NIC to the switching matrix and the switching matrix's on / off timing. The NIC sends messages to the switching matrix in the uplink direction at times specified by the controller; in the downlink direction, the NIC switches its CDR phase and receives messages from the switching matrix output at times specified by the controller, then parses and verifies the messages; the messages sent by the NIC are converted into optical signals by the optical modules and then transmitted to the optical switching matrix. Under the control of the main controller, the optical switching matrix opens at fixed times, allowing subsequent messages to pass through the switching matrix. Figure 4 As shown, each NIC includes control logic, SerDes (SERializer / DESerializer), and an optical module. The control logic module is used to control the overall service logic of the NIC, the SerDes is used to send and receive messages and is related to parameters such as the phase of the messages, and the optical module is used to perform photoelectric conversion.

[0061] Figure 5 A schematic flowchart of a method for measuring switching efficiency according to an embodiment of this application is shown.

[0062] According to this embodiment, the switching efficiency measurement method is applied to a switch, which performs switching between multiple transmitting ends and multiple receiving ends. The method includes steps S510 to S530, each of which is described in detail below.

[0063] S510: Measure the switching overhead time by sending message sequences from each transmitter to multiple receivers.

[0064] In this embodiment, the switching overhead time can refer to the time it takes for switches in the optical switching matrix to turn on and off to switch the optical signal to different links.

[0065] In this embodiment, by having each of the multiple transmitters send a message sequence to multiple receivers, the corresponding switchover overhead time for each transmitter can be calculated. Repeating this process multiple times yields multiple switchover overhead times for each transmitter.

[0066] As an example, when a switch has N transmitters and N receivers, in order to measure the switch switching overhead time by sending a sequence of messages from each transmitter to multiple receivers, a sequence of messages can be sent from the first to the Nth transmitter to each of the N receivers to obtain N switch switching overhead times for each of the N transmitters. This includes sending a sequence of messages from the first transmitter to the N receivers to obtain one switch switching overhead time for the first transmitter.

[0067] According to this example, starting from the first of N transmitters, a message sequence can be sent from the first transmitter to the N receivers, then from the second transmitter to the N receivers, and so on, until the Nth transmitter sends a message sequence to the N receivers. That is, by using a "one-to-many" approach, this process is repeated N times to obtain the switching overhead time for each of the N transmitters.

[0068] For the specific measurement and calculation method of the switch switching overhead time corresponding to the first transmitter in this example, please refer to the following text. Figures 6 to 9 The relevant descriptions should be understood. It should be understood that the measurement and calculation methods corresponding to the first transmitter are largely the same as those corresponding to the second to Nth transmitters, and the corresponding descriptions can be applied to the measurement and calculation of the switching time overhead corresponding to the second to Nth transmitters.

[0069] S520: Measure the phase switching overhead time by sending message sequences from multiple transmitters to each receiver.

[0070] In this embodiment, the phase switching overhead time can refer to the time it takes for the receiver to switch its phase (CDR phase) from a certain phase to the corresponding transmitter phase in order to correctly receive a message. For example, CDR technology is a core component of SerDes technology. The receiver modifies the phase of its local clock according to certain decision criteria so that the generated clock can be synchronized with the received data. Because the receiver may connect to each transmitter after the switch is switched, it needs to adjust its local phase according to the corresponding transmitter phase to achieve synchronization.

[0071] In this embodiment, by sending message sequences from multiple transmitters to each of the multiple receivers, the corresponding phase switching overhead time can be calculated for each receiver. Repeating this process multiple times yields multiple phase switching overhead times for each of the multiple receivers.

[0072] As an example, when a switch has N transmitters and N receivers, in order to measure the phase switching overhead time by sending message sequences from multiple transmitters to each receiver, message sequences can be sent from the N transmitters to the first to the Nth receivers, respectively, to obtain N phase switching overhead times for each of the N receivers. This includes sending message sequences from the N transmitters to the first receiver to obtain one phase switching overhead time for the first receiver.

[0073] According to this example, starting from the first receiver among N receivers, a message sequence can be sent from N transmitters to the first receiver, then from the N transmitters to the second receiver, and so on, until a message sequence is sent from the N transmitters to the Nth receiver. That is, by using a "many-to-one" approach, this process is repeated N times to obtain the N phase switching overhead times corresponding to the N receivers.

[0074] For the specific measurement and calculation method of the phase switching overhead time corresponding to the first receiver in this example, please refer to the following section. Figures 10 to 12 The relevant descriptions should be understood. The measurement and calculation methods corresponding to the first receiving end are roughly the same as those corresponding to the second to Nth receiving ends, and the corresponding descriptions can be applied to the measurement and calculation of the phase switching time overhead corresponding to the second to Nth receiving ends.

[0075] S530. Calculate the switching efficiency of the switch based on the switching overhead time and phase switching overhead time.

[0076] In this embodiment, the optical switching system contains switching overhead and phase switching overhead, which together determine the overall overhead of the switch. Therefore, by calculating the switching overhead and phase switching overhead for each transmitter and receiver separately, the overall overhead time of the switch can be calculated, thereby determining the maximum message length that the switch can transmit, and finally, the switching efficiency of the switch.

[0077] As an example, in order to calculate the switching efficiency of a switch based on the switching overhead time and the phase switching overhead time, we can first determine the maximum value among the N switching overhead times and the N phase switching overhead times; then, we can calculate the switching efficiency of the switch based on the maximum value.

[0078] In this example, switch switching and phase switching occur simultaneously, and the larger of the two overhead times can be used as the basis for calculating switching efficiency. Therefore, the maximum value among multiple switch switching times and multiple phase switching times can be used as the maximum overhead time of the switch, and then the switching efficiency can be calculated based on the maximum overhead time.

[0079] As an example, to calculate the switching efficiency of a switch based on the maximum value, we can first determine the maximum overhead time of the switch based on the maximum value; then, based on the maximum overhead time, we can calculate the maximum message length that the switch is allowed to transmit; finally, based on the maximum message length, we can calculate the switching efficiency of the switch.

[0080] Figure 6 This is a schematic flowchart illustrating a method for measuring the switching overhead time of a switch corresponding to the first transmitting end in a switching efficiency measurement method according to an embodiment of this application. Figure 7 Showing according to Figure 6 A schematic diagram of the hardware architecture of the measurement method in this embodiment. Figure 8 Showing according to Figure 6 A schematic diagram of the message sequence corresponding to the maximum duration of the header blank portion in the measurement method of the embodiment. Figure 9 Showing according to Figure 6 A schematic diagram of the message sequence corresponding to the minimum duration of the header blank portion in the measurement method of the embodiment.

[0081] like Figure 6 As shown, the method for measuring the switching overhead time of the first transmitter includes steps S210 to S250, which are described in detail below.

[0082] S610. Construct a message sequence, which includes multiple message periods of equal duration. Each message period includes a message portion and a blank portion. The header of the message sequence has a header blank portion with adjustable duration.

[0083] In this embodiment, before constructing the message sequence, the main controller can first control each sending end and receiving end to perform high-precision clock time synchronization.

[0084] In this embodiment, a message sequence can refer to a sequence consisting of a series of messages and idle portions, used to measure the switching efficiency of a switch. A message portion refers to the valid message portion of the message sequence, which contains message information. An idle portion refers to a portion that occupies a certain amount of space or time in the message sequence but does not contain valid information; it is only used as a placeholder. A header idle portion refers to the empty portion at the beginning of the message sequence, and its length is adjustable; it can be longer or shorter than the idle portion in the message period. For example, when adjusting the length of the header idle portion, one bit of idle data can be added or removed from the header idle portion, and the number of bits of idle data in the message sequence header can be encapsulated within the message.

[0085] See Figure 8 The image above shows a message sequence. Its left header includes a blank header section, followed by multiple message cycles. Each message cycle includes both a message portion and a blank header section. For example... Figure 8 As shown, the message sequence consists of four message cycles, with each message segment including message 1, message 2, message 3, and message 4, which are separated by blank spaces.

[0086] S620, Send the message sequence from the first sender.

[0087] In this embodiment, the first transmitter is transmitter 1. See also... Figure 7 This illustrates a "one-against-many" scenario. In this embodiment, the optical switching system has four transmitters and four receivers. The four transmitters include transmitter 1, and the four receivers include receiver 1, receiver 2, receiver 3, and receiver 4. Both transmitters and receivers can be in the form of optical network interface cards (NICs). In the optical switching matrix, transmitter 1 is connected to four switches, namely switches 1, 2, 3, and 4 shown in the figure, each switch being connected to a corresponding receiver. Since each transmitter is connected to four switches, the four transmitters are connected to a total of 16 switches. In other words, in this embodiment, the optical switching matrix has a total of 16 switches, but only four are shown in the figure.

[0088] In this embodiment, when sender 1 sends a message sequence, the master controller can send control information to sender 1, notifying sender 1 to send the message at time T0. At this time, sender 1 can parse the information sent by the master controller and extract the data transmission time T0. Then, sender 1 waits for the NIC's internal RTC (Real-Time Clock) to reach time T0. At time T, the padded message sequence is transmitted through SerDes (see...) periodically, with the message period as the time interval. Figure 4 Send it out. Optical module (see...) Figure 4 The SerDes electrical signal is converted into an optical signal and output to the optical switching matrix. At this time, only one message is valid for each receiver, and the receiver phase is not switched, but frozen to the CDR phase of the transmitter.

[0089] S630. Using the duration of the message period as the time interval, sequentially open the N switching switches corresponding to the N receiving ends to allow the message sequence to reach the N receiving ends respectively.

[0090] In this embodiment, the duration of the message period (Tslice) is determined based on system parameters. For example, if the optical switching system needs to complete a switch within 50 nanoseconds, then the message period can be determined to be 50 nanoseconds.

[0091] like Figure 8 As shown in the figure, the timing of the sequential opening of the switching switches corresponding to receivers 1 through 4 is illustrated. Specifically, switch 1 corresponding to receiver 1 (see...) Figure 7Switch 1 opens at time 1 and closes at time 1; switch 2 corresponding to receiver 2 opens at time 2 and closes at time 2; switch 3 corresponding to receiver 3 opens at time 3 and closes at time 3; switch 4 corresponding to receiver 4 opens at time 4 and closes at time 4. It can be seen that the opening and closing times of each switch are not strictly sequential according to the message period, but have delays and errors. For example, the duration between opening 1 and closing 1 is not strictly Tslice, and the time corresponding to opening 2 is not exactly the same as the time corresponding to closing 1. Such delays and errors are caused by the different arrival times of the control commands from the main controller to each switch, and the inconsistent times from receiving the control commands to actually executing the opening or closing operation for each switch. This delay and error will affect the calculation of switch switching overhead. Some technologies in the art do not consider this impact, resulting in a significant difference between the calculated overhead time and the actual overhead time in operation. In the embodiments of this application, this impact is fully considered, thereby making the calculation of switch switching overhead more accurate and more consistent with actual operating conditions.

[0092] In this embodiment, the main controller alternately configures switches 1 to 4 at time Tsw (i.e., open 1) to open the optical switching matrix. The path delay of the message sequence from transmission to arrival at the switch of the optical switching matrix is ​​Tpath. Tsw satisfies Tsw < T0 + Tpath < Tsw + Tslice.

[0093] S640. Adjust the duration of the blank portion of the header to determine the maximum and minimum duration of the blank portion of the header, while ensuring that all N receivers receive the complete message portion.

[0094] In this embodiment, when adjusting the duration of the header blank portion, the duration can be increased or decreased by 1 bit each time, starting from time T0. At this time, receivers 1 to 4 parse and verify the received message to determine whether the message portion in the message sequence is complete. If the message portion is incomplete, it means that the message portion in the current message sequence cannot pass completely through the optical switching matrix. If the message portion is still complete, the duration of the header blank portion can be further increased or decreased by 1 bit, and the message sequence can be sent again.

[0095] See Figure 8 and Figure 9 ,exist Figure 8 In this situation, the switch 3 corresponding to receiver 3 is already immediately adjacent to the tail of message 3 when it is turned off (i.e., off 3). This means that if one bit of data is added to the blank portion of the header and the next message sequence is sent, message 3 will experience tail dropping. Therefore, Figure 8The scenario shown represents the previous message sequence transmission before the discovery of the message sequence dropping its tail. At this point, [the following can be done / can ... Figure 8 The message transmission time Tb shown is recorded, where Tb = T0 + ΔTn, and ΔTn starts from time T0 and continues until... Figure 8 The occurrence of the shown situation indicates the number of bits added to the header blank portion. Tb is the maximum duration of the header blank portion.

[0096] exist Figure 9 In this situation, the switch 4 corresponding to receiver 4 is already adjacent to the header of message 4 when it is turned on (i.e., the moment switch 4 is turned on). This means that if one bit of data is removed from the header blank portion and the next message sequence is sent, message 4 will lose its header. Therefore, Figure 9 The scenario shown depicts the previous message sequence transmission before the packet header was detected being lost. At this point, [the following can be done / can ... Figure 9 The message transmission time Ta shown is recorded, where Ta = T0 - ΔTm, and ΔTm starts from time T0 and continues until... Figure 9 The occurrence of the shown situation indicates the number of bits reduced in the header blank space. Ta is the minimum duration of the header blank space.

[0097] S650. Calculate the switching overhead time for the first transmitter based on the maximum and minimum duration values.

[0098] In this embodiment, the switching overhead time corresponding to transmitter 1 can be calculated based on Tb and Ta. For example... Figure 8 and Figure 9 As shown in the diagrams, the sequences in both figures can pass completely through the optical switching matrix. At this point, the last complete sequence received can be parsed at the receiving end to obtain the message sequence number, allowing for the calculation of switching overhead. For example, the message sequence number can be parsed from the first message with a lost header, and the transmission time of the previous sequence, Ta = T0 - ΔTm, can be recorded; the message sequence number can be parsed from the first message with a lost tail, and the transmission time of the previous sequence, Tb = T0 + ΔTn, can be recorded.

[0099] As an example, in order to calculate the switching overhead time of the first sender corresponding to a switch based on the maximum and minimum duration values, we can first calculate the difference between the maximum and minimum duration values; then, calculate the sum of the difference and the duration of the message portion; finally, calculate the difference between the duration of the message period and the sum, thereby determining the switching overhead time of the first sender.

[0100] Based on this example, the switching overhead time of the current link can be calculated using the following formula based on Ta and Tb:

[0101] Tgt1 = Tslice - (Tb - Ta + Tpkt)

[0102] Where Tgt1 is the switching overhead time corresponding to the current sender 1, and Tpkt is the duration of the message portion in the message period.

[0103] Then, the calculated Tgt1 for transmitter 1 can be sent to the main controller. The main controller finds the maximum value Tgt_max among the switching overhead times Tgt1 to Tgt4 of each transmitter 1 to 4, which is the overall switching overhead time of the optical switching system.

[0104] Figure 10 This is a schematic flowchart illustrating a method for measuring the phase switching overhead time corresponding to the first receiving end in a switching efficiency measurement method according to an embodiment of this application. Figure 11 Showing according to Figure 10 A schematic diagram of the hardware architecture of the measurement method in this embodiment. Figure 12 Showing according to Figure 10 A schematic diagram of the message sequences corresponding to the earliest and latest phase switching times in the measurement method of the embodiment.

[0105] like Figure 10 As shown, the method for measuring the phase switching overhead time corresponding to the first receiving end includes steps S310 to S360, and each step is described in detail below.

[0106] S1010. Construct a message sequence. The message sequence includes multiple message cycles of equal duration. Each message cycle includes a message portion and a blank portion.

[0107] In this embodiment, before constructing the message sequence, the main controller can first control each sending end and receiving end to perform high-precision clock time synchronization.

[0108] See Figure 12 The message sequence consists of multiple message cycles (Tslices), each message cycle including a message portion (Tpkt) and a blank portion (idle). Figure 12 In the message sequence shown, there are four message parts (i.e., the part containing valid message data), namely message 1, message 2, message 3 and message 4, which are sent by sender 1, sender 2, sender 3 and sender 4 respectively.

[0109] S1020: Send N message sequences alternately from N sending ends.

[0110] In this embodiment, senders 1 to 4 alternately send message sequences, and the message sequence sent by each sender is as described above.

[0111] See Figure 11 The diagram illustrates a "many-to-one" scenario. In this embodiment, the message sequence is jointly transmitted by sender 1, sender 2, sender 3, and sender 4. Each sender transmits the corresponding message sequence to the corresponding switches in the optical switching matrix, including switch 1, switch 2, switch 3, and switch 4. Then, the messages are transmitted to receiver 1 via these switches. Figure 11 Only one receiver is shown (i.e., receiver 1). Those skilled in the art will understand that... Figure 11 The optical switching system shown may include four transmitters and four receivers. The four switches shown correspond to receiver 1, and the remaining receivers (not shown) each correspond to four switches. Therefore, Figure 11 In the optical switching system shown, the optical switching matrix should have 16 switching switches.

[0112] In this embodiment, the main controller sends control information to sending terminals 1 to 4, notifying them to start sending messages alternately at time T0 with a period of Tsilce.

[0113] S1030. Using the duration of the message period as the time interval, sequentially open the N switching switches corresponding to the N sending ends to allow the N message sequences to reach the first receiving end respectively.

[0114] In this embodiment, since each of the four transmitters 1 to 4 sends a message, it is necessary to turn on the four switching switches corresponding to the four transmitters in sequence so that the optical switching matrix can pass through the messages 1 to 4 in sequence, so that the messages 1 to 4 arrive at the receiver 1 in turn.

[0115] In this embodiment, the main controller alternately configures switches 1 to 4 at time Tsw (i.e., time 1 is turned on) with a period of Tslice to turn on the optical switching matrix. Here, Tsw satisfies Tsw < T0 + Tpath < Tsw + Tslice. Tpath is the path delay from the transmitter to the switch in the optical switching matrix.

[0116] S1040. Using the duration of the message period as the time interval, the first receiver switches phases sequentially to receive N message sequences in sequence.

[0117] like Figure 12 As shown, receiver 1 switches phases starting from Tph to correctly receive message 1 sent by sender 1; then switches phases at switch 2 to correctly receive message 2 sent by sender 2; then switches phases at switch 3 to correctly receive message 3 sent by sender 3; and then switches phases at switch 4 to correctly receive message 4 sent by sender 4.

[0118] In this embodiment, the controller sends control information to the receiver 1, instructing the receiver 1 to write the preset phases (i.e., phase 1, phase 2, phase 3 and phase 4) into the CDR circuit at time Tph with a period of Tslice.

[0119] In this embodiment, the time it takes for the message sequence to travel from the switch in the optical switching matrix to the receiver is Tdelay (see [link]). Figure 11 ), Tdelay satisfies Tph<Tsw+Tdelay<Tph+Tslice.

[0120] S1050. Adjust the timing when the first receiver starts switching phases to determine the earliest and latest times while ensuring that the first receiver fully receives the message portions of N message sequences.

[0121] like Figure 12 As shown, receiver 1 begins switching phases at time Tph+ΔTn, thus delaying the phase switching time. At this point, during the third phase switch (i.e., switch 3), the phase switch is completed right after the header of message 3. This means that if receiver 1 had started switching phases any later, message 3 would not have been received completely by receiver 1, and its header would have been lost. Therefore, time Tph+ΔTn is the latest possible time when receiver 1 begins switching phases.

[0122] like Figure 12 As shown, receiver 1 can also start switching phases at time Tph-ΔTm, thus advancing the phase switching time. At this point, during the fourth phase switching (i.e., switching 4), the moment the phase switching begins is right at the tail of message 3. This means that if receiver 1 started switching phases even earlier, message 3 would not be fully received by receiver 1 and its tail would be dropped. Therefore, Tph-ΔTm is the earliest time when receiver 1 begins switching phases.

[0123] In this embodiment, receiver 1 parses and verifies the received message on the receiving side to determine if the message is complete. If the message is found to be incomplete after verification, it indicates that either the header or the tail has been lost, and the previous message sequence transmission status needs to be recorded. If the message is found to be complete after verification, receiver 1 adjusts the timing of the phase switching and retransmits the message sequence.

[0124] S1060. Calculate the phase switching overhead time corresponding to the first receiver based on the earliest and latest times.

[0125] like Figure 12As shown, the two sequences below can pass completely through the optical switching matrix. At the receiving end, the last complete sequence is parsed to obtain the message sequence number, allowing for phase switching overhead calculation. For example, the message sequence number is parsed from the first message with a lost tail, and the transmission time of the previous sequence, Ta = Tph - ΔTm, is recorded; the message sequence number is parsed from the first message with a lost header, and the transmission time of the previous sequence, Tb = Tph + ΔTn, is recorded.

[0126] As an example, to calculate the phase switching overhead time corresponding to the first receiver based on the earliest and latest times, we can first calculate the difference between the earliest and latest times; then, calculate the sum of the difference and the duration of the message portion; finally, calculate the difference between the duration of the message period and the sum, thereby determining the phase switching overhead time corresponding to the first receiver.

[0127] In this embodiment, the switching overhead time of the current link can be calculated based on Ta and Tb according to the following formula:

[0128] Tskew = Tslice - (Tb - Ta + Tpkt)

[0129] Where Tskew1 represents the phase switching overhead corresponding to the first receiver (Receiver 1), and Tpkt represents the duration of the message portion in the message sequence.

[0130] Then, the calculated Tskew1 for receiver 1 can be sent to the main controller. The main controller finds the maximum value Tskew_max among the phase switching overhead times Tskew1 to Tskew4 for each receiver 1 to 4, which is the overall phase switching overhead time of the optical switching system.

[0131] Based on the foregoing Figure 5 The present application also provides a switching efficiency measuring device, the structural schematic of which is shown in the embodiment of the method described above. Figure 13 As shown. This device is used to perform the aforementioned... Figure 5 Each step in the process.

[0132] According to this embodiment, the switching efficiency measurement device 1300 includes a first measurement module 1310, a second measurement module 1320, and a calculation module 1330. The first measurement module 1310 is used to measure the switch switching overhead time by sending a sequence of messages from each transmitter to the plurality of receivers. The second measurement module 1320 is used to measure the phase switching overhead time by sending a sequence of messages from the plurality of transmitters to each receiver. The calculation module 1330 is used to calculate the switching efficiency of the switch based on the switch switching overhead time and the phase switching overhead time.

[0133] It should be noted that, Figure 13 The switching efficiency measuring device 1300 provided in the illustrated embodiment, when performing the switching efficiency measurement method, is only illustrated by the division of the above-described functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the switching efficiency measuring device 1300 provided in the above embodiment and... Figure 5 The embodiments of the exchange efficiency measurement methods shown belong to the same concept, and their specific implementation processes are detailed in the method embodiments, which will not be repeated here.

[0134] Figure 14 This is a schematic diagram of the hardware structure of a computing device 1400 provided in an embodiment of this application.

[0135] See Figure 14 The computing device 1400 includes a processor 1410, a memory 1420, a communication interface 1430, and a bus 1440, which are interconnected via the bus 1440. The processor 1410, memory 1420, and communication interface 1430 may also be connected using other methods besides the bus 1440.

[0136] The memory 1420 can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), flash memory, optical storage, hard disk, etc.

[0137] The processor 1410 may be a general-purpose processor, which can be a processor that performs specific steps and / or operations by reading and executing contents stored in memory (e.g., memory 1420). For example, the general-purpose processor may be a central processing unit (CPU). The processor 1410 may include at least one circuit to perform... Figure 5 The embodiments shown provide all or part of the steps of the exchange efficiency measurement method.

[0138] The communication interface 1430 includes input / output (I / O) interfaces, physical interfaces, and logical interfaces for interconnecting devices within the computing device 1400, as well as interfaces for interconnecting the computing device 1400 with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc. The communication interface 1430 can connect to external input and output devices. For example, input devices can be microphones or microphone arrays for capturing voice input signals; they can be communication network connectors for receiving acquired input signals from the cloud or other devices; and they can also include, for example, keyboards, mice, etc. Output devices can output various information externally, including determined distance information, direction information, etc. Output devices can include, for example, displays, speakers, printers, and communication networks and their connected remote output devices, etc.

[0139] The bus 1440 can be any type of communication bus used to interconnect the processor 1410, memory 1420 and communication interface 1430, such as a system bus.

[0140] The aforementioned devices can be disposed on separate chips, or at least partially or entirely on the same chip. Whether to dispose of the devices independently on different chips or integrate them on one or more chips often depends on the needs of the product design. This application does not limit the specific implementation of the aforementioned devices.

[0141] Figure 14 The computing device 1400 shown is merely exemplary. In its implementation, the computing device 1400 may also include other components, which will not be listed here.

[0142] Embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the switching efficiency measurement methods according to various embodiments of this application described above.

[0143] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0144] The concepts, principles, and ideas of this application have been described in detail above with reference to specific embodiments (including examples and instances). Those skilled in the art should understand that the embodiments of this application are not limited to those given above. After reading this application, those skilled in the art can make any possible improvements, substitutions, and equivalents to the steps, methods, apparatus, and components in the above embodiments, and such improvements, substitutions, and equivalents should be considered to fall within the scope of this application. The scope of protection of this application is limited to the claims.

Claims

1. A method for measuring exchange efficiency, characterized in that, Applied to a switch that performs switching between multiple transmitting ends and multiple receiving ends, the method includes: The switchover overhead time is measured by sending a sequence of messages from each sender to the plurality of receivers. The phase switching overhead time is measured by sending message sequences from the plurality of transmitters to each receiver. The switching efficiency of the switch is calculated based on the switch switching overhead time and the phase switching overhead time.

2. The method according to claim 1, characterized in that, The switch has N transmitting ends and N receiving ends. Measuring the switch switching overhead time by sending a sequence of messages from each transmitting end to the plurality of receiving ends includes: Send message sequences from the first to the Nth of the N transmitters to the N receivers respectively to obtain the N switch switching overhead times corresponding to the N transmitters; The step of measuring the phase switching overhead time by sending message sequences from the plurality of transmitters to each receiver includes: The message sequence is sent from the N transmitting ends to the first to the Nth receiving ends respectively, and the N phase switching overhead times corresponding to the N receiving ends are obtained respectively; The step of calculating the switching efficiency of the switch based on the switch switching overhead time and the phase switching overhead time includes: Determine the maximum value among the N switch switching overhead times and the N phase switching overhead times; The switching efficiency of the switch is calculated based on the maximum value.

3. The method according to claim 2, characterized in that, The step of sending message sequences from the 1st to the Nth of the N transmitters to the N receivers to obtain the N switchover overhead times corresponding to the N transmitters includes: Send a message sequence from the first of the N sending ends to the N receiving ends to obtain the switching overhead time of the first sending end; The step of sending a message sequence from the first of the N sending ends to the N receiving ends to obtain the switching overhead time corresponding to the first sending end includes: Construct a message sequence, the message sequence comprising multiple message periods of equal duration, each message period comprising a message portion and a blank portion, the header of the message sequence having a header blank portion of adjustable duration; The message sequence is sent from the first sending end; Using the duration of the message period as a time interval, the N switching switches corresponding to the N receiving ends are turned on sequentially to allow the message sequence to reach the N receiving ends respectively; The duration of the blank portion in the header is adjusted to determine the maximum and minimum duration of the blank portion in the header, while ensuring that all N receiving ends receive the complete portion of the message. Based on the maximum and minimum duration values, calculate the switching overhead time for the first transmitter.

4. The method according to claim 3, characterized in that, The step of calculating the switching overhead time corresponding to the first transmitting end based on the maximum and minimum duration values ​​includes: Calculate the difference between the maximum duration and the minimum duration; Calculate the sum of the difference and the duration of the message portion; The difference between the duration of the message period and the sum is calculated to determine the switching overhead time corresponding to the first sender.

5. The method according to claim 2, characterized in that, The step of sending message sequences from the N transmitters to the 1st to the Nth receivers from the N transmitters to obtain the N phase switching overhead times corresponding to the N receivers includes: Send a message sequence from the N transmitters to the first receiver among the N receivers to obtain the phase switching overhead time corresponding to the first receiver; The step of sending a message sequence from the N transmitting ends to the first receiving end among the N receiving ends to obtain a phase switching overhead time corresponding to the first receiving end includes: Construct a message sequence, the message sequence comprising multiple message periods of equal duration, each message period comprising a message portion and a blank portion; The N message sequences are alternately transmitted from the N transmitting ends; Using the duration of the message period as a time interval, the N switching switches corresponding to the N sending ends are opened sequentially to allow the N message sequences to reach the first receiving end respectively; Using the duration of the message period as a time interval, the first receiving end switches phases sequentially to receive the N message sequences in sequence; The timing at which the first receiving end begins switching phases is adjusted to determine the earliest and latest times of the timing, while ensuring that the first receiving end completely receives the message portions of the N message sequences. Based on the earliest time and the latest time, calculate the phase switching overhead time corresponding to the first receiving end.

6. The method according to claim 5, characterized in that, The step of calculating the phase switching overhead time corresponding to the first receiving end based on the earliest time and the latest time includes: Calculate the difference between the earliest time and the latest time; Calculate the sum of the difference and the duration of the message portion; The difference between the duration of the message period and the sum is calculated to determine the phase switching overhead time corresponding to the first receiving end.

7. The method according to claim 2, characterized in that, The step of calculating the switching efficiency of the switch based on the maximum value includes: Based on the maximum value, determine the maximum overhead time of the switch; Based on the maximum overhead time, calculate the maximum packet length that the switch is allowed to transmit; The switching efficiency of the switch is calculated based on the maximum message length.

8. A device for measuring exchange efficiency, characterized in that, Applied to a switch that enables switching between multiple transmitting ends and multiple receiving ends, the device includes: The first measurement module is used to measure the switch switching overhead time by sending a message sequence from each transmitter to the plurality of receivers. The second measurement module is used to measure the phase switching overhead time by sending a message sequence from the plurality of transmitters to each receiver. The calculation module is used to calculate the switching efficiency of the switch based on the switch switching overhead time and the phase switching overhead time.

9. A switch, characterized in that, The switch includes multiple transmitters, multiple receivers, a main controller, and a memory. The switch performs switching between the multiple transmitters and the multiple receivers. The main controller is used to execute a computer program stored in the memory to implement the switching efficiency measurement method according to any one of claims 1 to 7.

10. A computing device, characterized in that, The computing device includes a processor and a memory, the processor being configured to execute a computer program stored in the memory to implement the switching efficiency measurement method according to any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program for performing the switching efficiency measurement method according to any one of claims 1 to 7.

Images