CN116346918B - Performance bottleneck identification method and system for network event - Google Patents

Performance bottleneck identification method and system for network event Download PDF

Info

Publication number
CN116346918B
CN116346918B CN202310620214.8A CN202310620214A CN116346918B CN 116346918 B CN116346918 B CN 116346918B CN 202310620214 A CN202310620214 A CN 202310620214A CN 116346918 B CN116346918 B CN 116346918B
Authority
CN
China
Prior art keywords
network event
processed
timer
network
trigger
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310620214.8A
Other languages
Chinese (zh)
Other versions
CN116346918A (en
Inventor
李海兵
康江彬
叶子聪
李小壮
王柄淋
王妍兰
王文娟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianyi Cloud Technology Co Ltd
Original Assignee
Tianyi Cloud Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tianyi Cloud Technology Co Ltd filed Critical Tianyi Cloud Technology Co Ltd
Priority to CN202310620214.8A priority Critical patent/CN116346918B/en
Publication of CN116346918A publication Critical patent/CN116346918A/en
Application granted granted Critical
Publication of CN116346918B publication Critical patent/CN116346918B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/566—Grouping or aggregating service requests, e.g. for unified processing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/62—Establishing a time schedule for servicing the requests
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The embodiment of the specification mainly relates to the technical field of performance bottleneck identification, in particular to a method and a system for identifying the performance bottleneck of a network event. A method for identifying performance bottlenecks of network events comprises the following steps: collecting network events to be processed, and adding a certain number of network events to be processed into a network event list; setting a trigger and a timer, wherein the timer is provided with a target timing value; starting a timer and sequentially processing network events to be processed in a network event list; determining whether a performance bottleneck occurs through a trigger and a timer: if the network event to be processed is completed in the target timing value, judging that the network event has no performance bottleneck; and if the network event to be processed exceeds the target timing value, judging that the network event has performance bottleneck. The method and the system of the embodiment of the specification can enable the performance bottleneck of the network event to be found in time and enable the specific situation of the performance bottleneck to be obtained effectively.

Description

Performance bottleneck identification method and system for network event
Technical Field
The embodiment of the specification mainly relates to the technical field of performance bottleneck identification, in particular to a method and a system for identifying the performance bottleneck of a network event.
Background
The content distribution network comprises a plurality of groups of cache servers distributed in each region, wherein the cache servers are based on the secondary development of the nginx, collect network events through an epoll mechanism and serially process the network events in the process. Performance bottlenecks may occur if the cache server is relatively time consuming to process a network event. If the performance bottleneck cannot be found in time, the subsequent network events cannot be processed in time, and the satisfaction degree of the user is further affected.
In addition, the cache server consumes a long time for processing a certain network event, certain problems must exist, and operation and maintenance personnel need to analyze and solve the problems, so that similar situations are avoided in the later stage. However, the performance bottleneck is sporadic, and if not found at the time, the later period is difficult to reproduce, so that the operation and maintenance personnel can hardly obtain the specific condition of the performance bottleneck, and the corresponding problem cannot be analyzed.
Disclosure of Invention
Aiming at the problems existing in the prior art, the embodiment of the specification provides a method and a system for identifying the performance bottleneck of a network event, so as to solve the problems that the performance bottleneck of the network event in the prior art cannot be found in time and the specific situation of the performance bottleneck cannot be obtained effectively.
In a first aspect, an embodiment of the present disclosure provides a method for identifying a performance bottleneck of a network event, including the following steps:
collecting network events to be processed, and adding a certain number of network events to be processed into a network event list;
setting a trigger and a timer, wherein the timer is provided with a target timing value;
starting a timer and sequentially processing network events to be processed in a network event list;
determining whether a performance bottleneck occurs through a trigger and a timer: if the network event to be processed is completed in the target timing value, judging that the network event has no performance bottleneck; if the network event to be processed exceeds the target timing value, not all the network event to be processed is finished, judging that the network event has a performance bottleneck;
when judging that the network event has performance bottleneck, acquiring and storing current process stack information and stack variable information;
and deleting the trigger and the timer after all the network events to be processed in the network event list are completed.
Preferably, before collecting the network event to be processed, the method further comprises the following steps:
determining the number of network events to be processed for one cycle;
determining the number of network event adding pages according to the number of the network events to be processed;
and combining the corresponding number of network event adding pages according to the number of the network event adding pages to form a network event list, wherein each network event adding page is used for adding a network event to be processed.
Preferably, the setting trigger and the timer include:
registering a trigger, wherein the trigger is started when all network events to be processed in a network event list are completed, and the timer is deleted after the trigger is started;
a timer is registered which starts counting after start and generates an alarm signal when the timer count reaches a target value.
Preferably, determining whether a performance bottleneck occurs specifically includes:
state of the real-time scan timer: if the scanning timer is deleted, the network event to be processed is completely completed within a target timing value, and a first signal is formed; if the timer is scanned to generate an alarm signal, the network event to be processed is not completely finished beyond the target calibration value, and a second signal is formed.
Preferably, the deletion trigger and timer include:
after the first signal is acquired, deleting the trigger;
after the second signal is acquired, waiting for acquiring a third signal, and deleting the trigger and the timer after the third signal is acquired; and the third signal is formed after the network event to be processed exceeds the target timing value.
In a second aspect, embodiments of the present disclosure provide a performance bottleneck recognition system for network events, including:
the network event adding module is used for collecting network events to be processed and adding a certain number of network events to be processed into the network event list;
the trigger timer setting module is used for setting a trigger and a timer, and the timer is provided with a target timing value;
the network event processing module is used for starting a timer and sequentially processing network events to be processed in the network event list;
the performance bottleneck judging module is used for judging whether the performance bottleneck appears through a trigger and a timer: if the network event to be processed is completed in the target timing value, judging that the network event has no performance bottleneck; if the network event to be processed exceeds the target timing value, not all the network event to be processed is finished, judging that the network event has a performance bottleneck;
the performance bottleneck information acquisition module is used for acquiring and storing current process stack information and stack variable information when judging that a network event has a performance bottleneck;
and the trigger timer deleting module is used for deleting the trigger and the timer after all the network events to be processed in the network event list are completed.
Preferably, the method further comprises:
the network event number determining module is used for determining the number of network events to be processed in one cycle;
the network event adding page number determining module is used for determining the number of the network event adding pages according to the number of the network events to be processed;
and the network event list forming module is used for combining the network event adding pages with the corresponding number according to the network event adding page number to form a network event list, and each network event adding page is used for adding a network event to be processed.
Preferably, the trigger timer setting module specifically includes:
the trigger registration unit is used for registering a trigger, the trigger is started when all network events to be processed in the network event list are completed, and the timer is deleted after the trigger is started;
and the timer registration unit is used for registering a timer, starting to count after starting, and generating an alarm signal when the count of the timer reaches a target value.
Preferably, the performance bottleneck determination module specifically includes:
a timer state scanning unit, configured to scan the state of the timer in real time: if the scanning timer is deleted, the network event to be processed is completely completed in a target timing value; if the timer is scanned to generate an alarm signal, the network event to be processed is not completely finished beyond the target calibration value;
a first signal forming unit, configured to form a first signal when the network event to be processed is all completed within a target timing value;
and the second signal forming unit is used for forming a second signal when the network event to be processed exceeds the target timing value and is not completely finished.
Preferably, the trigger timer deleting module specifically includes:
the first deleting unit is used for deleting the trigger after the first signal is acquired;
the second deleting unit is used for waiting for acquiring a third signal after acquiring the second signal, and deleting the trigger and the timer after acquiring the third signal; and the third signal is formed after the network event to be processed exceeds the target timing value.
Advantageous effects
According to the performance bottleneck identification method and system, a proper number of network events to be processed can be processed one cycle by one cycle, whether the performance bottleneck occurs in the network events to be processed in the current cycle can be judged through the trigger and the timer, when the performance bottleneck occurs, related process stack information and stack variable information can be stored immediately, and an operation and maintenance person can be informed immediately, so that the operation and maintenance person can take corresponding solving measures in time, the problem that the user satisfaction is reduced due to the fact that the network events cannot be processed in time is avoided as much as possible, the specific situation of each performance bottleneck can be obtained almost effectively, and then the problem of each performance bottleneck can be analyzed and solved, and the subsequent generation of the performance bottleneck due to the same reason is avoided.
Drawings
Fig. 1 is a flow chart of a method for identifying a performance bottleneck of a network event according to an embodiment of the present disclosure;
fig. 2 is a schematic structural diagram of a performance bottleneck recognition system for network events according to an embodiment of the present disclosure.
Detailed Description
Embodiments of the present specification will be described in more detail below with reference to the accompanying drawings.
Example 1: a method for identifying performance bottlenecks of network events, as shown in fig. 1, comprises the following steps:
l1. collecting network events to be processed, and adding a certain number of network events to be processed into a network event list.
When a user initiates a request, a certain cache server in the cache server group closest to the user receives a network event, and the cache server is used for processing the network event and feeding back the processing result to the user.
The cache servers process the network events according to the received sequence of the network events, and generally one cache server has more network events to be processed at the same time (for example, 18 network events to be processed in sequence). The present embodiment requires that a certain number of network events to be processed be added to the network event list first, for example, 5 network events to be processed (1 st to 5 th network events to be processed) are added to the network event list. The network event list is set through the corresponding cache server.
The embodiment further includes the following steps before step L1:
and L11, determining the number of network events to be processed in one cycle. In this embodiment, the network events to be processed need to be processed cycle by cycle, so it is determined that each cycle needs to process several network events to be processed first, for example, the number of network events to be processed may be 5 in this embodiment.
In this embodiment, a loop needs to process a suitable number of network events to be processed, and if the number of network events processed in a loop is too small (for example, 1, 2), the processing efficiency of the network events is low; if the number of network event processing in one cycle is excessive (e.g., 30, 40), there may be two or more performance bottlenecks in one cycle, and when more than one second performance bottleneck occurs, the performance bottlenecks are not identified, so that the performance bottlenecks still cannot be found in time and the specific situation of the performance bottlenecks still cannot be effectively acquired.
And L12, determining the number of the network event adding pages according to the number of the network events to be processed. In this embodiment, one network event to be processed needs to be set on one network event adding page, so that one cycle needs to process several network events to be processed, and several network event adding pages need to be set correspondingly, for example, 5 network event adding pages may be set in this embodiment.
And L13, combining a corresponding number of network event adding pages according to the number of the network event adding pages to form a network event list, wherein each network event adding page is used for adding a network event to be processed. In this embodiment, the network event list is formed by combining network event adding pages, and when a cycle needs to process several network events to be processed, several network event adding pages are combined to form the network event list, for example, in this embodiment, 5 network event adding pages may be combined to form the network event list.
Returning to step L1, in this embodiment, only the network events to be processed need to be added to the network event list according to the sequence, and after the 5 network event adding pages all add the network events to be processed, the adding of the network events to be processed in one cycle is completed. In step L1, the specific number of the network events to be processed is not required, and only the network events to be processed need to be mechanically added into the network event list (when the last network event adding page completes the addition of the network events to be processed, the addition of the network events to be processed in one cycle is indicated), so that the addition operation of the network events to be processed in step L1 is simpler and more convenient and is less prone to error.
And L2, setting a trigger and a timer, wherein the timer is provided with a target calibration value. Wherein the setting trigger and timer specifically comprises:
l21. register a trigger that is started when all pending network events in the network event list are completed, and that deletes the timer after starting.
L22. register a timer that starts counting after start and generates an alarm signal when the timer counts up to a target value.
After the completion of step L1, a first trigger signal is generated, after the first trigger signal is received, the cache server registers a trigger and a timer immediately (how to register directly according to the prior art), after the registration of the trigger and the timer is completed, a second trigger signal is generated, and after the second trigger signal is received, step L3 is started.
In this embodiment, the timer may start to increase gradually from 0 until the target value is reached, and the target value may be 8 seconds, and then the timing process of the timer is 0,1 second, 2 seconds, and 3 seconds. When the timer times 8 seconds, the network event to be processed is not processed, and an alarm signal is generated. When the timer has not counted for 8 seconds and the network event to be processed has been processed, a third trigger signal is generated. After receiving the third trigger signal, the trigger is started immediately, the timer is deleted after the trigger is started, the timer cannot be continuously timed after being deleted, the time cannot be further timed to 8 seconds, and then an alarm signal cannot be generated.
And step L3, starting a timer and sequentially processing the network events to be processed in the network event list.
In this embodiment, the network event to be processed in the 1 st network event adding page is processed first, then the network event to be processed in the 2 nd network event adding page is processed after the processing, and then the network event to be processed in the 3 rd network event adding page is processed continuously. The specific processing mode of each network event to be processed can directly adopt the prior art.
L4. judging whether the performance bottleneck occurs or not through a trigger and a timer: if the network event to be processed is completed in the target timing value, judging that the network event has no performance bottleneck; and if the network event to be processed exceeds the target timing value, judging that the network event has performance bottleneck.
Under normal conditions, the 5 network events to be processed can be processed in one or two seconds, if the 5 network events to be processed are not processed yet in 8 seconds, the processing time of 1 network event to be processed is longer, and the probability is that performance bottleneck occurs.
In step L4 of this embodiment, determining whether the performance bottleneck occurs through the trigger and the timer specifically includes:
state of the real-time scan timer: if the scanning timer is deleted, the network event to be processed is completely completed within a target timing value, and a first signal is formed; if the timer is scanned to generate an alarm signal, the network event to be processed is not completely finished beyond the target calibration value, and a second signal is formed.
Step L4 and step L3 are performed synchronously, and the state of the timer is scanned in real time during the processing of the network event to be processed.
If the timer does not count for 8 seconds, the network event to be processed is processed, the trigger is started, the timer is deleted after the trigger is started, and the timer is scanned to be in a deleted state. I.e. if the scan timer is deleted, it means that the pending network event is all completed within the target timing value. At this time, a first signal is generated, and the first signal indicates that the network event to be processed has been completed within the target timing value, that is, the first signal indicates that the current cycle does not have a performance bottleneck of the network event. The first signal of the present embodiment may be a low level signal.
If the timer times 8 seconds, the pending network event has not been processed, an alarm signal is generated. I.e. if the timer is scanned to generate an alarm signal, the network event to be processed is not completely finished beyond the target timing value. At this time, a second signal is generated, and the second signal indicates that the network event to be processed is not completely completed beyond the target value, that is, the second signal indicates that the performance bottleneck of the network event occurs in the current cycle. The second signal of the present embodiment may be a high level signal.
And L5, when judging that the network event has a performance bottleneck, acquiring and storing current process stack information and stack variable information.
When the second signal appears, it indicates that the network event has a performance bottleneck, and at this time, the current process stack information and stack variable information need to be acquired immediately. After snapshot acquisition is completed, the snapshots may be sent and stored in the corresponding folders. A snapshot of the performance bottleneck is stored in a folder that may be named using the name of the corresponding cache server and the time of interception of the snapshot.
When the second signal is generated, the cache server also sends an alarm message to the operation and maintenance personnel, wherein the content of the alarm message can be that the XX cache server generates performance bottleneck in XX time, and the operation and maintenance personnel can immediately open the corresponding folder to check specific conditions after receiving the alarm message.
The operation and maintenance personnel can analyze the reasons of the performance bottleneck through the process stack information and the stack variable information, and solve the problem as soon as possible. If a certain performance bottleneck is not treated for 30 minutes, the problem can be solved by 5 minutes after intervention of operation and maintenance personnel, so that the performance bottleneck is ended by 5 minutes, and the performance bottleneck is not generated for the same reason later. If a certain performance bottleneck is not processed for any treatment, the performance bottleneck is maintained for 1 minute, after the intervention of operation and maintenance personnel, the problem is not found yet, and the performance bottleneck is finished, but the operation and maintenance personnel can find the problem and solve the problem by analyzing the process stack information and the stack variable information, so that the performance bottleneck cannot be generated for the same reason in the follow-up.
In addition, if the problem of a certain performance bottleneck is troublesome and cannot be solved quickly, the collection of new network events to be processed can be suspended, and the rest of network events to be processed can be scheduled to other cache servers for processing.
And L6, deleting the trigger and the timer after all the network events to be processed in the network event list are completed. Wherein, L6 specifically includes:
and L61, deleting the trigger after the first signal is acquired. In step L4, if the current cycle does not have a performance bottleneck for the network event, the network event to be processed in the network event list has been completed within the target timing value, a first signal is generated and the corresponding timer has been deleted. So in step L61, if the first signal is obtained, only the trigger needs to be deleted, since the timer has already been deleted.
L62, after the second signal is obtained, waiting for obtaining a third signal, and after the third signal is obtained, deleting the trigger and the timer; and the third signal is formed after the network event to be processed exceeds the target timing value. In step L4, if the current cycle has a performance bottleneck at the occurrence of a network event, a second signal is generated and the corresponding timer has not been deleted. So in step L62, if the second signal is obtained, it indicates that a performance bottleneck has occurred, at which time it is necessary to wait for the performance bottleneck to end and for all pending network events in the current loop to complete.
When the performance bottleneck is finished (the performance bottleneck is possibly automatically finished after being maintained for 2 minutes; or the operation and maintenance personnel solve the corresponding problem to finish the performance bottleneck), the network events to be processed in the network event list can be processed quickly. When the network event to be processed is completed, a third signal is generated, and the third signal may be a signal composed of two high levels and one low level. If the third signal is obtained in step L6, the network event to be processed indicating the present cycle is processed, and only the trigger and the timer need to be deleted.
After step L61 or step L62 is completed, it means that the current entire cycle is completed, and at this time, the process may proceed to the next cycle, and steps L1 to L6 may be repeated.
According to the method for identifying the performance bottleneck of the network event, a proper number of network events to be processed can be processed one cycle by one cycle, whether the performance bottleneck occurs in the network events to be processed in the current cycle can be judged through the trigger and the timer, when the performance bottleneck occurs, related process stack information and stack variable information can be stored immediately, and operation and maintenance personnel can be informed immediately, so that the operation and maintenance personnel can take corresponding solving measures in time, the problem that the user satisfaction is reduced because the network event cannot be processed in time is avoided as much as possible, the specific situation of each performance bottleneck can be obtained almost effectively, and then the problem of each performance bottleneck can be analyzed and solved, and the subsequent generation of the performance bottleneck due to the same reason is avoided.
Example 2: a performance bottleneck recognition system for network events, for executing the performance bottleneck recognition method of embodiment 1, as shown in fig. 2, comprising: the system comprises a network event adding module, a trigger timer setting module, a network event processing module, a performance bottleneck judging module, a performance bottleneck information acquiring module and a trigger timer deleting module.
The network event adding module is used for collecting the network events to be processed and adding a certain number of the network events to be processed into the network event list. The performance bottleneck recognition system of the embodiment further includes: the system comprises a network event number determining module, a network event adding page number determining module and a network event list forming module. The network event number determining module is used for determining the number of network events to be processed in one cycle. And the network event adding page quantity determining module is used for determining the network event adding page quantity according to the network event quantity to be processed. The network event list forming module is used for combining the network event adding pages with the corresponding number according to the number of the network event adding pages to form a network event list, and each network event adding page is used for adding a network event to be processed.
The trigger timer setting module is used for setting a trigger and a timer, and the timer is provided with a target timing value. The trigger timer setting module specifically comprises: a trigger registration unit and a timer registration unit. The trigger registration unit is used for registering a trigger, the trigger is started when all the network events to be processed in the network event list are completed, and the timer is deleted after the trigger is started. The timer registration unit is used for registering a timer, the timer starts to count after being started, and an alarm signal is generated when the timer counts up to a target calibration value.
The network event processing module is used for starting a timer and sequentially processing network events to be processed in the network event list.
The performance bottleneck judging module is used for judging whether the performance bottleneck occurs or not through the trigger and the timer: if the network event to be processed is completed in the target timing value, judging that the network event has no performance bottleneck; and if the network event to be processed exceeds the target timing value, judging that the network event has performance bottleneck. The performance bottleneck judging module specifically comprises: a timer state scanning unit, a first signal forming unit and a second signal forming unit. The timer state scanning unit is used for scanning the state of the timer in real time: if the scanning timer is deleted, the network event to be processed is completely completed in a target timing value; if the timer is scanned to generate an alarm signal, the network event to be processed is not completely finished beyond the target calibration value. The first signal forming unit is used for forming a first signal when the network event to be processed is completely completed within a target timing value. The second signal forming unit is used for forming a second signal when the network event to be processed exceeds the target timing value and is not completely completed.
And the performance bottleneck information acquisition module is used for acquiring and storing current process stack information and stack variable information when the performance bottleneck of the network event is judged.
The trigger timer deleting module is used for deleting the trigger and the timer after all the network events to be processed in the network event list are completed. The trigger timer deleting module specifically comprises: a first deletion unit and a second deletion unit. The first deleting unit is used for deleting the trigger after the first signal is acquired. The second deleting unit is used for waiting for acquiring a third signal after acquiring the second signal, and deleting the trigger and the timer after acquiring the third signal; and the third signal is formed after the network event to be processed exceeds the target timing value.
According to the system for identifying the performance bottlenecks of the network events, a proper number of network events to be processed can be processed one cycle by one cycle, whether the performance bottlenecks occur in the network events to be processed in the current cycle can be judged through the trigger and the timer, when the performance bottlenecks occur, related process stack information and stack variable information can be stored immediately, and operation and maintenance personnel can be informed immediately, so that the operation and maintenance personnel can take corresponding solving measures in time, the problem that the user satisfaction is reduced due to the fact that the network events cannot be processed in time is avoided as much as possible, the specific situation of each performance bottleneck can be obtained almost effectively, and then the problem of each performance bottleneck can be analyzed and solved, and the subsequent performance bottlenecks caused by the same reasons are avoided.
While certain embodiments of the present description have been shown in the accompanying drawings, it is to be understood that the present description may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided to provide a more thorough and complete understanding of the present description. It should be understood that the drawings and examples of the present specification are for illustrative purposes only and are not intended to limit the scope of the present specification.

Claims (2)

1. A method for identifying performance bottlenecks of network events comprises the following steps:
collecting network events to be processed, and adding a certain number of network events to be processed into a network event list;
setting a trigger and a timer, wherein the timer is provided with a target timing value;
the setting trigger and timer specifically include:
registering a trigger, wherein the trigger is started when all network events to be processed in a network event list are completed, and the timer is deleted after the trigger is started;
registering a timer, starting to count after starting the timer, and generating an alarm signal when the timer counts up to a target calibration value;
starting a timer and sequentially processing network events to be processed in a network event list;
determining whether a performance bottleneck occurs through a trigger and a timer: if the network event to be processed is completed in the target timing value, judging that the network event has no performance bottleneck; if the network event to be processed exceeds the target timing value, not all the network event to be processed is finished, judging that the network event has a performance bottleneck;
the determining whether the performance bottleneck occurs specifically includes:
state of the real-time scan timer: if the scanning timer is deleted, the network event to be processed is completely completed within a target timing value, and a first signal is formed; if the timer is scanned to generate an alarm signal, the network event to be processed is not completely finished beyond the target calibration value, and a second signal is formed;
when judging that the network event has performance bottleneck, acquiring and storing current process stack information and stack variable information;
after all the network events to be processed in the network event list are completed, deleting the trigger and the timer;
the delete trigger and timer include:
after the first signal is acquired, deleting the trigger;
after the second signal is acquired, waiting for acquiring a third signal, and deleting the trigger and the timer after the third signal is acquired; the third signal is formed after the network event to be processed exceeds the target timing value;
before collecting the pending network events, further comprising the steps of:
determining the number of network events to be processed for one cycle;
determining the number of network event adding pages according to the number of the network events to be processed;
and combining the corresponding number of network event adding pages according to the number of the network event adding pages to form a network event list, wherein each network event adding page is used for adding a network event to be processed.
2. A performance bottleneck recognition system for network events, comprising:
the network event adding module is used for collecting network events to be processed and adding a certain number of network events to be processed into the network event list;
the trigger timer setting module is used for setting a trigger and a timer, and the timer is provided with a target timing value;
the trigger timer setting module specifically comprises:
the trigger registration unit is used for registering a trigger, the trigger is started when all network events to be processed in the network event list are completed, and the timer is deleted after the trigger is started;
the timer registration unit is used for registering a timer, starting to count after the timer is started, and generating an alarm signal when the timer counts up to a target calibration value;
the network event processing module is used for starting a timer and sequentially processing network events to be processed in the network event list;
the performance bottleneck judging module is used for judging whether the performance bottleneck appears through a trigger and a timer: if the network event to be processed is completed in the target timing value, judging that the network event has no performance bottleneck; if the network event to be processed exceeds the target timing value, not all the network event to be processed is finished, judging that the network event has a performance bottleneck;
the performance bottleneck judging module specifically comprises:
a timer state scanning unit, configured to scan the state of the timer in real time: if the scanning timer is deleted, the network event to be processed is completely completed in a target timing value; if the timer is scanned to generate an alarm signal, the network event to be processed is not completely finished beyond the target calibration value;
a first signal forming unit, configured to form a first signal when the network event to be processed is all completed within a target timing value;
a second signal forming unit, configured to form a second signal when the network event to be processed exceeds a target value and is not completely completed;
the performance bottleneck information acquisition module is used for acquiring and storing current process stack information and stack variable information when judging that a network event has a performance bottleneck;
the trigger timer deleting module is used for deleting the trigger and the timer after all the network events to be processed in the network event list are completed;
the trigger timer deleting module specifically comprises:
the first deleting unit is used for deleting the trigger after the first signal is acquired;
the second deleting unit is used for waiting for acquiring a third signal after acquiring the second signal, and deleting the trigger and the timer after acquiring the third signal; the third signal is formed after the network event to be processed exceeds the target timing value;
the network event number determining module is used for determining the number of network events to be processed in one cycle;
the network event adding page number determining module is used for determining the number of the network event adding pages according to the number of the network events to be processed;
and the network event list forming module is used for combining the network event adding pages with the corresponding number according to the network event adding page number to form a network event list, and each network event adding page is used for adding a network event to be processed.
CN202310620214.8A 2023-05-30 2023-05-30 Performance bottleneck identification method and system for network event Active CN116346918B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310620214.8A CN116346918B (en) 2023-05-30 2023-05-30 Performance bottleneck identification method and system for network event

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310620214.8A CN116346918B (en) 2023-05-30 2023-05-30 Performance bottleneck identification method and system for network event

Publications (2)

Publication Number Publication Date
CN116346918A CN116346918A (en) 2023-06-27
CN116346918B true CN116346918B (en) 2023-10-10

Family

ID=86879104

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310620214.8A Active CN116346918B (en) 2023-05-30 2023-05-30 Performance bottleneck identification method and system for network event

Country Status (1)

Country Link
CN (1) CN116346918B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7797598B1 (en) * 2006-11-14 2010-09-14 Xilinx, Inc. Dynamic timer for testbench interface synchronization
CN103092682A (en) * 2011-10-28 2013-05-08 浙江大华技术股份有限公司 Asynchronous network application program processing method
CN103942104A (en) * 2014-04-23 2014-07-23 北京金山网络科技有限公司 Task managing method and device
CN111813552A (en) * 2020-07-16 2020-10-23 济南浪潮数据技术有限公司 Scheduling execution method, device and medium based on multi-thread task
CN114416324A (en) * 2022-02-15 2022-04-29 康键信息技术（深圳）有限公司 Task triggering method and device, computer equipment and storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5390134B2 (en) * 2008-07-18 2014-01-15 インターナショナル・ビジネス・マシーンズ・コーポレーション Information management server, information processing system, communication method, and program

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7797598B1 (en) * 2006-11-14 2010-09-14 Xilinx, Inc. Dynamic timer for testbench interface synchronization
CN103092682A (en) * 2011-10-28 2013-05-08 浙江大华技术股份有限公司 Asynchronous network application program processing method
CN103942104A (en) * 2014-04-23 2014-07-23 北京金山网络科技有限公司 Task managing method and device
CN111813552A (en) * 2020-07-16 2020-10-23 济南浪潮数据技术有限公司 Scheduling execution method, device and medium based on multi-thread task
CN114416324A (en) * 2022-02-15 2022-04-29 康键信息技术（深圳）有限公司 Task triggering method and device, computer equipment and storage medium

Also Published As

Publication number Publication date
CN116346918A (en) 2023-06-27

Similar Documents

Publication Publication Date Title
CN110569214B (en) Index construction method and device for log file and electronic equipment
CN109656999B (en) Method, device, storage medium and apparatus for synchronizing large data volume data
CN110719332B (en) Data transmission method, device, system, computer equipment and storage medium
CN112434039A (en) Data storage method, device, storage medium and electronic device
CN107590490A (en) Unanimous vote face information acquisition method, device and the computer-readable recording medium of invoice
CN116346918B (en) Performance bottleneck identification method and system for network event
CN111078688B (en) Method for micro-service health check and storage medium
CN111124791A (en) System testing method and device
CN114328122A (en) IO full life cycle time delay monitoring method and related device
CN114281819A (en) Data query method, device, equipment and storage medium
CN110543509B (en) Monitoring system, method and device for user access data and electronic equipment
CN109189813B (en) Data sharing method and device
CN111242595A (en) Information exchange method and device between ERP platform and payment platform
CN115599652A (en) Method, device, equipment and storage medium for testing system performance parameters
CN113641455B (en) Message conversion method and platform
CN111800409B (en) Interface attack detection method and device
CN114595146A (en) AB test method, device, system, electronic equipment and medium
CN112131180B (en) Data reporting method, device and storage medium
CN115292335A (en) Transaction processing method and device and electronic equipment
CN112764988A (en) Data segmentation acquisition method and device
CN113704652B (en) Data processing method, device, transfer station and data processing system
CN111984844A (en) Method and system for automatically supplementing graph based on big data
CN113326511B (en) File repair method, system, equipment and medium
CN112069359B (en) Method for dynamically filtering abnormal data of snapshot object comparison result
CN114979100B (en) Cloud resource inspection method and related device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant