Flow block compression method in network tester
Through the flow block compression method, the flow template and flow block information are recorded, and similar flow blocks are classified and marked according to the flow jump type and merged to generate normal and compressed flow block groups, which solves the problem of resource limitation of network tester and realizes efficient and low-cost network testing.
Patent Information
- Application Number
- CN202510665014.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-22
- Publication Date
- 2025-09-23
AI Technical Summary
Due to resource constraints, existing network testers have a limited number of flow blocks, resulting in test failure when the test configuration exceeds the upper limit. In addition, existing solutions are costly and time-sensitive.
By recording flow templates and flow block information, counting the total number of flow blocks and performing compression processing, classifying and marking flow blocks according to flow jump type, merging or optimizing similar flow blocks, generating normal and compressed flow block groups, sending normal flow block groups to the FPGA, and extracting the original information of the compressed flow block group for jump comparison, the final configuration flow block information is sent to the FPGA.
This improves the applicability of the tester without modifying the design, saves costs, ensures the efficiency and accuracy of traffic simulation, and adapts to larger-scale network testing.
Smart Images
Figure CN120692206A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, in particular to a stream block compression method in a network tester. Background Art
[0002] Network testers are now widely used in the R&D, operation, and maintenance of network chips and network equipment. Their core traffic module defines flow-related configurations and sends them to the FPGA to generate traffic. Traffic functions are typically represented by several flow templates, each of which aggregates multiple flow blocks. The FPGA sends traffic in flow blocks, and each flow block generates multiple flows and data frames by configuring flow transitions. However, due to resource constraints, the number of flow blocks that can be sent is limited. If the specified limit is exceeded, the sent traffic will be erroneous. However, actual network test configurations are often complex, making it easy to exceed the configuration limit, resulting in test failure.
[0003] Existing solutions mostly involve expanding FPGA capacity, increasing specification limits, or streamlining test configurations to meet specification limits. However, these solutions suffer from technical issues such as high expansion costs and difficulty meeting test requirements in terms of timeliness. Summary of the Invention
[0004] The purpose of the present invention is to provide a stream block compression method in a network tester to solve the technical problems in the prior art of high expansion cost and difficulty in meeting test requirements in terms of timeliness, and to achieve the technical effect of improving the applicability of the tester and saving costs without modifying the design.
[0005] A stream block compression method in a network tester provided by the present invention adopts the following technical solution: Records the information of each flow template and the aggregated flow block information under the sending port.
[0006] Count and determine the total number of stream blocks of the current sending port and perform stream block compression processing.
[0007] The compressed flow blocks are classified and marked according to the flow hopping type, and the flow blocks of each flow template are divided into a common flow block group and a compressed flow block group.
[0008] The common stream block group configuration is sent to the FPGA, original stream block information of the compressed stream block group is obtained, and a compressed transition group is extracted according to the original stream block information.
[0009] A jump comparison is performed based on the compressed stream block group and the compressed jump group, and the original jump of each stream block is converted into a final jump.
[0010] Based on the final jump, the block ID of the compressed stream block group is used as an identifier, and the stream block information is configured to the FPGA to complete the stream sending operation.
[0011] In a feasible implementation, counting and determining the total number of stream blocks of the current sending port and performing stream block compression processing include: Get the upper limit of the number of stream blocks supported by the FPGA to which the sending port belongs.
[0012] Determine whether the total number of current stream blocks exceeds the upper limit of the supported number of stream blocks. If so, perform stream block compression processing.
[0013] In a feasible implementation, the flow blocks of each flow template are divided into a common flow block group and a compressed flow block group, including: Uncompressed common stream blocks are individually grouped into common stream block groups, and each common stream block group is assigned a unique block ID.
[0014] The compressed stream blocks with the same stream jump characteristics are merged into the compressed stream block group, and all the compressed stream blocks in the same compressed stream block group share the same block ID.
[0015] In a feasible implementation, obtaining original stream block information of the compressed stream block group and extracting the compressed jump group according to the original stream block information includes: Each compressed stream block group is traversed to obtain the original jump and protocol header data of each stream block to form the original stream block information.
[0016] Taking the first stream block of the compressed stream block group as a reference, the protocol header data of each stream block in the compressed stream block group is traversed and compared with the protocol header data of other stream blocks in the same compressed stream block group one by one.
[0017] If they are inconsistent, the data range of the inconsistent position is recorded as a compression jump and stored in the compression jump group.
[0018] In a feasible implementation, the protocol header data includes at least one of Ethernet, IPv4, VLAN, PPPoE, IPv6, MPLS, and VXLAN.
[0019] In a feasible implementation, performing jump comparison between the compressed stream block group and the compressed jump group, and converting the original jump of each stream block into a final jump, includes: Each stream block in the compressed stream block group and each compressed transition in the compressed transition group are traversed, and a starting position of the original transition is compared with a starting position of the compressed transition.
[0020] If the starting position of the original transition is the same as the starting position of the compressed transition, the value of the original transition is directly used as the final transition.
[0021] If the starting position of the original transition is different but the data range includes the data range of the compressed transition, the original transition data within the compressed transition range is intercepted as the final transition.
[0022] If the starting position of the original transition is different and the data range does not include the compressed transition range, the data within the compressed transition range is directly used to generate the final transition.
[0023] In a feasible implementation, the compressed stream blocks are classified and marked according to the stream hopping type, including: dividing into at least one of a source address hopping type group, a destination address hopping type group and a no hopping type group.
[0024] In summary, the present invention discloses a flow block compression method in a network tester, comprising: recording the self-information of each flow template under the sending port and the aggregated flow block information thereof, and counting the total number of flow blocks of the current sending port, and performing flow block compression processing. The compressed flow blocks are classified and marked according to the flow jump type, and the flow blocks of the flow template are divided into ordinary flow block groups and compressed flow block groups. The ordinary flow block group configuration is sent to the FPGA, and the original flow block information of the compressed flow block group is obtained at the same time, and the compressed jump group is extracted based on the information. The jump conditions of the compressed flow block group and the compressed jump group are compared, and the original jump of each flow block is converted into the final jump. Finally, the block ID of the compressed flow block group is used as an identifier, and the flow block information is configured to the FPGA to complete the flow sending operation. The flow block compression method in the network tester disclosed by the present invention solves the technical problems of high expansion cost and difficulty in meeting test requirements in timeliness, and achieves the technical effect of improving the applicability of the tester and saving costs without modifying the design. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 It is a flow chart of a stream block compression method in the network tester of the present invention; Figure 2 The present invention is a schematic diagram of a flow chart of extracting compressed jump groups in a stream block compression method of a network tester. DETAILED DESCRIPTION
[0026] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention, rather than all, are shown in the drawings.
[0027] Example, Figure 1The flowchart of a stream block compression method in a network tester of the present invention is shown below. The stream block compression method in the network tester includes: Records the information of each flow template and the aggregated flow block information under the sending port.
[0028] Specifically, a flow template is a configuration unit used to define network traffic characteristics, including key flow parameters such as source address, destination address, protocol type, port information, etc.; a flow block is the smallest configuration unit of traffic during network testing. A flow template is usually composed of multiple flow blocks, and each flow block simulates complex network traffic by jumping in a specific way (such as IP address change, port number change, etc.).
[0029] Specifically, the sending port is the physical or logical interface used by the network tester to send traffic. It is responsible for generating and sending test traffic according to the configuration of the flow template and flow block. It records the information of all flow templates under the sending port, including their names, traffic patterns, configuration parameters, etc., and simultaneously records the information of all aggregated flow blocks under the flow template, such as the total number of flow blocks, specific jump methods, parameters of each flow block, etc., to provide a basis for subsequent analysis and processing.
[0030] Count and determine the total number of stream blocks of the current sending port and perform stream block compression processing.
[0031] Specifically, the total number of flow blocks refers to the sum of the number of flow blocks contained in all flow templates under the current sending port. FPGA has fixed hardware resources and storage capabilities. The total number of flow blocks determines whether flow block compression is required.
[0032] In some embodiments, counting and determining the total number of stream blocks of the current sending port and performing stream block compression processing include: Obtain the upper limit of the number of stream blocks supported by the FPGA to which the sending port belongs; determine whether the current total number of stream blocks exceeds the upper limit of the number of stream blocks supported, and if so, perform stream block compression processing.
[0033] Specifically, the upper limit of the number of stream blocks supported by an FPGA refers to the maximum number of stream blocks that the FPGA can process under the current configuration. This is because FPGAs (field programmable gate arrays) usually have fixed hardware resources and storage capabilities. Exceeding the upper limit of the number of stream blocks supported may result in data loss or processing failure.
[0034] Specifically, the upper limit of the number of stream blocks supported by the FPGA to which the sending port belongs is obtained and compared with the current total number of stream blocks. If the total number of stream blocks does not exceed the upper limit, traffic sending can be performed normally without the need for stream block compression.
[0035] Specifically, when the total number of flow blocks exceeds the FPGA's processing capacity, similar flow blocks are merged, deleted, or optimized to reduce the number of flow blocks to meet the FPGA's processing capacity constraints. Compression methods can include merging similar flow blocks (for example, if multiple flow blocks only vary slightly in the source IP address or port number, they can be merged into a single flow block, using a mask to represent the hop range), reducing unnecessary flow blocks (removing low-priority flow blocks), and optimizing hop rules (reducing meaningless parameter hops, such as optimizing step size or reducing unnecessary random hop ranges).
[0036] Dynamically adjusting the flow block configuration through this step helps ensure proper allocation of FPGA resources and prevents processing anomalies caused by an excessive number of flow blocks. Furthermore, flow block compression improves the efficiency of traffic simulation and reduces unnecessary computational overhead, enabling the network tester to simulate larger, more realistic network traffic within limited hardware resources.
[0037] The compressed flow blocks are classified and marked according to the flow hopping type, and the flow blocks of each flow template are divided into a common flow block group and a compressed flow block group.
[0038] Specifically, the flow hopping type refers to how key parameters of a flow block change during packet generation. For example, traffic changes can manifest themselves in different hopping patterns for fields such as the source IP address, destination IP address, port number, protocol type, and data payload, including sequential hopping, random hopping, and range hopping. Different hopping types determine the complexity of the traffic pattern.
[0039] Specifically, after the stream block compression is completed, the stream blocks are classified according to the stream jump type and corresponding tags are added to them so as to distinguish different types of stream blocks in subsequent data analysis and optimization processing.
[0040] Specifically, a normal flow block group refers to flow blocks that have not undergone compression optimization. These flow blocks retain their original definitions and are unaffected by flow block compression. A compressed flow block group refers to flow blocks that have undergone flow block compression optimization. These flow blocks are reduced in number through merging and optimizing transition rules to accommodate the processing power of the FPGA. Generally, compressed flow block groups are used to simulate large-scale traffic patterns and represent multiple similar flow blocks more efficiently.
[0041] In some implementations, the compressed stream blocks are classified and marked according to the stream hopping type, including: being divided into at least one of a source address hopping type group, a destination address hopping type group, and a no hopping type group.
[0042] Specifically, the source address hopping type group refers to the hopping of the flow block, which is mainly reflected in the change of the source IP address (Source IP). It is suitable for simulating scenarios where multiple clients access the same server, such as DDoS testing and load balancing testing. For example, when simulating multiple different hosts sending traffic to the same target server, different flow blocks can be created by adjusting the source address.
[0043] Specifically, the destination address hopping type group refers to the hopping of the flow block mainly occurring in the destination IP address (DestinationIP). It is suitable for simulating scenarios where a single device sends data to multiple devices, such as CDN (Content Delivery Network) simulation and multicast testing. For example, when simulating a server sending data streams to multiple clients, the destination address will change with different flow blocks.
[0044] Specifically, the "no-jump" type group refers to a flow block with no changes in source and destination addresses. This may involve only adjustments to the data payload, protocol type, or port number. This group is suitable for testing fixed end-to-end connections, such as validating specific protocols or analyzing traffic stability. For example, this group is used for test traffic on fixed end-to-end connections.
[0045] This step aims to clarify the hopping patterns of different flow blocks and optimize traffic generation strategies. Classification and labeling allow for more precise selection of required flow blocks during subsequent traffic scheduling, improving test flexibility and controllability.
[0046] In some embodiments, dividing the flow blocks of each flow template into a normal flow block group and a compressed flow block group includes: Uncompressed ordinary stream blocks are separately organized into ordinary stream block groups, and a unique block ID is assigned to each ordinary stream block group; compressed stream blocks with the same stream jump characteristics are merged into the compressed stream block group, and all compressed stream blocks in the same compressed stream block group share the same block ID.
[0047] Specifically, the block ID is a number used to uniquely identify a flow block group, ensuring accurate identification during subsequent traffic scheduling. For ordinary flow block groups, each independent flow block is assigned a unique block ID. For compressed flow block groups, all merged flow blocks share the same block ID to reduce storage redundancy.
[0048] Specifically, the compressed stream blocks are screened to identify stream blocks with the same stream jump characteristics, such as: stream blocks with only the source address changed but other parameters are the same, stream blocks with only the destination address changed but other parameters are the same, and stream blocks with only the port number jumped but other parameters are the same; then, these mergeable stream blocks are classified into the same compressed stream block group according to the classification marks of the jump characteristics to reduce storage redundancy. At the same time, a shared block ID is assigned to each compressed stream block group to simplify data management.
[0049] Through the above steps, the division of normal flow block groups and compressed flow block groups can dynamically adjust the traffic replay strategy according to different testing requirements, such as simulating different traffic patterns (random, large-scale concurrency, etc.); among them, the compressed flow block group merges duplicate data and only stores key information to avoid unnecessary redundant storage; by sharing block IDs, batch traffic can be quickly called.
[0050] The common stream block group configuration is sent to the FPGA, original stream block information of the compressed stream block group is obtained, and a compressed transition group is extracted according to the original stream block information.
[0051] Specifically, the configuration information of an uncompressed group of common flow blocks is sent to the FPGA via a control interface (such as PCIe, I2C, or AXI bus). After receiving the configuration, the FPGA stores the flow blocks according to their contents and uses these flow blocks for traffic playback during traffic simulation or testing. The configuration information typically includes traffic characteristics such as source / destination IP addresses, port numbers, protocol type, and data payload.
[0052] Specifically, the raw stream block information of a compressed stream block group refers to the detailed data of the stream blocks before they are merged. This information is used to further analyze and extract stream transition characteristics based on the original form of the compressed stream blocks. For example, a compressed stream block group may consist of multiple stream blocks with different source addresses but otherwise identical characteristics. This raw stream block information contains the specific attributes of each individual stream block.
[0053] In some embodiments, as Figure 2 As shown, obtaining the original stream block information of the compressed stream block group and extracting the compressed jump group according to the original stream block information includes: Traverse each compressed stream block group, obtain the original jump and protocol header data of each stream block, and form the original stream block information; take the first stream block of the compressed stream block group as the benchmark, traverse the protocol header data of each stream block in the compressed stream block group, and compare them with the protocol header data of other stream blocks in the same compressed stream block group one by one; if there is inconsistency, record the data range of the inconsistent position as a compressed jump, and store it in the compressed jump group.
[0054] Specifically, each compressed stream block group is first traversed to obtain the original jump and protocol header data of each stream block, thereby forming complete original stream block information as the basic data for subsequent jump extraction, where the protocol header data includes but is not limited to: IP layer header fields (such as version, source / destination address, TTL, etc.); transport layer header fields (such as TCP / UDP port number, sequence number, flag bit, etc.); application layer protocol fields (such as HTTP method, URI, Header field, etc.).
[0055] Next, taking the first stream block in the compressed stream block group as the benchmark, the protocol header data of other stream blocks in the same group are traversed in turn, and the protocol header data of each stream block is compared with the protocol header data of the benchmark stream block (the first stream block) at the field level.
[0056] Specifically, if a field value is found to be inconsistent with the baseline stream block, the data offset and data range of the field are marked as a jump region. All jump regions and their corresponding change values are recorded as compressed jumps, and the compressed jumps are stored in a compressed jump group. This process ensures that all differences in the compressed stream block group are accurately captured, providing a basis for subsequent data processing and configuration optimization.
[0057] Exemplarily, each compressed jump entry may include: a jump field identifier (such as a field name or offset address); a jump value (the actual value of the variant field in the current stream block); a jump length (the number of bytes occupied by the field); and a corresponding stream block number (indicating which stream block in the compression group the jump belongs to).
[0058] Through the above steps, the protocol header data differences in the compressed stream block group can be accurately identified and extracted, thereby forming a detailed compression jump group, which can maintain the accuracy and integrity of the stream block configuration while meeting the FPGA resource constraints.
[0059] In some implementations, the protocol header data includes at least one of Ethernet, IPv4, VLAN, PPPoE, IPv6, MPLS, and VXLAN.
[0060] For example, in a data center network environment, the compressed flow block may include the encapsulation structure of VXLAN over IPv6 over Ethernet; in a broadband access network, the compressed flow block may include the protocol stack of PPPoE over VLAN over IPv4.
[0061] By supporting the parsing and comparison of multiple protocol header data, the adaptability and versatility of the compressed jump extraction mechanism in complex network environments are significantly improved, ensuring that jump fields can be accurately extracted in multiple encapsulation formats, thereby achieving efficient data restoration, behavior identification and security auditing.
[0062] A jump comparison is performed based on the compressed stream block group and the compressed jump group, and the original jump of each stream block is converted into a final jump.
[0063] In some embodiments, performing a jump comparison between the compressed stream block group and the compressed jump group, and converting an original jump of each stream block into a final jump, includes: Traverse each stream block in the compressed stream block group and each compressed jump in the compressed jump group, and compare the starting position of the original jump with the starting position of the compressed jump; if the starting position of the original jump is the same as the starting position of the compressed jump, the value of the original jump is directly used as the final jump; if the starting position of the original jump is different but the data range includes the data range of the compressed jump, the original jump data within the compressed jump range is intercepted as the final jump; if the starting position of the original jump is different and the data range does not include the compressed jump range, the data within the compressed jump range is directly used to generate the final jump.
[0064] Specifically, the final transition refers to the transition information finally determined for each stream block after the compressed transition group is compared with the original transition, and is used to accurately define the changes in each protocol field in the data packet.
[0065] Specifically, jump comparison processing is performed based on the compressed stream block group and the compressed jump group to convert the original jump information of each stream block into final jump information, thereby achieving accurate restoration of the compressed data: first, traverse each stream block in the compressed stream block group, and at the same time compare each compressed jump record in the compressed jump group, and perform the following comparison operations: obtain the starting position and data length of the current original jump, obtain the starting position and data length of the current compressed jump, compare whether the starting offset positions of the two are the same, and whether there is an inclusion or intersection relationship between the data ranges.
[0066] Furthermore, based on the comparison results, one of the following three situations is used to determine the final jump information: Case 1: Same Starting Position: If the starting position of the original transition matches the starting position of the compressed transition, the field has not shifted before and after compression, and the system can directly use the original transition value as the final transition value. For example, if the original transition starts at offset = 40 and has a length of 4 bytes, and the compressed transition also starts at offset = 40 and has a length of 4 bytes, the original transition value will be used directly as the final transition value.
[0067] Case 2: The starting positions are different, but the original jump range includes the compressed jump range: If the starting position of the original jump is different from the starting position of the compressed jump, but the data range of the original jump completely includes the data range of the compressed jump, it means that the jump value was clipped or offset during the compression process. In this case, the data corresponding to the compressed jump range is intercepted from the original jump value as the final jump. For example, if the original jump range is offset = 36-44 and the compressed jump range is offset = 40-44, the portion of offset = 40-44 from the original jump value is intercepted as the final jump value.
[0068] Case 3: Different starting positions and data range not included: If the starting positions of the original transition and the compressed transition are different, and the data range of the original transition does not include the compressed transition data range, this indicates that the field has undergone structural changes or has been added / replaced during the compression process. In this case, the data within the compressed transition range is directly used to generate the final transition. For example, if the original transition range is offset = 20-24 and the compressed transition range is offset = 36-40, the value in the compressed transition is directly used as the final transition value (offset = 36-40).
[0069] Through the above method, the final jump information of each stream block can be accurately generated based on the compressed stream block group and the compressed jump group, wherein the accurate extraction of the jump value is ensured through the joint comparison of the position and the range.
[0070] Based on the final jump, the block ID of the compressed stream block group is used as an identifier, and the stream block information is configured to the FPGA to complete the stream sending operation.
[0071] Furthermore, based on the final transition information extracted in the previous step, the block ID of the compressed stream block group is used as an identifier. The final transition data of each stream block is associated with the block ID and then uniformly configured. Specifically, the final transition data after transition comparison and conversion is classified according to the block ID, and a corresponding configuration file or data stream is generated. This is then sent to the FPGA via a dedicated interface, achieving precise control of complex network traffic.
[0072] In summary, the stream block compression method in the network tester provided by the present invention has the following technical effects: By recording the self-information of each flow template under the sending port and its aggregated flow block information, and counting the total number of flow blocks of the current sending port, flow block compression processing is performed; the compressed flow blocks are classified and marked according to the flow jump type, and the flow blocks of the flow template are divided into ordinary flow block groups and compressed flow block groups; the ordinary flow block group configuration is sent to the FPGA, and the original flow block information of the compressed flow block group is obtained at the same time, and the compressed jump group is extracted based on the information; the jump conditions of the compressed flow block group and the compressed jump group are compared, and the original jump of each flow block is converted into the final jump; finally, the block ID of the compressed flow block group is used as an identifier, the flow block information is configured to the FPGA, and the flow sending operation is completed, thereby achieving the technical effect of improving the applicability of the tester and saving costs without modifying the design.
[0073] It should be understood that the embodiments of this specific embodiment are preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Identical components are denoted by the same reference numerals. Therefore, any equivalent changes made based on the structure, shape, and principles of the present invention are intended to be within the scope of protection of the present invention.
Claims
1. A stream block compression method in a network tester, characterized in that: include: Record the information of each flow template and the aggregated flow block information under the sending port; Count and determine the total number of stream blocks of the current sending port and perform stream block compression processing; Classify and mark the compressed flow blocks according to the flow hopping type, and divide the flow blocks of each flow template into a common flow block group and a compressed flow block group; Sending the common stream block group configuration to the FPGA, obtaining the original stream block information of the compressed stream block group, and extracting the compressed transition group according to the original stream block information; Perform jump comparison between the compressed stream block group and the compressed jump group, and convert the original jump of each stream block into a final jump; Based on the final jump, the block ID of the compressed stream block group is used as an identifier, and the stream block information is configured to the FPGA to complete the stream sending operation.
2. A stream block compression method in a network tester according to claim 1, characterized in that: Count and determine the total number of stream blocks on the current sending port and perform stream block compression processing, including: Get the upper limit of the number of stream blocks supported by the FPGA to which the sending port belongs; Determine whether the total number of current stream blocks exceeds the upper limit of the supported number of stream blocks. If so, perform stream block compression processing.
3. A stream block compression method in a network tester according to claim 2, characterized in that: The flow blocks of each flow template are divided into common flow block groups and compressed flow block groups, including: Group uncompressed common stream blocks into common stream block groups, and assign a unique block ID to each common stream block group; The compressed stream blocks with the same stream jump characteristics are merged into the compressed stream block group, and all the compressed stream blocks in the same compressed stream block group share the same block ID.
4. A stream block compression method in a network tester according to claim 3, characterized in that: Obtaining original stream block information of the compressed stream block group, and extracting a compressed jump group according to the original stream block information, comprising: Traversing each compressed stream block group, obtaining original transition and protocol header data of each stream block, and forming the original stream block information; Taking the first stream block of the compressed stream block group as a reference, traversing the protocol header data of each stream block in the compressed stream block group, and successively comparing the protocol header data of other stream blocks in the same compressed stream block group; If they are inconsistent, the data range of the inconsistent position is recorded as a compression jump and stored in the compression jump group.
5. A stream block compression method in a network tester according to claim 4, characterized in that: The protocol header data includes at least one of Ethernet, IPv4, VLAN, PPPoE, IPv6, MPLS, and VXLAN.
6. A stream block compression method in a network tester according to claim 4, characterized in that: Performing jump comparison between the compressed stream block group and the compressed jump group, and converting the original jump of each stream block into a final jump, comprising: Traversing each stream block in the compressed stream block group and each compressed transition in the compressed transition group, and comparing the starting position of the original transition with the starting position of the compressed transition; If the starting position of the original jump is the same as the starting position of the compressed jump, the value of the original jump is directly used as the final jump; If the starting position of the original jump is different but the data range includes the data range of the compressed jump, the original jump data within the compressed jump range is intercepted as the final jump; If the starting position of the original transition is different and the data range does not include the compressed transition range, the data within the compressed transition range is directly used to generate the final transition.
7. A stream block compression method in a network tester according to claim 1, characterized in that: Classifying and marking the compressed stream blocks according to the stream jump type, including: dividing into at least one of a source address jump type group, a destination address jump type group and a no-jump type group.