Method and system for accurately monitoring response time of server

Abstract

The invention provides a method and a system for accurately monitoring response time of a server, and belongs to the field of distributed system performance monitoring. The method comprises the following steps: firstly, capturing network flow round-trip data packets related to a server and a client session, respectively analyzing the captured round-trip data packets to judge whether the data packets are request flow or response flow, and extracting respective timestamps; judging whether the traffic is from a server side or a client side according to the configuration file; if the client side comes from the server side, setting a time point at which the client side request information is captured as T1 and a time point at which the server response information is captured as T2 according to the timestamps of the request flow and the response flow, and calculating the actual response time Tresponse of the server as T2-T1; if the data packet comes from the client side, Tresponse is calculated to be T2-T1-RTT, and RTT is round-trip time delay of the round-trip data packet. According to the method, the real-time performance, the accuracy and the precision of monitoring the response time of the server are improved by judging the deployment position and carrying out differential calculation.

Description

Technical Field

[0001] The present invention belongs to the field of distributed system performance monitoring, and particularly relates to a method and system for accurately monitoring the server response time. Background Art

[0002] With the in-depth digital transformation, servers, as the core infrastructure of enterprises, their response time directly determines the user experience and business continuity. Research shows that a web page loading delay exceeding 2 seconds will lead to a 47% increase in the user churn rate, and for every 100 ms increase in the application programming interface response time, the conversion rate of e-commerce platforms will decrease by 7%. Especially in fields with extremely high real-time requirements such as finance and healthcare, server response delays may cause transaction failures or data synchronization interruptions, resulting in direct economic losses. At the same time, with the popularization of technologies such as the Internet of Things, 5G communication, and edge computing, the server deployment environment has become increasingly complex, and cross-regional distribution and multi-node collaboration have become the norm. Traditional tools such as Ping and Traceroute can only measure network connectivity and are difficult to reflect the application processing time consumption; the analysis method based on application logs has problems such as log delay, loss, and invasive modification. Therefore, accurately monitoring and analyzing the server response time has become the core requirement for optimizing performance and ensuring service reliability.

[0003] In the prior art, traditional tools (such as Ping and Traceroute) are generally used to monitor the server response time. However, with the popularization of distributed systems, microservices, and cloud-native technologies, servers mostly have complex architectures with high concurrency, facing multi-level and cross-node request processing, and multiple links such as network, computing, and storage may have a superimposed impact on the server response time, resulting in fluctuations in the response time. And the existing measurement methods based on the above traditional monitoring tools do not consider the influence of different measurement positions (such as the server side or the client side) on the measurement results, resulting in data deviation. It can be seen that the existing measurement of server response time has serious data deviation and is not accurate. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, the present invention aims to provide a method and system for accurately monitoring the server response time. By distinguishing whether the round-trip data packet comes from the server side or the client side, different server response time calculation methods are adopted for differential calculation, so as to accurately calculate the server response time, improve the accuracy, real-time performance, and universality of monitoring, and at the same time reduce the invasiveness of time monitoring.

[0005] To achieve the above object, the embodiments of the present invention adopt the following technical solutions:

[0006] In the first aspect, the embodiments of the present invention provide a method for accurately monitoring the server response time, including the following steps:

[0007] Step S1, capture the round-trip data packets of the network traffic related to the server-client session;

[0008] Step S2, respectively analyze the captured round-trip data packets, and determine whether the data packet is request traffic or response traffic according to the data packet timing relationship, TCP flag bits and characteristic fields, and extract their respective timestamps;

[0009] Step S3, determine whether the traffic comes from the server side or the client side according to the predefined information in the configuration file; if it comes from the server side, go to Step S4; if it comes from the client side, go to Step S5;

[0010] Step S4, according to the timestamps of the captured request traffic and response traffic, set the time point when the client request information is captured as T1, and the time point when the server response information is captured as T2, and calculate the actual server response time T response as T2 - T1;

[0011] Step S5, according to the timestamps of the captured request traffic and response traffic, set the time point when the client request information is captured as T1, and the time point when the server response information is captured as T2, and the round-trip delay of the round-trip data packet is RTT; calculate the actual server response time T response as T2 - T1 - RTT.

[0012] As a preferred embodiment of the present invention, in Step S1, the round-trip data packets of the network traffic related to the server-client session are captured in a mirroring manner.

[0013] As a preferred embodiment of the present invention, the mirroring manner is implemented by setting a layer 2 manageable switch.

[0014] As a preferred embodiment of the present invention, when Step S2 judges the data packet according to the data packet timing relationship, TCP flag bits and characteristic fields, it specifically includes:

[0015] When the traffic target port is the standard port of the server or the first data packet has a SYN field, the header field contains the specified target server domain name, and the client device information is identified, it is determined that the data packet is request traffic;

[0016] When the traffic target port is a temporary port or the TCP flag bit contains a SYN+ACK field, the header field contains the field identifying the server software, and the format of the response body is declared, it is determined that the data packet is response traffic.

[0017] As a preferred embodiment of the present invention, the predefined content in the configuration file in Step S3 includes the type, protocol, port number, IP address of the application to which the traffic belongs, and the location of traffic collection.

[0018] In a second aspect, an embodiment of the present invention further provides a system for accurately monitoring the server response time, the system includes: a traffic collection module, a data packet parsing module, a location identification module, a server scenario calculation module, a client scenario calculation module, and a display module; wherein,

[0019] The traffic collection module is used to capture the network traffic round-trip data packets related to the server-client session;

[0020] The data packet parsing module is used to respectively parse the captured round-trip data packets, judge whether the data packet is a request traffic or a response traffic according to the data packet time sequence relationship, TCP flag bits and characteristic fields, and extract their respective timestamps;

[0021] The location identification module is used to judge whether the traffic comes from the server side or the client side according to the predefined information in the configuration file; if it comes from the server side, the server scenario calculation module is started; if it comes from the client side, the client scenario calculation module is started;

[0022] The server scenario calculation module is used to set the time point when the client request information is captured as T1 and the time point when the server response information is captured as T2 according to the timestamps of the captured round-trip data packets, and calculate the actual server response time T response as T2 - T1 and send the calculated server response time T response to the display module;

[0023] The client scenario calculation module is used to set the time point when the client request information is captured as T1 and the time point when the server response information is captured as T2 according to the timestamps of the captured round-trip data packets, and the round-trip delay of the round-trip data packet is RTT; calculate the actual server response time T response as T2 - T1 - RTT and send the calculated server response time T response to the display module;

[0024] The display module is used to display the captured data packets and the corresponding server response time.

[0025] As a preferred embodiment of the present invention, the traffic collection module captures the network traffic round-trip data packets related to the server-client session in a mirroring manner.

[0026] As a preferred embodiment of the present invention, the traffic collection module is implemented by a layer-2 manageable switch.

[0027] As a preferred embodiment of the present invention, the data packet parsing module is further used for:

[0028] When the traffic target port is the standard port of the server, or the first data packet has a SYN field, the packet header field contains the specified target server domain name, and the client device information is identified, the data packet is determined to be a request traffic;

[0029] When the traffic target port is a temporary port, or the TCP flag contains the SYN+ACK field, the packet header field contains the field identifying the server software, and the format of the response body is declared, the data packet is determined to be a response traffic.

[0030] As a preferred embodiment of the present invention, the predefined content in the configuration file in the location identification module includes the type, protocol, port number, IP address, URL information of the application to which the traffic belongs, and the location where the traffic is collected.

[0031] The technical solutions provided by the embodiments of the invention have the following beneficial effects:

[0032] The method and system for accurately monitoring the server response time provided by the embodiments of the present invention first analyze the request traffic and response traffic in the round-trip data packets based on the captured network traffic round-trip data packets related to the server-client session, and extract their respective timestamps; then, according to the configuration file matching judgment, confirm whether the data packet comes from the client side or the server side, and match the appropriate algorithm to accurately calculate the server response time; when it comes from the server side, the difference between the time point when the server response is captured and the time point when the client request information is captured is used as the actual server response time; when it comes from the client side, the difference between the time point when the server response information is captured and the time point when the client request information is captured is used, and then the round-trip delay of the round-trip data packet is subtracted as the actual server response time. The present invention improves the accuracy of the monitoring result by judging the deployment location and performing differential calculations, breaks through the limitation of traditional tools (such as Ping) that only measure the network layer delay, realizes the end-to-end full-link time-consuming decomposition, improves the accuracy, real-time and universality of the server response time monitoring, and at the same time reduces the invasiveness of the time monitoring.

[0033] Of course, it is not necessary for any product or method implementing the present invention to achieve all the above-mentioned advantages at the same time. Brief Description of the Drawings

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0035] Figure 1 It is the flowchart of the method for accurately monitoring the server response time described in the embodiments of the present invention. Detailed implementation manners

[0036] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can also be combined with each other.

[0037] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In the description of the present invention, the terms "first", "second", "third", "fourth", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0038] Based on the problem of monitoring the server response time, the embodiments of the present invention provide a method and system for accurately monitoring the server response time. After traffic access, firstly, traffic analysis is performed through a configuration file to obtain the monitoring location, and the server side and the client side are distinguished; differential calculations are performed based on different locations to calculate the server response time respectively, and different characteristics of each location are fully considered during the calculation, so as to achieve accurate monitoring of the server response time.

[0039] As Figure 1 shown, the method for accurately monitoring the server response time includes the following steps:

[0040] Step S1, capturing the network traffic round-trip data packets related to the server-client session.

[0041] In this step, for the relevant network traffic round-trip data packets, mirroring is used for capture. Specifically, the mirroring function of the network traffic round-trip data packets can be implemented by setting a layer-2 manageable switch. Through the mirroring operation, different source traffic information is copied to a specified port.

[0042] Specifically, traffic collection, that is, capturing data packets, is performed by setting a traffic collection module. When performing data packet capture, there are two cases. When the monitored object is on the client side, the traffic collection module is deployed on the client side, and when the monitored object is on the server side, the traffic collection module is deployed on the server side. The specific deployment of the traffic collection module is based on the actual monitoring situation.

[0043] Step S2: Parse the captured round-trip data packets respectively, and determine whether the data packet is request traffic or response traffic according to the data packet timing relationship, TCP flag bits, and characteristic fields, and extract their respective timestamps.

[0044] In this step, when judging the data packet according to the data packet timing relationship, TCP flag bits, and characteristic fields, it specifically includes:

[0045] When the traffic target port is the standard port of the server or the SYN field exists in the first data packet, the message header field contains the specified target server domain name, and the client device information is identified, it is determined that the current data packet is request traffic;

[0046] When the traffic target port is a temporary port or the TCP flag bit contains the SYN+ACK field, the message header field contains the field identifying the server software, and the format of the response body is declared, it is determined that the data packet is response traffic.

[0047] Steps S1 and S2 realize the real-time monitoring of the server response time by capturing the round-trip data packets of network traffic and parsing and extracting timestamps, breaking through the limitation of traditional tools (such as Ping) that only measure the network layer delay. At the same time, when parsing the data packets, multiple standards such as the data packet timing relationship, TCP flag bits, and characteristic fields are introduced to judge whether the data packet is request traffic or response traffic, improving the accuracy of the judgment.

[0048] Step S3: Judge whether the traffic comes from the server side or the client side according to the predefined information in the configuration file; if it comes from the server side, go to step S4; if it comes from the client side, go to step S5.

[0049] In this step, according to the configuration file, it is automatically judged whether the traffic comes from the server side or the client side; it can be predefined in the configuration file, and the predefined content includes the type, protocol, port number, IP address, URL information of the application to which the traffic belongs, and the location where the traffic is collected. The location where the traffic is collected is the deployment location of the traffic collection module, which is generally defaulted to the server side; if the traffic collection module is located on the client side, it can be defined as the client side. The purpose of the configuration file is to dynamically adjust the calculation method of the server response time according to the deployment location of the traffic collection module. Different calculation formulas are referenced for different deployment locations, so as to calculate the server response time more accurately. This step uses the configuration file to predefine information such as the type, protocol, port number, and IP address of the application to which the traffic belongs, and can dynamically adjust the monitoring strategy according to different deployment locations, thereby improving the flexibility and adaptability of the monitoring.

[0050] Step S4: According to the timestamps of the extracted request traffic and response traffic, set the time point when the client request information is captured as T1, and the time point when the server response information is captured as T2, and calculate the actual server response time T response It is T2 - T1.

[0051] In this step, the calculation formula for the actual server response time is:

[0052] The actual server response time = the difference between the time point T1 when the client request information is captured and the time point T2 when the server response is captured; that is

[0053] T response = T2 - T1

[0054] In the server-side scenario: The client initiates a server request with a non-zero TCP payload, and the server makes an application-layer response to this request. At this time, it is considered that both are in the same local area network, thus ignoring the network latency.

[0055] Step S5: According to the timestamps of the extracted request traffic and response traffic, set the time point when the client request information is captured as T1, the time point when the server response information is captured as T2, and the round-trip delay of the round-trip data packet as RTT; calculate the actual server response time T response It is T2 - T1 - RTT.

[0056] In this step, the actual server response time T response The calculation formula is as follows:

[0057] T response = T2 - T1 - RTT

[0058] In the client-side scenario, it belongs to a public network environment with large fluctuations. At this time, the server response time includes network latency. By deducting the round-trip time (RTT) of the network, the influence of the transmission path on the response time is eliminated, so as to achieve accurate monitoring of the server response time.

[0059] It can be seen that the embodiment of the present invention calculates the server response time differently according to the traffic source (server side or client side), directly calculates the response time on the server side, and considers the round-trip delay (RTT) on the client side to improve the accuracy of monitoring; at the same time, data monitoring is carried out by a non-invasive method, avoiding interference with the normal business process and improving the reliability of monitoring.

[0060] Based on the same idea, the embodiment of the present invention also provides a system for accurately monitoring the server response time. The system includes: a traffic collection module, a data packet parsing module, a location identification module, a server scenario calculation module, a client scenario calculation module, and a display module.

[0061] Among them, the traffic collection module is used to capture the network traffic round-trip data packets related to the server-client session;

[0062] The data packet parsing module is used to parse the captured round-trip data packets respectively, and judge whether the data packet is a request traffic or a response traffic according to the data packet timing relationship, TCP flag bits and characteristic fields, and extract their respective timestamps;

[0063] The location identification module is used to judge whether the traffic comes from the server side or the client side according to the predefined information in the configuration file; if it comes from the server side, the server scenario calculation module is started; if it comes from the client side, the client scenario calculation module is started;

[0064] The server scenario calculation module is used to set the time point when the client request information is captured as T1 and the time point when the server response information is captured as T2 according to the timestamps of the extracted request traffic and response traffic, and calculate the actual server response time T response as T2 - T1 and send the calculated server response time T response to the display module;

[0065] The client scenario calculation module is used to set the time point when the client request information is captured as T1 and the time point when the server response information is captured as T2 according to the timestamps of the extracted request traffic and response traffic, and the round-trip delay of the round-trip data packet is RTT; calculate the actual server response time T response as T2 - T1 - RTT and send the calculated server response time T response to the display module;

[0066] The display module is used to display the captured data packets and the corresponding server response times.

[0067] In this embodiment, each module is implemented by a processor, and a memory is appropriately added when storage is required. Among them, the processor may be, but is not limited to, a microprocessor MPU, a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), other programmable logic devices, discrete gate, transistor logic devices, discrete hardware components, etc. The memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0068] In the above embodiment, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.).

[0069] In addition, it should be noted that the monitoring system for the server response time in this embodiment corresponds to the monitoring method for the server response time. The description and limitation of the method also apply to the system, and will not be repeated here.

[0070] As can be seen from the above technical solutions, for the method and system for accurately monitoring the server response time provided by the embodiments of the present invention, first, the request traffic and response traffic in the round-trip data packet are analyzed, and their respective timestamps are extracted; then, according to the matching judgment of the configuration file, it is confirmed whether the data packet comes from the client side or the server side, and a suitable algorithm is matched to accurately calculate the server response time; when it comes from the server side, the difference between the time point when the server response is captured and the time point when the client request information is captured is used as the actual server response time; when it comes from the client side, the difference between the time point when the server response information is captured and the time point when the client request information is captured is subtracted by the round-trip delay of the round-trip data packet as the actual server response time. By judging the deployment location and performing differential calculations, the present invention improves the accuracy of the monitoring results, breaks through the limitation of traditional tools (such as Ping) that only measure the network layer delay, realizes the end-to-end full-link time-consuming decomposition, improves the accuracy, real-time performance and universality of the server response time monitoring, and at the same time reduces the invasiveness of the time monitoring.

[0071] The above description is only the preferred embodiment of the present invention and the explanation of the applied technical principle, and is not intended to limit the scope of the present invention claimed, but only represents the preferred embodiment of the present invention. Those skilled in the art should understand that the scope of the invention involved in the present invention is not limited to the technical solution formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present invention.

Claims

1. A method for accurately monitoring the server response time, characterized in that, It includes the following steps: Step S1, capture the round-trip data packets of the network traffic related to the server-client session; Step S2, respectively parse the captured round-trip data packets, and judge whether the data packets are request traffic or response traffic according to the packet timing relationship, TCP flag bits and characteristic fields, and extract their respective timestamps; Step S3, judge whether the traffic comes from the server side or the client side according to the predefined information in the configuration file; If it comes from the server side, go to step S4; if it comes from the client side, go to step S5; Step S4. According to the timestamps of the extracted request traffic and response traffic, set the time point when the client request information is captured as T1, and the time point when the server response information is captured as T2, and calculate the actual server response time T response as T2 - T1; Step S5, according to the timestamps of the extracted request traffic and response traffic, set the time point when the client request information is captured as T1, the time point when the server response information is captured as T2, and the round-trip delay of the round-trip data packets as RTT; Calculate the actual response time T of the server response It is T2 - T1 - RTT.

2. The method for precisely monitoring the server response time according to claim 1, characterized in that, In step S1, the round-trip data packets of the network traffic related to the server-client session are captured in a mirroring manner.

3. The method for accurately monitoring the server response time as claimed in claim 2, wherein, The mirroring manner is implemented by setting a layer-2 manageable switch.

4. The method for accurately monitoring the server response time according to claim 1, wherein When step S2 judges the data packets according to the packet timing relationship, TCP flag bits and characteristic fields, it specifically includes: When the traffic target port is the standard port of the server or the first data packet has a SYN field, the header field contains the specified target server domain name, and the client device information is identified, it is determined that the data packet is request traffic; When the traffic target port is a temporary port or the TCP flag bit contains the SYN+ACK field, the header field contains the field identifying the server software, and the format of the response body is declared, it is determined that the data packet is response traffic.

5. The method for accurately monitoring the server response time according to claim 4, characterized in that, The predefined content in the configuration file in step S3 includes the type, protocol, port number, IP address, URL information of the application to which the traffic belongs, and the location where the traffic is collected.

6. A system for precisely monitoring the server response time, characterized in that, The system includes: a traffic collection module, a data packet parsing module, a location identification module, a server scenario calculation module, a client scenario calculation module, and a display module; among them, The traffic collection module is used to capture the round-trip data packets of the network traffic related to the server-client session; The data packet parsing module is used to respectively parse the captured round-trip data packets, and judge whether the data packets are request traffic or response traffic according to the packet timing relationship, TCP flag bits and characteristic fields, and extract their respective timestamps; The location identification module is used to judge whether the traffic comes from the server side or the client side according to the predefined information in the configuration file; if it comes from the server side, start the server scenario calculation module; if it comes from the client side, start the client scenario calculation module; The server scenario calculation module is used to calculate the actual response time T of the server according to the timestamps of the extracted request traffic and response traffic. Set the time point when the client request information is captured as T1, and the time point when the server response information is captured as T2 response as T2 - T1 and send the calculated server response time T response to the display module; The client scenario calculation module is used to set the time point when the client request information is captured as T1, the time point when the server response information is captured as T2, and the round-trip delay of the round-trip data packet as RTT according to the timestamps of the extracted request traffic and response traffic; calculate the actual server response time T response as T2 - T1 - RTT and send the calculated server response time T response to the display module; The display module is used to display the captured data packets and the corresponding server response time.

7. The system for accurately monitoring the server response time according to claim 6, wherein, The traffic collection module captures the round-trip data packets of the network traffic related to the server-client session in a mirroring manner.

8. The system for accurately monitoring the server response time according to claim 7, characterized in that, The traffic collection module is implemented by a layer-2 manageable switch.

9. The system for precisely monitoring the server response time according to claim 6, wherein The data packet parsing module is further used for: When the traffic target port is the standard port of the server or the first data packet has a SYN field, the header field contains the specified target server domain name, and the client device information is identified, it is determined that the data packet is request traffic; When the traffic target port is a temporary port, or the TCP flag contains the SYN+ACK field, the packet header field contains the field identifying the server software, and the format of the declared response body is specified, the data packet is determined to be response traffic.

10. The system for accurately monitoring the server response time according to claim 9, characterized in that The predefined content in the configuration file in the location identification module includes the type of the application to which the traffic belongs, the protocol, the port number, the IP address, the URL information, and the location where the traffic is collected.

Images