Method and system for analyzing business transaction penetration time delay based on transaction decoding, and medium

Abstract

The invention discloses a method and a system for analyzing business transaction penetration delay based on transaction decoding, and a medium. The method comprises the following steps: an intermediate service receives a transaction request sent by an access end; according to the transaction request, based on mirror traffic transaction decoding, obtaining request starting time, response starting time, transaction processing time and transaction distributed link id of each transaction of each intermediate service; associating business transactions of each intermediate service and the target service based on the transaction distributed link id, and calculating an uplink penetration time delay, a downlink penetration time delay and a processing time delay of the business transactions in each intermediate service and / or the target service according to the access topological graph; wherein the access topological graph refers to the access relationship of the business transaction. According to the invention, based on mirror image traffic transaction decoding, the problems of code development and invasion to a service system are solved, and the problem of system resource occupation is avoided.

Description

Technical Field

[0001] This invention relates to the field of TCP / IP network transmission, and specifically to a method, system, and medium for analyzing the latency of business transactions based on transaction decoding. Background Technology

[0002] Currently, business transactions typically require multiple service accesses to complete. The current method for pinpointing latency in business transactions requires developers of the business system to create tracking code logic in the business code, add corresponding tracking code or install an agent program, and then collect the tracking data and perform latency analysis.

[0003] However, it is intrusive to business systems, and developing the event tracking code requires additional work. It also requires allocating additional system resources for the collection and transmission of event tracking data. Especially for high-performance businesses, the additional system resource consumption can lead to a decrease in the throughput of the business system.

[0004] In view of the above, this application is hereby submitted. Summary of the Invention

[0005] The technical problem this invention aims to solve is the latency penetration issue in existing location-based business transactions. These methods require adding corresponding tracking code or installing an agent program to the business code. However, this is intrusive to the business system, involves additional workload in developing the tracking code, and requires allocating additional system resources for data collection and transmission. Especially for high-performance services, this additional system resource consumption can lead to a decrease in business system throughput. The purpose of this invention is to provide a method, system, and medium for analyzing business transaction latency penetration based on transaction decoding. Based on mirrored traffic transaction decoding, it solves the problems of code development and intrusion into the business system, and avoids the problem of excessive system resource consumption.

[0006] This invention is achieved through the following technical solution:

[0007] In a first aspect, the present invention provides a method for analyzing transaction penetration latency based on transaction decoding, the method comprising:

[0008] The intermediate service receives transaction requests sent by the access end;

[0009] Based on the transaction request, and using mirrored traffic transaction decoding, the request start time, response start time, transaction processing time, and transaction distributed link ID of each transaction for each intermediate service are obtained.

[0010] Business transactions are associated with each intermediate service and target service based on the transaction distributed link ID, and the uplink penetration latency, downlink penetration latency and processing latency of business transactions in each intermediate service and / or target service are calculated according to the access topology.

[0011] The access topology diagram refers to the access relationships of business transactions.

[0012] Furthermore, based on the transaction request and mirrored traffic transaction decoding, the request start time, response start time, transaction processing time, and transaction distributed link ID of each transaction for each intermediate service are obtained, including:

[0013] Based on the transaction request, the mirror traffic of each intermediate service is collected through several traffic collection points deployed in the network; the traffic collection points are deployed at the traffic exit or traffic entry point of the network device corresponding to each intermediate service.

[0014] Based on the mirrored traffic, identify the application traffic of each intermediate service according to service rules;

[0015] Based on application traffic and transaction configuration, the system decodes transactions at each traffic collection point to obtain the request start time, response start time, transaction processing time, and transaction distributed link ID for each transaction in each intermediate service.

[0016] Furthermore, the intermediate services include a first intermediate service and a second intermediate service; the specific steps for decoding transactions based on mirrored traffic are as follows:

[0017] The first intermediary service receives transaction requests sent by the access end;

[0018] Based on the transaction request, the mirror traffic of the first intermediate service is collected through the first traffic collection point to obtain the first mirror traffic;

[0019] Based on the first mirror traffic, the first mirror traffic is directed to the traffic analysis device, which analyzes and identifies the first mirror traffic to obtain the first request start time, first response start time, first transaction processing time, and first transaction distributed link ID of the first intermediate service transaction.

[0020] The transaction request carries the first transaction distributed link ID to form a first forwarding request, and the first forwarding request is sent to the second intermediate service;

[0021] The second intermediate service receives the first forwarding request and collects the mirror traffic of the second intermediate service through the second traffic collection point to obtain the second mirror traffic;

[0022] Based on the second mirror traffic, the second mirror traffic is directed to the traffic parsing device, which then parses and identifies the second mirror traffic to obtain the second request start time, the second response start time, the second transaction processing time, and the second transaction distributed link ID of the second intermediate service transaction.

[0023] The first forwarding request carries the second transaction distributed link ID to form a second forwarding request, and the second forwarding request is sent to the target service.

[0024] Furthermore, the first mirror traffic is analyzed and identified, including:

[0025] Parse the session information in the first mirror traffic and extract the session header information;

[0026] Based on the header information and service rules, identify the application traffic of the first intermediate service to obtain the first application traffic.

[0027] Based on the first application traffic and the transaction configuration, the transaction is decoded at the first traffic collection point to extract the load information of the first intermediate service session and obtain the first request start time, first response start time, first transaction processing time and first transaction distributed link ID of the first intermediate service transaction.

[0028] The transaction configuration includes the matching conditions for the same transaction, the fields to be extracted, and the extraction method.

[0029] Furthermore, it also includes:

[0030] The target service processes the second forwarding request, obtains the processing result, and returns the processing result to the second service.

[0031] The second service will process the results into a first response and forward it to the first service.

[0032] The first service will generate a second response from the first response result and return it to the access point.

[0033] Furthermore, the uplink penetration latency, downlink penetration latency, and processing latency of each intermediate service and / or target service include:

[0034] The uplink penetration delay of the first intermediary service = T2 - T1, where T2 is the start time of the transaction request for the second intermediary service and T1 is the start time of the transaction request for the first intermediary service.

[0035] Downlink penetration latency of the first intermediary service = T6 - T5, where T6 is the start time of the transaction response for the first intermediary service and T5 is the start time of the transaction response for the second intermediary service;

[0036] The uplink penetration delay of the second intermediary service = T3 - T2, where T3 is the start time of the transaction request for the target service and T2 is the start time of the transaction request for the second intermediary service.

[0037] Downlink penetration latency of the second intermediary service = T5 - T4, where T5 is the start time of the transaction response of the second intermediary service and T4 is the start time of the transaction response of the target service;

[0038] The processing delay of the first intermediary service = T6 - T1, where T6 is the start time of the transaction response of the first intermediary service and T1 is the start time of the transaction request of the first intermediary service.

[0039] The processing latency of the second intermediary service = T5 - T2, where T5 is the start time of the transaction response of the second intermediary service, and T2 is the start time of the transaction request of the second intermediary service;

[0040] The processing latency of the target service = T4 - T3, where T4 is the start time of the transaction response of the target service and T3 is the start time of the transaction request of the target service.

[0041] Furthermore, the method also includes:

[0042] Construct latency metrics with P95 and P99 quantiles;

[0043] The overall access quality of the service system is assessed based on latency metrics and the processing latency of each intermediate service and / or target service.

[0044] Secondly, this invention also provides a system for analyzing transaction penetration latency based on transaction decoding, the system comprising:

[0045] The acquisition unit is used by the intermediate service to receive transaction requests sent by the access end;

[0046] The transaction decoding unit is used to decode the mirrored traffic transaction based on the transaction request to obtain the request start time, response start time, transaction processing time and transaction distributed link ID of each transaction for each intermediate service.

[0047] The latency calculation unit is used to associate business transactions with various intermediate services and target services based on the transaction distributed link ID, and to calculate the uplink penetration latency, downlink penetration latency and processing latency of business transactions in various intermediate services and / or target services according to the access topology map; where the access topology map refers to the access relationship of business transactions.

[0048] Furthermore, the execution process of the transaction decoding unit is as follows:

[0049] Based on the transaction request, the mirror traffic of each intermediate service is collected through several traffic collection points deployed in the network; the traffic collection points are deployed at the traffic exit or traffic entry point of the network device corresponding to each intermediate service.

[0050] Based on the mirrored traffic, the application traffic of each intermediate service is identified according to service rules; the intermediate services include the first intermediate service and the second intermediate service.

[0051] Based on application traffic and transaction configuration, the system decodes transactions at each traffic collection point to obtain the request start time, response start time, transaction processing time, and transaction distributed link ID for each transaction in each intermediate service.

[0052] Thirdly, the present invention provides a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the above-mentioned method for analyzing business transaction penetration latency based on transaction decoding.

[0053] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0054] This invention relates to a method, system, and medium for analyzing transaction penetration latency in business transactions based on transaction decoding. This invention does not require data embedding in the corresponding business, and is non-intrusive and imperceptible to the business system. This invention is based on application identification and transaction parsing using bypass traffic mirroring, and realizes distributed ID association based on the parsed transactions. It has no impact on the performance or functionality of the business system and will not occupy the system resources of the machine where the business system is located. Attached Figure Description

[0055] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0056] Figure 1 This is a flowchart of the method for analyzing transaction penetration latency based on transaction decoding in this invention;

[0057] Figure 2 This is a flowchart illustrating the implementation of the method for analyzing transaction penetration latency based on transaction decoding in this invention.

[0058] Figure 3 This is a system structure block diagram of the present invention based on transaction decoding analysis of business transaction penetration latency. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0060] This invention primarily achieves application identification and transaction decoding using network traffic data. Specifically, it identifies and parses application transactions based on bypass traffic mirroring, and associates distributed IDs based on the parsed transactions. This method is used in real-world business access scenarios to analyze business transaction penetration latency based on distributed link IDs.

[0061] Example 1

[0062] like Figure 1 and Figure 2 As shown, the present invention provides a method for analyzing transaction penetration latency based on transaction decoding, the method comprising:

[0063] Step 1: The intermediate service receives the transaction request sent by the access end;

[0064] Step 2: Based on the transaction request, decode the mirrored traffic transaction to obtain the request start time, response start time, transaction processing time, and transaction distributed link ID for each transaction from each intermediate service; there can be 1, 2, 3, or more intermediate services.

[0065] Step 3: Associate business transactions of each intermediate service and target service based on the transaction distributed link ID, and calculate the uplink penetration delay, downlink penetration delay and processing delay of business transactions in each intermediate service and / or target service according to the access topology.

[0066] This embodiment is implemented using a network comprising two intermediate services (Service 1 and Service 2) and a target service (Service 3). Each service has a traffic collection point (probe) at either the traffic egress or ingress point of the corresponding network device. The mirrored traffic of the network device is collected through these traffic collection points. The network device is a switch or router in the network topology. The access topology diagram refers to the access relationship of business transactions, i.e., the access sequence from the access point -> Service 1 -> Service 2 -> Service 3 -> Service 2 -> Service 1 -> Access Point.

[0067] In this embodiment, step 2 specifically includes:

[0068] Based on the transaction request, the mirror traffic of each intermediate service is collected through several traffic collection points deployed in the network; the traffic collection points are deployed at the traffic exit or traffic entry point of the network device corresponding to each intermediate service.

[0069] Based on the mirrored traffic, the application traffic of each intermediate service is identified according to the service rules; the intermediate services include the first intermediate service (service 1) and the second intermediate service (service 2).

[0070] Based on application traffic and transaction configuration, the system decodes transactions at each traffic collection point to obtain the request start time, response start time, transaction processing time, and transaction distributed link ID for each transaction in each intermediate service.

[0071] Specifically, the steps for decoding transactions based on mirrored traffic are as follows:

[0072] The first intermediary service (service 1) receives transaction requests sent by the access end;

[0073] Based on the transaction request, the mirror traffic of the first intermediate service is collected through the first traffic collection point to obtain the first mirror traffic;

[0074] Based on the first mirror traffic, the first mirror traffic is directed to the traffic analysis device, which analyzes and identifies the first mirror traffic to obtain the first request start time, first response start time, first transaction processing time, and first transaction distributed link ID of the first intermediate service transaction.

[0075] The transaction request carries the first transaction distributed link ID to form a first forwarding request, and the first forwarding request is sent to the second intermediate service (service 2).

[0076] The second intermediate service receives the first forwarding request and collects the mirror traffic of the second intermediate service through the second traffic collection point to obtain the second mirror traffic;

[0077] Based on the second mirror traffic, the second mirror traffic is directed to the traffic parsing device, which then parses and identifies the second mirror traffic to obtain the second request start time, the second response start time, the second transaction processing time, and the second transaction distributed link ID of the second intermediate service transaction.

[0078] The first forwarding request carries the second transaction distributed link ID to form a second forwarding request, and the second forwarding request is sent to the target service.

[0079] Specifically, the process of parsing and identifying the first mirror traffic is similar to that of parsing and identifying the second mirror traffic. Here, only the steps for parsing and identifying the first mirror traffic are given, which include:

[0080] Parse the session information in the first mirror traffic and extract the session header information (client, client port, server, server port);

[0081] Based on the header information and service rules (usually IP or IP + port), the application traffic of the first intermediate service is identified to obtain the first application traffic;

[0082] Based on the first application traffic and the transaction configuration, the transaction is decoded at the first traffic collection point to extract the load information of the first intermediate service session and obtain the first request start time, first response start time, first transaction processing time and first transaction distributed link ID of the first intermediate service transaction.

[0083] The transaction configuration includes the matching conditions for the same transaction, the fields to be extracted, and the extraction method.

[0084] This embodiment also includes:

[0085] The target service processes the second forwarding request, obtains the processing result, and returns the processing result to the second service.

[0086] The second service will process the results into a first response and forward it to the first service.

[0087] The first service will generate a second response from the first response result and return it to the access point.

[0088] Furthermore, the uplink penetration latency, downlink penetration latency, and processing latency of each intermediate service and / or target service include:

[0089] The uplink penetration delay of the first intermediary service = T2 - T1, where T2 is the start time of the transaction request for the second intermediary service and T1 is the start time of the transaction request for the first intermediary service.

[0090] Downlink penetration latency of the first intermediary service = T6 - T5, where T6 is the start time of the transaction response for the first intermediary service and T5 is the start time of the transaction response for the second intermediary service;

[0091] The uplink penetration delay of the second intermediary service = T3 - T2, where T3 is the start time of the transaction request for the target service and T2 is the start time of the transaction request for the second intermediary service.

[0092] Downlink penetration latency of the second intermediary service = T5 - T4, where T5 is the start time of the transaction response of the second intermediary service and T4 is the start time of the transaction response of the target service;

[0093] The processing delay of the first intermediary service = T6 - T1, where T6 is the start time of the transaction response of the first intermediary service and T1 is the start time of the transaction request of the first intermediary service.

[0094] The processing latency of the second intermediary service = T5 - T2, where T5 is the start time of the transaction response of the second intermediary service, and T2 is the start time of the transaction request of the second intermediary service;

[0095] The processing latency of the target service = T4 - T3, where T4 is the start time of the transaction response of the target service and T3 is the start time of the transaction request of the target service.

[0096] In practice, Figure 2 In this context, assuming a transaction requires access to two intermediate services before reaching the target service, the transaction process is as follows:

[0097] 1. The access point sends a transaction request;

[0098] 2. Service 1 receives the transaction request, decodes the transaction based on the mirrored traffic, generates the transaction distributed link ID, and sends the transaction request with this transaction distributed link ID to form the first intermediate request, and forwards the first intermediate request to Service 2;

[0099] 3. Service 2 receives the first intermediate request, reuses the transaction distributed link ID of Service 1 or regenerates a transaction distributed link ID based on the mirrored traffic transaction decoding, and carries the transaction distributed link ID or the regenerated transaction distributed link ID in the first intermediate request to form a second intermediate request, and forwards the second intermediate request to Service 3.

[0100] 4. Service 3 processes the second intermediate transaction request and returns the result to Service 2;

[0101] 5. Service 2 generates the first response result and returns it to Service 1;

[0102] 6. Service 1 generates a second response result and returns it to the access end.

[0103] Specifically, the decoding of mirrored traffic transactions includes:

[0104] Traffic collection points (probes) collect mirrored traffic from network devices and redirect it to traffic analysis devices. These devices analyze and identify the mirrored traffic, extracting session information and its headers (client, client port, server, server). Based on service rules (usually IP address or IP address + port), they identify the traffic of each intermediate service. According to transaction configurations, they extract the load information for each service session, completing transaction decoding for the corresponding service at each traffic collection point. Transaction configurations include: matching conditions for the same transaction, fields to be extracted, and extraction methods.

[0105] Based on the above steps, the request start time, response start time, transaction processing time, and transaction distributed link ID of each transaction for each service can be obtained.

[0106] Specifically, each transaction for each service is correlated in real time based on the transaction's distributed link ID, and the uplink latency of the business transaction at each traffic collection point is calculated according to the business access relationship. Figure 2 In the diagram: T1, T2, and T3 represent the start times of transaction requests for services 1, 2, and 3, respectively; T4, T5, and T6 represent the start times of transaction responses for services 3, 2, and 1, respectively, with service 3 being the target service. The latency data for each part is described below:

[0107] The uplink penetration latency of service 1 = T2 - T1;

[0108] The downlink penetration latency for service 1 is T6 - T5;

[0109] The uplink penetration latency for service 2 = T3 - T2;

[0110] The downlink penetration latency for service 2 is T5 - T4.

[0111] The processing latency of service 1 = T6 - T1;

[0112] The processing latency of service 2 = T5 - T2;

[0113] The processing latency of service 3 = T4 - T3.

[0114] Based on the above latency data, the latency of each transaction in each service can be obtained; the overall latency of each service can be analyzed from the perspective of service; latency indicators with P95 and P99 quantiles can be defined to evaluate the overall access quality of the service system; and the access latency of each transaction can be tracked from the perspective of a single transaction.

[0115] The advantages of this invention are as follows:

[0116] (1) Existing methods require data tracking in the corresponding business, which will cause intrusion into the business system. However, the present invention does not require data tracking in the corresponding business, and has no intrusion into the business system and is imperceptible.

[0117] (2) The present invention is based on the application identification transaction parsing of bypass traffic mirroring, and realizes the association of distributed IDs based on the parsed transactions. It has no impact on the performance and function of the business system and will not occupy the system resources of the machine where the business system is located.

[0118] (3) No modification is required to the existing system, and the implementation risk is extremely low; it is language and framework independent, and can be monitored as long as the business is written in Java, Go, Python or C++, as long as it communicates through the network; it captures data at the network level, including network latency, and can more realistically reflect the end-to-end user experience.

[0119] Example 2

[0120] like Figure 3 As shown, the difference between this embodiment and Embodiment 1 is that this embodiment provides a system based on transaction decoding analysis of business transaction penetration latency. This system includes:

[0121] The acquisition unit is used by the intermediate service to receive transaction requests sent by the access end;

[0122] The transaction decoding unit is used to decode the mirrored traffic transaction based on the transaction request to obtain the request start time, response start time, transaction processing time and transaction distributed link ID of each transaction for each intermediate service.

[0123] The latency calculation unit is used to associate business transactions with various intermediate services and target services based on the transaction distributed link ID, and to calculate the uplink penetration latency, downlink penetration latency and processing latency of business transactions in various intermediate services and / or target services according to the access topology map; where the access topology map refers to the access relationship of business transactions.

[0124] As a further implementation, the execution process of the transaction decoding unit is as follows:

[0125] Based on the transaction request, the mirror traffic of each intermediate service is collected through several traffic collection points deployed in the network; the traffic collection points are deployed at the traffic exit or traffic entry point of the network device corresponding to each intermediate service.

[0126] Based on the mirrored traffic, the application traffic of each intermediate service is identified according to service rules; the intermediate services include the first intermediate service and the second intermediate service.

[0127] Based on application traffic and transaction configuration, the system decodes transactions at each traffic collection point to obtain the request start time, response start time, transaction processing time, and transaction distributed link ID for each transaction in each intermediate service.

[0128] As a further implementation, the specific steps for decoding transactions based on mirrored traffic are as follows:

[0129] The first intermediary service receives transaction requests sent by the access end;

[0130] Based on the transaction request, the mirror traffic of the first intermediate service is collected through the first traffic collection point to obtain the first mirror traffic;

[0131] Based on the first mirror traffic, the first mirror traffic is directed to the traffic analysis device, which analyzes and identifies the first mirror traffic to obtain the first request start time, first response start time, first transaction processing time, and first transaction distributed link ID of the first intermediate service transaction.

[0132] The transaction request carries the first transaction distributed link ID to form a first forwarding request, and the first forwarding request is sent to the second intermediate service;

[0133] The second intermediate service receives the first forwarding request and collects the mirror traffic of the second intermediate service through the second traffic collection point to obtain the second mirror traffic;

[0134] Based on the second mirror traffic, the second mirror traffic is directed to the traffic parsing device, which then parses and identifies the second mirror traffic to obtain the second request start time, the second response start time, the second transaction processing time, and the second transaction distributed link ID of the second intermediate service transaction.

[0135] The first forwarding request carries the second transaction distributed link ID to form a second forwarding request, and the second forwarding request is sent to the target service.

[0136] The execution process of each unit can be carried out according to the method flow steps of transaction decoding analysis of business transaction penetration latency in Embodiment 1, and will not be described in detail in this embodiment.

[0137] Meanwhile, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for analyzing business transaction penetration latency based on transaction decoding.

[0138] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for analyzing transactional business transaction penetration latency based on transactional decoding, characterized in that, The method comprises: The intermediate service receives a transaction request sent by the access end; According to the transaction request, the mirror traffic transaction decoding is based on, the request start time, the response start time, the transaction processing time and the transaction distributed link id of each transaction of each intermediate service are obtained; Based on the transaction distributed link id, the business transaction of each intermediate service and the target service is associated, and according to the access topology graph, the uplink penetration delay, the downlink penetration delay and the processing delay of the business transaction in each intermediate service and / or target service are calculated; The access topology graph refers to the access relationship of the business transaction.

2. The method for analyzing business transaction penetration latency based on transaction decoding according to claim 1, wherein, According to the transaction request, the mirror traffic transaction decoding is based on, the request start time, the response start time, the transaction processing time and the transaction distributed link id of each transaction of each intermediate service are obtained, comprising: According to the transaction request, the mirror traffic of each intermediate service is collected through a plurality of traffic collection points deployed in the network; the traffic collection points are deployed at the traffic outlet or traffic inlet of the network equipment corresponding to each intermediate service; According to the mirror traffic, the application traffic of each intermediate service is identified based on the service rule; And according to the application traffic, the transaction decoding of each traffic collection point is carried out based on the transaction configuration, and the request start time, the response start time, the transaction processing time and the transaction distributed link id of each transaction of each intermediate service are obtained.

3. The method for decoding and analyzing business transaction penetration latency based on transactions as claimed in claim 2, wherein, The intermediate service includes a first intermediate service and a second intermediate service; the specific steps of mirror traffic transaction decoding are: The first intermediate service receives a transaction request sent by the access end; According to the transaction request, the mirror traffic of the first intermediate service is collected through the first traffic collection point to obtain the first mirror traffic; According to the first mirror traffic, the first mirror traffic is introduced to the traffic analysis device, and the first request start time, the first response start time, the first transaction processing time and the first transaction distributed link id of the first intermediate service transaction are obtained by the analysis device analyzing and identifying the first mirror traffic; The transaction request carries the first transaction distributed link id to form a first forwarding request, and the first forwarding request is sent to the second intermediate service; The second intermediate service receives the first forwarding request, collects the mirror traffic of the second intermediate service through the second traffic collection point to obtain the second mirror traffic; According to the second mirror traffic, the second mirror traffic is introduced to the traffic analysis device, and the second request start time, the second response start time, the second transaction processing time and the second transaction distributed link id of the second intermediate service transaction are obtained by the analysis device analyzing and identifying the second mirror traffic; The first forwarding request carries the second transaction distributed link id to form a second forwarding request, and the second forwarding request is sent to the target service.

4. The method for analyzing transactional business transaction penetration latency based on transactional decoding according to claim 3, wherein, The analysis and identification of the first mirror traffic comprises: The session information in the first mirror traffic is analyzed, and the header information of the session information is extracted; According to the header information, the application traffic of the first intermediate service is identified based on the service rule to obtain the first application traffic; According to the first application flow, based on the transaction configuration, transaction decoding of the first flow collection point is performed, load information of the first intermediate service session is extracted, and a first request start time, a first response start time, a first transaction processing time and a first transaction distributed link id of the first intermediate service transaction are obtained; The transaction configuration includes matching conditions of the same transaction, fields to be extracted and extraction methods.

5. The method for analyzing transactional business transaction penetration latency based on transactional decoding as claimed in claim 3 wherein, Further comprising: The target service processes the second forwarding request to obtain a processing result; And the processing result is returned to the second service; The second service forms a first response result from the processing result and forwards it to the first service; The first service forms a second response result from the first response result and returns it to the access end.

6. The method for analyzing transactional business transaction penetration latency based on transactional decoding as claimed in claim 3 wherein, The uplink penetration delay, the downlink penetration delay and the processing delay of each intermediate service and / or target service include: The uplink penetration delay of the first intermediate service is T2-T1, T2 is the transaction request start time of the second intermediate service, and T1 is the transaction request start time of the first intermediate service; The downlink penetration delay of the first intermediate service is T6-T5, T6 is the transaction response start time of the first intermediate service, and T5 is the transaction response start time of the second intermediate service; The uplink penetration delay of the second intermediate service is T3-T2, T3 is the transaction request start time of the target service, and T2 is the transaction request start time of the second intermediate service; The downlink penetration delay of the second intermediate service is T5-T4, T5 is the transaction response start time of the second intermediate service, and T4 is the transaction response start time of the target service; The processing delay of the first intermediate service is T6-T1, T6 is the transaction response start time of the first intermediate service, and T1 is the transaction request start time of the first intermediate service; The processing delay of the second intermediate service is T5-T2, T5 is the transaction response start time of the second intermediate service, and T2 is the transaction request start time of the second intermediate service; The processing delay of the target service is T4-T3, T4 is the transaction response start time of the target service, and T3 is the transaction request start time of the target service.

7. The method for decoding and analyzing business transaction penetration latency based on transactions as claimed in claim 1, wherein, The method further comprises: Building a delay index of P95, P99 and other quantiles; According to the delay index and the processing delay of each intermediate service and / or target service, the overall access quality of the service system is evaluated.

8. A system for analyzing transactional business transaction penetration latency based on transactional decoding, characterized in that, The system comprises: An acquisition unit configured to receive, by an intermediate service, a transaction request sent by an access end; A transaction decoding unit configured to, according to the transaction request, based on mirror flow transaction decoding, obtain a request start time, a response start time, a transaction processing time and a transaction distributed link id of each transaction of each intermediate service; A delay calculation unit configured to associate business transactions of each intermediate service and a target service based on the transaction distributed link id, and calculate uplink penetration delay, downlink penetration delay and processing delay of the business transactions in each intermediate service and / or target service according to an access topology diagram; wherein the access topology diagram refers to an access relationship of the business transactions.

9. The system for transaction-based decoding analysis of business transaction penetration latency according to claim 8, wherein, The execution process of the transaction decoding unit is: According to the transaction request, mirror traffic of each intermediate service is collected through a plurality of traffic collection points deployed in a network; the traffic collection points are deployed at a traffic outlet or a traffic inlet of a network device corresponding to each intermediate service; According to the mirror traffic, application traffic of each intermediate service is identified based on a service rule; the intermediate service includes a first intermediate service and a second intermediate service; And according to the application traffic, transaction decoding of each traffic collection point is performed based on a transaction configuration to obtain a request start time, a response start time, a transaction processing time and a transaction distributed link id of each transaction of each intermediate service.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the method for analyzing business transaction penetration delay based on transaction decoding according to any one of claims 1 to 7.

Images