CN107819642B - Self-adaptive heartbeat method and system based on distribution - Google Patents

Self-adaptive heartbeat method and system based on distribution Download PDF

Info

Publication number
CN107819642B
CN107819642B CN201710832684.5A CN201710832684A CN107819642B CN 107819642 B CN107819642 B CN 107819642B CN 201710832684 A CN201710832684 A CN 201710832684A CN 107819642 B CN107819642 B CN 107819642B
Authority
CN
China
Prior art keywords
detection
adaptive
period
self
heartbeat
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
CN201710832684.5A
Other languages
Chinese (zh)
Other versions
CN107819642A (en
Inventor
施亚虎
石海龙
崔莉
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.)
Institute of Computing Technology of CAS
Original Assignee
Institute of Computing Technology of CAS
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 Institute of Computing Technology of CAS filed Critical Institute of Computing Technology of CAS
Priority to CN201710832684.5A priority Critical patent/CN107819642B/en
Publication of CN107819642A publication Critical patent/CN107819642A/en
Application granted granted Critical
Publication of CN107819642B publication Critical patent/CN107819642B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route
    • H04L43/103—Active monitoring, e.g. heartbeat, ping or trace-route with adaptive polling, i.e. dynamically adapting the polling rate
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/12—Network monitoring probes
    • 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
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • 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
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • 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
    • H04L67/14—Session management
    • H04L67/143—Termination or inactivation of sessions, e.g. event-controlled end of session
    • H04L67/145—Termination or inactivation of sessions, e.g. event-controlled end of session avoiding end of session, e.g. keep-alive, heartbeats, resumption message or wake-up for inactive or interrupted session
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/09—Mapping addresses
    • H04L61/25—Mapping addresses of the same type
    • H04L61/2503—Translation of Internet protocol [IP] addresses
    • H04L61/2521—Translation architectures other than single NAT servers

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Cardiology (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Small-Scale Networks (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention relates to a distributed-based adaptive heartbeat method for maintaining network connection of Internet of things equipment, which comprises the following steps: the method comprises the steps that the Internet of things equipment establishes and maintains network connection with an Internet of things cloud platform through a local area network, and self-adaptive parameters are initialized; the method comprises the steps that the equipment of the Internet of things sends adaptive parameter request broadcast to a local area network, and updates adaptive parameters and adaptive heartbeat cycles according to adaptive parameter sharing broadcast responded by other equipment in the local area network; and determining a plurality of detection values to be detected through an exponential growth mode or a linear growth mode, dividing the detection values into a plurality of periodic detection tasks, and allocating the periodic detection tasks to a plurality of Internet of things devices in the local area network for detection. The invention solves the problems of multiple detection times and long time consumption in the process of acquiring the self-adaptive heartbeat cycle by the Internet of things equipment, and provides an efficient detection termination strategy to balance detection errors and detection efficiency.

Description

Self-adaptive heartbeat method and system based on distribution
Technical Field
The invention provides a distributed-based self-adaptive heartbeat method aiming at the problem of maintaining network connection of Internet of things equipment, and an optimal heartbeat period is obtained through cooperative cooperation between the Internet of things equipment in the same local area network.
Background
With the gradual maturity and practicability of the leveling access protocol of the internet of things (such as CoAP, MQTT, LwM2M, EBHTTP, etc.), accessing a device to a cloud platform (such as xivery, Waston IoT, OneNET, YeeLink, etc.) to access the device in real time becomes a mainstream architecture.
Due to limitations of existing internet infrastructure (e.g., limited IP addresses and port resources), most of internet of things devices are usually located in an enterprise local area network or a home local area network, and there is no independent public network IP, and a network connection is established to access the internet through port mapping by using a Network Address Translation (NAT), but based on security considerations, an enterprise network or a cell network is often provided with firewalls, which periodically remove unusual port mapping information, and when an internet of things device does not use the port mapping for communication with the outside world within a NAT aging time period (agingT), the port mapping information will be invalid, and the outside world will not be able to communicate with the device by using the mapping, that is, the network connection is disconnected.
Therefore, in order to ensure bidirectional real-time access between the cloud platform of the internet of things and the equipment, the heartbeat packet needs to be sent between the equipment and the platform at regular time so as to maintain the network connection of the equipment of the internet of things. Most existing internet of things systems (such as xivery, Waston IoT, OneNET, YeeLink, etc.) adopt a uniform and fixed heartbeat cycle (such as using TCP long connection and adopting a fixed-cycle heartbeat mechanism to maintain long connection), and in order to ensure adaptability to different network environments, the heartbeat cycle is generally set to a shorter time to ensure that the network connection of equipment does not fail. The system of the Internet of things adopting the heartbeat mode with uniform and fixed heartbeat cycle can provide stable service of the Internet of things at the initial development stage of the Internet of things with less equipment. With the rapid growth of Internet of things devices, the manner of maintaining network connection occupies a large amount of cloud platform resources (server memory, CPU time, Internet bandwidth, etc.), thereby affecting other service functions of the cloud platform.
For the above problems, some research achievements are currently made at home and abroad, for example, in "research and application of MQTT communication protocol based on adaptive heartbeat mechanism" (university of south china 2015), by researching heartbeat strategy used by MQTT, an adaptive heartbeat mechanism is proposed, which uses dichotomy to quickly find an optimal heartbeat value, can detect NAT aging time period (agingT) of network environment where internet of things equipment is currently located, and dynamically adjusts heartbeat period of equipment according to the period to maintain network connection, so that cost of heartbeat of equipment is minimum, however, because a fixed threshold is adopted as a detection termination condition in a detection process, when NAT aging time period (agingT) is large, detection of network detection is performedThe number of times of measurement is large, the consumed time is long, and because the connection used for equipment access and periodic detection is the same, if the detection result is failure, the connection is correspondingly failure, and the platform cannot use the connection to access the equipment of the internet of things. Another problem is that when using the dichotomy to determine the detection values, the detection values at the beginning of the detection are
Figure BDA0001409071220000021
Compared with the initial heartbeat value MinT, the detection value is larger, the heartbeat can be updated in a longer time, the number of heartbeat packets sent in the period is larger, and the heartbeat overhead is larger.
In the adaptive heartbeat Method proposed in A Cost-Effective Method to Keep Availability of Man Cloud-Connected Devices (Ajitomi D, Kawazoe H, Minami K, et al. International conference on Cloud computing. IEEE,2015:1-8.), an extra connection is established for NAT aging time period (agingT) detection, so the connection for device access can be continuously Effective without affecting the device access. However, the detection value for network detection in this method is fixed, although the detection times can be controlled within the fixed times and the detection time can be controlled, the detected NAT aging time period (agingT) error is too large, and thus the obtained adaptive heartbeat period is often not optimal.
Related patents such as CN105978757A, CN103685241A, CN103209089A, CN104144159A, CN102843250A, CN103684815A, and CN106452973A all perform periodic detection independently through devices or terminals during the execution of the disclosed periodic detection of adaptive heartbeats, the devices or terminals need to perform multiple detections, the number of communications is large, the overhead is large, the time consumption is long, and the detection efficiency and the error are not optimized correspondingly. Particularly, related patents CN105978757A, CN103685241A, and CN103209089A store the NAT expiration time period detected by the device or the terminal in the server, and when the device or the terminal having the same IP address needs to acquire the period next time, the period is directly acquired from the server without further detection, so that the overhead can be reduced; however, the above patent does not consider the case of multi-level routing under the same gateway: the public network IP of the final NAT mapping of the two local area networks is the same, but because the intermediate layer routers passing from the device or the terminal to the gateway are different, and the NAT aging time periods of different routers may be different, the NAT aging time periods of the device or the terminal may be different, so the NAT aging time period method obtained by the server disclosed in the above patent is not suitable for the case where different intermediate layer routers exist.
The existing self-adaptive heartbeat method needs each internet of things device to independently perform periodic detection to obtain a self-adaptive heartbeat period, and each device needs to perform multiple detections in the period, so that the communication times are more, and a large amount of network bandwidth resources and platform resources (server memory, CPU time and the like) are occupied; moreover, since the detection of different detection values needs to be performed sequentially, multiple detections will consume a lot of time; meanwhile, the existing period detection method has no efficient detection termination strategy, the error between the self-adaptive heartbeat period obtained by the detection method adopting the fixed detection value and the optimal value is large, the detection method adopting the fixed detection termination threshold value is likely to have excessive detection times and low efficiency, wherein the detection termination threshold value is the minimum difference value of two adjacent detection values, and the period detection is terminated when the detection termination threshold value is smaller than the threshold value.
In summary, the conventional method has the problems of high cost, long time consumption, and incapability of effectively controlling errors and efficiency, and therefore, it is necessary to provide an adaptive method capable of solving the above problems.
Disclosure of Invention
Aiming at the problems, the invention provides a distributed self-adaptive heartbeat method to balance detection errors and detection efficiency, and aims to solve the problems of more detection times and long time consumption in the process of acquiring a self-adaptive heartbeat cycle by equipment of the Internet of things.
The optimal adaptive heartbeat period is related to the NAT aging time period (agingT), network delay and other factors, is determined by a network where the equipment is located, and is not influenced by the Internet of things equipment, so that the optimal adaptive heartbeat periods of the Internet of things equipment in the same local area network are the same. Based on the characteristics, the invention provides a distributed self-adaptive heartbeat method, which comprises the following steps:
1) and a parameter synchronization mechanism: the Internet of things equipment in the same local area network realizes the sharing of self-adaptive parameters through broadcasting, and the equipment adjusts the self-adaptive heartbeat period T according to the self-adaptive parameters and executes distributed period detection.
The adaptive parameters (Params) include: the method comprises the following steps of a period detection lower limit (lowerT), a period detection upper limit (upperT), the number (m) of devices actually participating in distributed period detection, a distributed adaptive process (AS) and a countdown time (CD) for completing the execution of the last detection task in a round of detection in a fast convergence stage.
The distributed adaptive procedure (AS) is divided into: four phases of not-started (NS), fast-updated (FU), fast-converged (FC) and Completed (CP), wherein the FU and FC phases require distributed periodic probing.
2) The distributed periodic detection method comprises the following steps: fig. 1 is a schematic diagram of device independent period detection and distributed period detection in a Local Area Network (LAN), where each device in an existing adaptive heartbeat method needs to independently complete a whole period detection process (denoted by TASK), the whole process needs to detect multiple detection values (t1.. Tk), and different detections need to be performed sequentially, which consumes a long time and is costly. The NAT aging time periods (agingT) of all the devices in the same network are the same, therefore, the periodic detection results of all the devices are the same, the whole detection process is divided into a plurality of subtasks by distributed periodic detection and is distributed to a plurality of devices to be executed, each subtask has 0 to a plurality of detection values, therefore, different detection values can be detected simultaneously, the detection results are shared, the periodic detection speed can be increased, the total detection times can be reduced, and the expenses of an object end, a network end and a cloud end caused by the fact that each device sends heartbeat packets are reduced.
The periodic detection process is as follows: the method comprises the steps that a detection value testT is detected, equipment firstly adopts a port to send a mapping establishment request (initP) to a platform, NAT mapping to the port is established, the testT does not use the port for communication any more, after testT time, a mapping test request (testP) is sent from the platform to the port, if the equipment can receive the request, the testT is smaller than a failure period, otherwise, the testT is larger than the failure period, and the approximate value of the failure period can be converged finally through detection on different detection values.
3) The "fast update" mechanism: supposing that m devices actually participate in the distributed periodic detection, supposing that the detection value of the detection task executed by the device Dj at the ith time is t [ i [ [ i ]][j]This detection belongs to the ith round of detection. In the initial stage of distributed period detection, the detection lower limit lowerT and the heartbeat period T of the device are both the minimum heartbeat period MinT, and if the detection value is testT, the device needs to update the heartbeat period of the device after the testT time, during which the heartbeat period needs to be sent
Figure BDA0001409071220000041
If the testT is large, each device sends multiple heartbeat packets in a small heartbeat cycle, which not only causes large overhead, but also causes a slow update speed of the heartbeat cycle of the device.
By adopting a 'fast update' mechanism, testT increases according to the exponential of 2 to distribute the detection tasks, and the initial value is twice of the detection lower limit lowerT (namely MinT), namely:
t[1][1]＝2×lowerT,
t[1][2]＝2×t[1][1],
t[i][j]＝2×t[i][j-1],
(where i > 0, j > 0, ti [ i ] [0] is ti [ i-1] [ m ])
If the detection result of the device Dj to the T [ i ] [ j ] is valid, updating the lower T and the T, and allocating a detection task (if any) with a detection value T [ i +1] [ j ] to the Dj; and if the detection result of Dj on t [ i ] [ j ] is invalid, ending all unfinished detection tasks in advance, ending the execution of the 'fast update' and updating the upperT. Assuming that the real value of the NAT aging time period is agingT, when the execution of the fast update is finished, the total detection times are test _ num, and the number of detection rounds is loop _ num, then:
when agingT is greater than or equal to upperT
Figure BDA0001409071220000051
Figure BDA0001409071220000052
T＝upperT (3)
When agingT < upperT
Figure BDA0001409071220000053
Figure BDA0001409071220000054
Figure BDA0001409071220000055
In the "fast update" execution process, each device has 1 mapping establishment request (initP), so the total number of mapping establishment requests initP _ num is m; there are 1 testP per probe, so the total number of mapping test requests testP _ num is testP _ num.
Fig. 2 shows an implementation of the "fast update" mechanism when m is 3, lowerT is 2, agingT is 59, and upperT is 70. The task matrix t obtained by using a 'fast update' mechanism to perform detection task allocation is as follows:
Figure BDA0001409071220000056
the detection values detected by the 1 st round of the devices D1, D2 and D3 are T [1] [1] 4, T [1] [2] 8 and T [1] [3] 16 respectively, the detection result of the D1 on T [1] [1] is effective, the lowerT and the T are updated to T [1] [1], and the device D1 continues to select the detection value T [2] [1] for execution; similarly, when D2 and D3 finish the detection of T1 < 2 > and T1 < 3 >, updating lowerT and T, and selecting corresponding detection values T2 < 2 > and T2 < 3 >; when D1 completes probing t [2] [1], no probing task has been assigned to D1; when D2 completes the detection of T [2] [2], the detection result is invalid, so D3 ends the detection of T [2] [3] early, D3 no longer sends testP, the execution of "fast update" ends and updates upperT ═ T [2] [2] ═ 64, and T ═ lowerT ═ T [2] [1] ═ 32.
4) The "fast convergence" mechanism: when the failure value testT _ failed is detected for the first time, the distributed periodic detection enters the later stage, T and testT are both large at the moment, one round of detection consumes long time, and if the detection rounds are more, the convergence speed of the distributed periodic detection is low. And a 'fast convergence' mechanism is adopted, the detection interval is minimized in each detection round, the detection rounds are reduced, and the detection is finished quickly. The detection interval at the beginning of one round of detection is (lowerT, upperT), the interval width testSize ═ upperT-lowerT, assuming that agingT obeys uniform distribution on (lowerT, upperT), the time detected by the device Dj is t [, ]][j]The time interval between two preceding and succeeding detections is t][j]-t[][j-1](wherein t 2][0]loweT). Then after the detection of the current round is finished, the width of the converged detection interval is testSizeNew, and the mathematical expectation is thatWhere Δ Tm +1 is upperT,
Figure BDA0001409071220000062
if and only if
Figure BDA0001409071220000063
When E (testSizeNew) takes a minimum value of
Figure BDA0001409071220000064
Based on the above analysis, in the "fast convergence" mechanism, testT grows linearly according to the arithmetic progression rule with the tolerance of
Figure BDA0001409071220000065
Rounding up may result in a tolerance other than 0. By adopting the strategy, E (testSizeNew) can obtain the approximate value of the minimum value in one detection round, a smaller detection interval can be converged when each detection round is finished, the periodic detection process can be finished through fewer detection rounds, the cost is lower, and the consumed time is shorter.
Fig. 3 is an implementation of the "fast convergence" mechanism in fig. 2 after the "fast update" mechanism is executed. In this case, the detection end threshold P is 1, m is 3, T is 32, lowerT is 32, agingT is 59, upperT is 64, testSize is 32. In this case, for the 3 rd round of detection, the task vector obtained by using the "fast convergence" mechanism to perform detection task allocation is t [3] [ 404856 ]. It can be seen that the detection results of D1, D2 and D3 for T [3] [1], T [3] [2] and T [3] [3] are all valid, and when the round of detection is finished, T is 56, lower is 56, upper is 64, testsizeenew is 8, the detection interval is reduced to 1/4, and the number of times the device sends heartbeat packets during the round of detection is small, namely 1. And in the same way, the task vectors detected in the 4 th round and the 5 th round are t [4] [ [ 586062 ] and t [5] [ [59], respectively.
Before the 5 th round of detection starts, T is 58, lowerT is 58, upperT is 60, testSize is 2, and there is only one detection value 59 between lowerT and upperT, so the task vector T [5] [ ] is only one-dimensional, and only a detection task needs to be allocated to D1, and the detection value T [5] [1] ═ 59. Although the agingT is 59, in consideration of factors such as network delay, the detection result of T [5] [1] by D1 should be invalid, so after the 5 th round of detection is finished, T is 58, lowertt is 58, and upperT is 59, and because testSize is 1, the detection end threshold P is reached, and there is no need to perform detection again, the execution of the "fast convergence" mechanism is finished, the convergence is finished, and the section of convergence (lowertv, upperT) is (58, 59), and the detection value agingT' of agingT is 58.
5) The "dynamic threshold" strategy: when the NAT aging time period agingT is smaller, the self-adaptive heartbeat period is smaller after distributed period detection is finished, the cost caused by sending heartbeat packets is larger, and the detection termination threshold value (P) is reduced as much as possible to obtain the maximum self-adaptive heartbeat period so as to reduce the heartbeat cost; when the agingT is large, the self-adaptive heartbeat period is large after the distributed period detection is finished, the cost caused by the heartbeat is small, if the detection termination threshold value (P) is small, the time consumed in the fast convergence process is long, the detection completion time can be greatly shortened by properly increasing the detection termination threshold value (P), and the heart tripping cost cannot be obviously increased.
An exponentially growing "dynamic threshold" may be:
when MinT is less than or equal to agingT and less than or equal to 2MinT, detecting a termination threshold value (P) and taking minP;
when 2MinT is less than or equal to agingT and less than or equal to 4MinT, the detection termination threshold (P) is 2 minP;
when 4MinT is less than or equal to agingT and less than or equal to 8MinT, taking 4minP as the detection termination threshold (P); and so on.
6) The strategy of 'dynamic maximum detection equipment number': in the "fast update" phase, at most
Figure BDA0001409071220000071
Detecting the devices simultaneously; in the "fast convergence" phase, at most
Figure BDA0001409071220000072
The devices can be detected simultaneously. The distributed periodic detection process is carried out in a self-organizing mode, a unified control center is not provided, and a competition mode is adopted in the stages of 'fast update' and 'fast convergence' to distribute detection values.
Specifically, the invention relates to a distributed-based adaptive heartbeat method, which comprises the following steps:
the method comprises the steps of initialization, wherein the first equipment establishes and maintains network connection with an Internet of things cloud platform through a local area network to which the first equipment belongs, and initialization setting is carried out on self-adaptive parameters of the first equipment;
and a period detection step, namely determining a plurality of detection values of the NAT aging time period to be detected through an exponential growth mode or a linear growth mode, dividing the detection values into a plurality of period detection tasks, and allocating the plurality of period detection tasks to a plurality of first devices in the local area network for detection.
The initialization step specifically comprises the following steps:
a network connection step, in which the first device sends a network connection establishment request to the cloud platform, after receiving the request, the cloud platform stores the NAT mapping address of the first device as the source IP address of the request, stores the mapping port of the network connection port of the first device as the source port of the request, and sends an access request to the source IP address and the source port when the cloud platform needs to access the first device;
a parameter initialization step, after network connection is established, initializing and setting the self-adaptive heartbeat cycle and the self-adaptive parameters of the first equipment; the self-adaptive parameters comprise a periodic detection lower limit, a periodic detection upper limit, the number of devices actually participating in the distributed periodic detection and a distributed self-adaptive process, wherein the distributed self-adaptive process comprises an uninitiated stage, a fast updating stage, a fast convergence stage and a completed stage;
a network maintaining step, in which the first device sends a heartbeat packet to the cloud platform to maintain network connection, wherein the sending time interval of the heartbeat packet is the initialized self-adaptive heartbeat period;
and a parameter synchronization step, in which the first device sends a self-adaptive parameter request broadcast to the local area network, a second device in the local area network sends a self-adaptive parameter sharing broadcast to the local area network after receiving the request broadcast, and the first device updates the self-adaptive parameters and the self-adaptive heartbeat cycle according to the received sharing broadcast.
Wherein, the period detection step specifically comprises:
the first equipment sends a mapping establishment request to the cloud platform through a port, establishes NAT mapping to the port, suspends communication, sends a mapping test request to the port from the cloud platform after the testT time, and if the first equipment receives the request, the testT is less than a failure period, and a corresponding detection result is valid; if the first device does not receive the request, the testT is greater than the failure period, and the corresponding detection result fails; wherein testT is the detection value.
The distributed adaptive heartbeat method specifically comprises the following steps of:
a step of allocating a fast update task, when the distributed adaptive process is in an un-started stage or a fast update stage, the distributed adaptive process is executedSetting the state as a fast update stage, and starting fast update detection; setting the detection value testT to 2nXlowert, assigning the detection value to m first devices of the local area network for periodic detection; if the detection result is valid, updating the detection lower limit lowerT of the period to be testT, updating the self-adaptive heartbeat period and sending self-adaptive parameter updating broadcast to the local area network, if lowerT is MaxT, updating the distributed self-adaptive process to be a finished stage, terminating period detection, otherwise, re-performing the rapid updating task allocation step; if the detection result is invalid, updating the periodic detection upper limit uptert to testT, sending adaptive parameter updating broadcast to the local area network, updating the distributed adaptive process to be a fast convergence stage, and periodically detecting to enter a fast convergence task allocation step;
wherein the content of the first and second substances,
Figure BDA0001409071220000081
m is less than or equal to MaxM 1; m and n are positive integers, m is the number of the actual devices participating in the distributed period detection, MinT is the minimum heartbeat period, MaxT is the maximum heartbeat period, MaxM1 is the maximum number of the devices participating in the distributed period detection in the fast updating stage, lowerT is the period detection lower limit, and the lowerT initial value is MinT.
The distributed adaptive heartbeat method specifically comprises the following steps of:
a fast convergence task allocation step, when the distributed self-adaptive process is in a fast convergence stage, if t is less than or equal to P, the distributed self-adaptive process is updated to be in a finished stage, and distributed periodic detection is terminated; if T is greater than P, starting fast convergence detection, setting the detection value testT as lowerT + nxDELTA T, and distributing the detection value to m first devices in the local area network for periodic detection; if the detection result is valid, updating the periodic detection lower limit lowerT to testT, sending adaptive parameter updating broadcast to the local area network, and performing the fast convergence task allocation step again after waiting for t time; if the detection result is invalid, updating the periodic detection upper limit upperT to testT, sending adaptive parameter updating broadcast to the local area network, and repeating the fast convergence task allocation step;
wherein, when 2n-1×MinT≤agingT≤2nX MinT, P is 2n-1×minP；t＝upperT-lowerT，
Figure BDA0001409071220000091
m≤MaxM2，m and n are positive integers, m is the number of the devices actually participating in the distributed period detection, minP is a minimum detection termination threshold value, P is a detection termination threshold value, upperT is the period detection upper limit, lowerT is the period detection lower limit, agingT is the NAT aging time period, and MaxM2 is the maximum number of the devices participating in the period detection in the fast convergence stage.
The invention also relates to a distributed self-adaptive heartbeat system, which comprises:
the initialization module is used for establishing and maintaining network connection with the Internet of things cloud platform through the local area network of the first device and carrying out initialization setting on self-adaptive parameters of the first device;
and the period detection module is used for determining a plurality of detection values of the NAT aging time period to be detected through an exponential growth mode or a linear growth mode, dividing the detection values into a plurality of period detection tasks and allocating the plurality of period detection tasks to a plurality of first devices in the local area network for detection.
The initialization module specifically comprises:
a network connection module, configured to send a network connection establishment request to the cloud platform by the first device, where after receiving the request, the cloud platform stores the NAT mapping address of the first device as a source IP address of the request, stores a mapping port of a network connection port of the first device as a source port of the request, and sends an access request to the source IP address and the source port when the cloud platform needs to access the first device;
the parameter initialization module is used for initializing and setting the self-adaptive heartbeat cycle and the self-adaptive parameters of the first equipment after the network connection is established; the self-adaptive parameters comprise a periodic detection lower limit, a periodic detection upper limit, the number of devices actually participating in the distributed periodic detection and a distributed self-adaptive process, wherein the distributed self-adaptive process comprises an uninitiated stage, a fast updating stage, a fast convergence stage and a completed stage;
a network maintaining module, configured to send a heartbeat packet to the cloud platform by the first device to maintain network connection, where a sending time interval of the heartbeat packet is the initialized adaptive heartbeat period;
and the parameter synchronization module is used for sending a self-adaptive parameter request broadcast to the local area network by the first equipment, sending a self-adaptive parameter sharing broadcast to the local area network after receiving the request broadcast by the second equipment in the local area network, and updating the self-adaptive parameter and the self-adaptive heartbeat cycle by the first equipment according to the received sharing broadcast.
Wherein, the period detection module specifically comprises:
the detection module is used for sending a mapping establishment request to the cloud platform through a port by the first equipment, establishing NAT mapping to the port, suspending communication, sending a mapping test request to the port from the cloud platform after the testT time, and if the first equipment receives the request, the testT is less than a failure period and the corresponding detection result is valid; if the first device does not receive the request, the testT is greater than the failure period, and the corresponding detection result fails; wherein testT is the detection value.
The first task allocation module is used for determining a plurality of detection values of the NAT aging time period to be detected in an exponential growth mode, performing detection task allocation and calling the detection module to perform self-adaptive heartbeat period detection.
And the second task allocation module is used for determining a plurality of detection values of the NAT aging time period to be detected in a linear growth mode, allocating detection tasks and calling the detection module to perform self-adaptive heartbeat period detection.
According to the invention, by utilizing the characteristic that the optimal heartbeat cycles of the Internet of things devices in the same local area network are the same, the complex and time-consuming periodic detection process is distributed to a plurality of Internet of things devices for execution, and the real-time sharing of the result is realized through parameter synchronization, so that the periodic detection communication times and time consumption are reduced; the detection termination threshold value of the period detection is dynamically adjusted according to the size range of the optimal heartbeat period, so that the detection error and the detection efficiency can be effectively controlled; and by controlling the maximum number of detection devices actually participating in the distributed periodic detection, the broadcast communication traffic caused by competition is effectively reduced, and the broadcast storm is avoided.
Drawings
Fig. 1 shows an independent period detection diagram (left) and a distributed period detection diagram (right).
FIG. 2 is a diagram of an example implementation of the "fast update" mechanism.
FIG. 3 is a diagram of an example implementation of the "fast convergence" mechanism.
Fig. 4 is a flow chart of an embodiment.
Fig. 5 contention listening flow chart.
FIG. 6 is a diagram illustrating an exemplary probing task.
Wherein the reference numerals are:
step S1 is a network connection setup,
step S2 is to initialize the parameters,
step S3 is for network connection maintenance,
step S4 is the parameter synchronization,
step S5 is initiated for distributed cycle detection,
step S6 is initiated for "fast update", including: in the step S601/step S602,
step S7 is initiated for "fast convergence", and includes: in the step S701/step S702,
step S8 is the distributed cycle detection termination.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more clearly understood, the following describes the distributed adaptive heartbeat method in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The present invention adopts a broadcast mode to perform communication between devices in a local area network, to realize self-organization between the devices, and further to realize a distributed-based adaptive heartbeat method, so that a communication protocol capable of supporting broadcast communication, such as an IP protocol of a network layer, a UDP protocol of a transmission layer, and a CoAP protocol of an application layer, needs to be adopted, a specific implementation mode for implementing the present invention by adopting the CoAP protocol is given below, and fig. 4 is a flow chart of the specific implementation mode of the present invention, which includes the following steps:
step S1: and establishing network connection. The device adopts a network connection port (CoAP default port 5683) to send a network connection establishment request to the platform, and the platform stores a source IP address and a source port of the request when receiving the request, wherein the source IP address and the source port are respectively an NAT mapping address of the device and a mapping port of the network connection port; when the platform needs to access the device, an access request is sent to the source IP address and source port.
Step S2: and initializing parameters. After the network connection is successfully established, initial values of the lowerT and the adaptive heartbeat period T are set to be a minimum heartbeat period (MinT), an upper T is set to be a maximum heartbeat period (MaxT), m is set to be 0, an AS is set to be NS, and a CD is set to be 0.
Step S3: the network connection is maintained. After the parameters are initialized, the equipment adopts a network connection port to send heartbeat packets to the platform to maintain network connection, and the sending time interval of the heartbeat packets is a self-adaptive heartbeat period T.
Step S4: and synchronizing the parameters. After initializing the parameters, the device sends a parameter synchronization request broadcast, updates the adaptive parameters (Params) of the device according to the received response, and updates the adaptive heartbeat period T according to the period detection lower limit (lowerT). The device receiving the broadcast randomly backs off for a certain time (which should be less than the preset maximum back-off time maxRT), if other devices respond to the request in the period, the request is not responded, otherwise, the parameter synchronization response broadcast is sent.
Step S5: distributed periodic probing is initiated. If the distributed adaptive process (AS) is Not Started (NS) or Fast Update (FU), go to step S6; if the Fast Convergence (FC) is detected, the process proceeds to step S7; if the result is Complete (CP), the process proceeds to step S8.
Step S6: the "fast update" starts: set AS FU, start contention listening and proceed to step S601.
Step S601: the "fast update" race. Entering into a 'fast update' stage, the equipment firstly obtains distributed numbers (SIDs) through competition, and the 'fast update' stage is most needed
Figure BDA0001409071220000121
The devices perform the detection. m is the known total number of devices participating in probing, and if m is MaxM, the devices do not compete any more, set SID to 0 and go to step S602; otherwise, the SID contended by the device is m +1, the device sends a contention broadcast, the broadcast carries the SID to be contended, if the contention failure response broadcast is not received, the contention success is indicated, the SID is m +1, m is increased by 1, and the step S602 is entered; otherwise, the contention fails, the adaptive parameters are obtained from the contention failure response broadcast and updated, and step S601 is executed again.
Step S602: "fast update" probing. If the SID is 0, the device only monitors the adaptive parameter updating broadcast of other devices; otherwise: the detection values of the detection tasks in the 'fast update' stage are 2MinT, 4MinT, 8MinT, and so on,
Figure BDA0001409071220000122
if the number of the detection tasks is already finished in the 'fast update' stage, if the number is 0, the detection tasks are not finished yet, and at the moment, the equipment selects the detection task with the SID (detection value is testT) to execute according to the SID; if not, it indicates that the device has completed the detection task, and since the detection task is completed in sequence from small to large according to the detection value, if the device can execute the detection task, the q + m-th detection task (the detection value is testT) is selected to be executed. If the detection result is valid, updating lowerT to testT and updating the adaptive heartbeat period T, and simultaneously sending adaptive parameter updating broadcast, if lowerT is equal to MaxT, entering step S8, otherwise, entering step S602 again; and if the detection result is invalid, updating the upperT to the testT, simultaneously sending adaptive parameter updating broadcast, and entering the step S7. Receiving adaptive parametersUpdating the self-adaptive parameters and the heartbeat cycle by the broadcast updating equipment according to the broadcast content, and if the detection result is invalid, entering the step S7; if the updated lowerT is equal to MaxT, the process proceeds to step S8.
Step S7: the "fast convergence" starts: if (upperT-lowerT) ≦ P, where P is the detection termination threshold corresponding to the "fast convergence" stage, then enter step S8; otherwise: AS is set to FC, SID and m are set to 0, contention snooping is started and the process proceeds to step S701.
Step S701: "fast convergence" competition. The device first obtains the distributed number (SID) through competition, and the fast convergence stage is needed at most
Figure BDA0001409071220000131
The devices perform the detection. m is the known total number of devices participating in probing, and if m is MaxM, the devices do not compete any more, set SID to 0 and go to step S702; otherwise, the SID contended by the device is m +1, the device sends a contention broadcast, the broadcast carries the SID to be contended, if the contention failure response broadcast is not received, the contention success is indicated, the SID is m +1, m is increased by 1, and the step S702 is entered; otherwise, the contention fails, the adaptive parameters are obtained from the contention failure response broadcast and updated, and step S7 is executed again.
Step S702: "fast convergence" detection: if the SID is 0, the device only monitors the adaptive parameter updating broadcast of other devices; if SID>0, then: the detection value of the detection task in the fast convergence stage is lowerT + Δ T, lowerT +2 Δ T, lowerT +3 Δ T, and so on, wherein
Figure BDA0001409071220000132
At this time, the device selects the second SID probe task (with a probe value of testT) to execute according to the SID. If the detection result is valid, updating lowerT to testT and updating the adaptive heartbeat period T, simultaneously sending adaptive parameter updating broadcast, and re-entering the step S7 after waiting for (upperT-lowerT) time; if the detection result is invalid, update the upperT to testT, and simultaneously send the adaptive parameter update broadcast, and immediately enter step S7. Apparatus for receiving adaptive parameter update broadcast, updating adaptive parameter according to broadcast contentNumber of heart beats, and if the updated upperT is smaller than that before the update, the process immediately proceeds to step S7.
Step S8: distributed periodic probing terminates: and setting the AS to be CP, and setting the SID and m to be 0, and ending the distributed periodic detection.
And (3) contention monitoring: fig. 5 is a flow chart of contention listening. If the competitive SID is the same as the SID of the equipment, the equipment which receives the competitive broadcast immediately sends a competitive failure response broadcast, and the response carries all self-adaptive parameters which are kept by the equipment; if the competitive SID is larger than the MaxM maintained by the equipment, randomly retreating for a certain time (which is smaller than the preset maximum retreat time maxRT), and if other equipment responds to the request in the period, no responding to the request is performed, otherwise, a competition failure response broadcast is sent, and all self-adaptive parameters maintained by the equipment are carried in the response; if the competitive SID is larger than m maintained by the device, setting m as the competitive SID and not responding to the competitive broadcast; otherwise, the contention broadcast is ignored.
And (3) executing a detection task: fig. 6 is a diagram illustrating an example of the execution of the device Dj detection task. The "/adpt" resources of the platform are used to assist the device in performing the probing tasks. The network connection port (5683) of the device is long-lived, requiring the use of another port (5684) as a probe execution port for executing probe tasks.
Device DjThe method comprises the steps of deciding a detection task to be executed according to a distributed periodic detection task allocation mechanism (a 'fast update' mechanism is adopted in a 'fast update' stage), wherein a detection value is testT (60 seconds in a legend), Dj needs to detect the testT, firstly, mapping (initP) is established for a detection execution port, namely, a mapping request (realized by GET) is sent to a platform through the detection execution port, so that an upper network establishes mapping on a detection execution port of a device, the legend is 9084, and the platform responds the mapping port 9084 to the device. After the testT seconds, the device sends a mapping test request (testP, realized by PUT) to the platform through the network connection port, the mapping test request needs to carry a mapping port 9084, the platform forwards the mapping test request to the mapping port and receives a response, the mapping port is proved to be effective, and a mapping test result is respondedIf the equipment is required to be provided, the equipment updates the adaptive parameter lowerT and synchronizes the adaptive parameter when receiving the mapping test response that the Code is VALID, and meanwhile updates the adaptive heartbeat period T; the device again decides on the detection task to be performed and on the increased detection value testT (illustrated as 120 seconds), DjThe probing for a new testT continues. After waiting for 120 seconds, the device performs mapping test on the mapping port 9084 again, because the failure cycle is less than 120 seconds, the mapping port fails, and the platform cannot receive a response to the forwarded mapping test request, the mapping test response with Code NOT _ FOUND is sent to the device, and the device updates the adaptive parameter upperT and sends adaptive parameter update broadcast after receiving the response; the third decision (the "fast convergence" phase using the "fast convergence" mechanism) of the device is to take the probe task to be performed and the testT (illustrated as 100 seconds), D, reducedjAnd re-sending the mapping request to establish mapping (initP) for the detection execution port, continuing to perform mapping test on the new testT, and repeating the steps until the self-adaption process is finished. By detecting a plurality of detection values, the size relationship between each detection value and the failure cycle is determined, the failure cycle can be converged, and the detection of the failure cycle is realized.

Claims (10)

1. A distributed-based adaptive heartbeat method, comprising:
the method comprises the steps of initialization, wherein the first equipment establishes and maintains network connection with an Internet of things cloud platform through a local area network to which the first equipment belongs, and initialization setting is carried out on self-adaptive parameters of the first equipment;
and a period detection step, namely determining a plurality of detection values of the NAT aging time period to be detected through an exponential growth mode and a linear growth mode, dividing the detection values into a plurality of period detection tasks, and allocating the plurality of period detection tasks to a plurality of first devices in the local area network for detection.
2. The distributed based adaptive heartbeat method of claim 1, in which the initialization step specifically includes:
a network connection step, in which the first device sends a network connection establishment request to the cloud platform, after receiving the request, the cloud platform stores the NAT mapping address of the first device as the source IP address of the request, stores the mapping port of the network connection port of the first device as the source port of the request, and sends an access request to the source IP address and the source port when the cloud platform needs to access the first device;
a parameter initialization step, after network connection is established, initializing and setting the self-adaptive heartbeat cycle and the self-adaptive parameters of the first equipment; the self-adaptive parameters comprise a periodic detection lower limit, a periodic detection upper limit, the number of devices actually participating in the distributed periodic detection and a distributed self-adaptive process, wherein the distributed self-adaptive process comprises an uninitiated stage, a fast updating stage, a fast convergence stage and a completed stage;
a network maintaining step, in which the first device sends a heartbeat packet to the cloud platform to maintain network connection, wherein the sending time interval of the heartbeat packet is the initialized self-adaptive heartbeat period;
and a parameter synchronization step, in which the first device sends a self-adaptive parameter request broadcast to the local area network, a second device in the local area network sends a self-adaptive parameter sharing broadcast to the local area network after receiving the request broadcast, and the first device updates the self-adaptive parameters and the self-adaptive heartbeat cycle according to the received sharing broadcast.
3. The distributed based adaptive heartbeat method of claim 1, in which the period detection step specifically includes:
the first equipment sends a mapping establishment request to the cloud platform through a port, establishes NAT mapping to the port, suspends communication, sends a mapping test request to the port from the cloud platform after the testT time, and if the first equipment receives the request, the testT is less than a failure period, and a corresponding detection result is valid; if the first device does not receive the request, the testT is greater than the failure period, and the corresponding detection result fails; wherein testT is the detection value.
4. The distributed-based adaptive heartbeat method of claim 1 or 3, wherein the exponential growth mode in the periodic detection step specifically includes:
a fast update task allocation step, when the distributed self-adaptive process is a non-starting stage or a fast update stage, setting the state of the distributed self-adaptive process as a fast update stage, and starting fast update period detection; setting the detection value testT to 2nXlowert, assigning the detection value to m first devices of the local area network for periodic detection; if the detection result is valid, updating the detection lower limit lowerT of the period to be testT, updating the adaptive heartbeat period to be testT and sending adaptive parameter updating broadcast to the local area network, if lowerT is MaxT, updating the distributed adaptive process to be a finished stage, terminating period detection, otherwise, re-performing the rapid updating task allocation step; if the detection result is invalid, updating the periodic detection upper limit uptert to testT, sending adaptive parameter updating broadcast to the local area network, updating the distributed adaptive process to be a fast convergence stage, and periodically detecting to enter a fast convergence task allocation step;
wherein the content of the first and second substances,
Figure FDA0002277035620000021
m and n are positive integers, m is the number of the first devices actually participating in the distributed period detection, MinT is the minimum heartbeat period, MaxT is the maximum heartbeat period, MaxM1 is the maximum number of the devices participating in the distributed period detection in the fast update stage, lowerT is the period detection lower limit, and the lowerT initial value is MinT.
5. The distributed-based adaptive heartbeat method of claim 4, wherein the linear growth mode in the periodic detection step specifically includes:
a fast convergence task allocation step, when the distributed self-adaptive process is in a fast convergence stage, if t is less than or equal to P, the distributed self-adaptive process is updated to be in a finished stage, and distributed periodic detection is terminated; if T is greater than P, starting fast convergence detection, setting the detection value testT as lowerT + nxDELTA T, and distributing the detection value to m first devices in the local area network for periodic detection; if the detection result is valid, updating the periodic detection lower limit lowerT to testT, sending adaptive parameter updating broadcast to the local area network, and performing the fast convergence task allocation step again after waiting for t time; if the detection result is invalid, updating the periodic detection upper limit upperT to testT, sending adaptive parameter updating broadcast to the local area network, and repeating the fast convergence task allocation step;
wherein, when 2n-1×MinT≤agingT≤2nX MinT, P is 2n-1×minP；t＝upperT-lowerT，
Figure FDA0002277035620000022
minP is a minimum detection termination threshold value, P is a detection termination threshold value, upperT is the period detection upper limit, lowerT is the period detection lower limit, agingT is the NAT aging time period, and MaxM2 is the maximum number of devices participating in period detection in the fast convergence stage.
6. A distributed based adaptive heartbeat system, comprising:
the initialization module is used for establishing and maintaining network connection with the Internet of things cloud platform through the local area network of the first device and carrying out initialization setting on self-adaptive parameters of the first device;
and the period detection module is used for determining a plurality of detection values of the NAT aging time period to be detected through an exponential growth mode and a linear growth mode, dividing the detection values into a plurality of period detection tasks and allocating the plurality of period detection tasks to a plurality of first devices in the local area network for detection.
7. The distributed based adaptive heartbeat system of claim 6, in which the initialization module specifically includes:
a network connection module, configured to send a network connection establishment request to the cloud platform by the first device, where after receiving the request, the cloud platform stores the NAT mapping address of the first device as a source IP address of the request, stores a mapping port of a network connection port of the first device as a source port of the request, and sends an access request to the source IP address and the source port when the cloud platform needs to access the first device;
the parameter initialization module is used for initializing and setting the self-adaptive heartbeat cycle and the self-adaptive parameters of the first equipment after the network connection is established; the self-adaptive parameters comprise a periodic detection lower limit, a periodic detection upper limit, the number of devices actually participating in the distributed periodic detection and a distributed self-adaptive process, wherein the distributed self-adaptive process comprises an uninitiated stage, a fast updating stage, a fast convergence stage and a completed stage;
a network maintaining module, configured to send a heartbeat packet to the cloud platform by the first device to maintain network connection, where a sending time interval of the heartbeat packet is the initialized adaptive heartbeat period;
and the parameter synchronization module is used for sending a self-adaptive parameter request broadcast to the local area network by the first equipment, sending a self-adaptive parameter sharing broadcast to the local area network after receiving the request broadcast by the second equipment in the local area network, and updating the self-adaptive parameter and the self-adaptive heartbeat cycle by the first equipment according to the received sharing broadcast.
8. The distributed based adaptive heartbeat system of claim 6, in which the period detection module specifically includes:
the detection module is used for sending a mapping establishment request to the cloud platform through a port by the first equipment, establishing NAT mapping to the port, suspending communication, sending a mapping test request to the port from the cloud platform after the testT time, and if the first equipment receives the request, the testT is less than a failure period and the corresponding detection result is valid; if the first device does not receive the request, the testT is greater than the failure period, and the corresponding detection result fails; wherein testT is the detection value.
9. The distributed based adaptive heartbeat system of claim 6 or 8 wherein the period detection module further includes:
the first task allocation module is used for determining a plurality of detection values of the NAT aging time period to be detected in an exponential growth mode, performing detection task allocation and calling the detection module to perform self-adaptive heartbeat period detection.
10. The distributed based adaptive heartbeat system of claim 9, wherein the period detection module further includes:
and the second task allocation module is used for determining a plurality of detection values of the NAT aging time period to be detected in a linear growth mode, allocating detection tasks and calling the detection module to perform self-adaptive heartbeat period detection.
CN201710832684.5A 2017-09-15 2017-09-15 Self-adaptive heartbeat method and system based on distribution Active CN107819642B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710832684.5A CN107819642B (en) 2017-09-15 2017-09-15 Self-adaptive heartbeat method and system based on distribution

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710832684.5A CN107819642B (en) 2017-09-15 2017-09-15 Self-adaptive heartbeat method and system based on distribution

Publications (2)

Publication Number Publication Date
CN107819642A CN107819642A (en) 2018-03-20
CN107819642B true CN107819642B (en) 2020-02-28

Family

ID=61607582

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710832684.5A Active CN107819642B (en) 2017-09-15 2017-09-15 Self-adaptive heartbeat method and system based on distribution

Country Status (1)

Country Link
CN (1) CN107819642B (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109286525B (en) * 2018-09-28 2022-02-25 昆明能讯科技有限责任公司 Double-computer backup method based on MQTT communication and heartbeat between main and standby
CN109602413B (en) * 2018-12-06 2022-01-18 Oppo广东移动通信有限公司 Heartbeat detection method, heartbeat detection device, storage medium and server
CN111698098A (en) * 2019-03-15 2020-09-22 北京京东尚科信息技术有限公司 Communication method, apparatus and computer-readable storage medium
CN109981800A (en) * 2019-04-19 2019-07-05 广州鲁邦通物联网科技有限公司 Method, heartbeat packet transmitting terminal and the interactive system of terminal quick sensing NAT keep-alive time
CN110336708A (en) * 2019-05-24 2019-10-15 重庆科技学院 A kind of elastic keep-alive system and method for Virtual cross-domain communication
CN110971701B (en) * 2019-12-10 2022-08-23 广州番禺职业技术学院 Internet of things communication method and device
CN111953569B (en) * 2020-08-27 2022-04-29 浪潮电子信息产业股份有限公司 State information reporting method, device, equipment and medium
EP3968600A1 (en) * 2020-09-11 2022-03-16 Volkswagen Ag Controlling a communication between a vehicle and a backend device
CN112165517B (en) * 2020-09-22 2022-09-20 成都知道创宇信息技术有限公司 Return source detection method and device, storage medium and electronic equipment
CN113905050B (en) * 2021-08-30 2023-07-18 成都市联洲国际技术有限公司 Method, device and system for detecting internet access information
CN113873017B (en) * 2021-09-06 2023-12-26 绿盟科技集团股份有限公司 Heartbeat cycle adjusting method, device, client and server
CN114826982B (en) * 2022-04-08 2023-08-18 浙江大学 Self-adaptive heartbeat packet adjusting method in micro-service scene
CN115102885B (en) * 2022-06-17 2024-05-14 中建八局第二建设有限公司 Variable-speed heartbeat method for low-power-consumption Internet of things equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103684815A (en) * 2012-09-03 2014-03-26 中国移动通信集团公司 Keep-alive method, device and system for data transmission link
CN103870377A (en) * 2012-12-11 2014-06-18 深圳市腾讯计算机系统有限公司 Method and device for detecting MySQL operation information
CN104243719A (en) * 2013-06-08 2014-12-24 中国移动通信集团公司 Heartbeat period determining method, system and device for long connection in mobile communication network

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103684815A (en) * 2012-09-03 2014-03-26 中国移动通信集团公司 Keep-alive method, device and system for data transmission link
CN103870377A (en) * 2012-12-11 2014-06-18 深圳市腾讯计算机系统有限公司 Method and device for detecting MySQL operation information
CN104243719A (en) * 2013-06-08 2014-12-24 中国移动通信集团公司 Heartbeat period determining method, system and device for long connection in mobile communication network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
"An untold story of middleboxes in ce11u1ar networks";Wand Zhao}uan}，Qian Zhiyun，Xu Qiang，et a1;《Proc of the 11th ACM SIGCOMM Conf.New York:ACM》;20110802;374-385 *
"一种基于自适应心跳机制的MQTT通信协议的研究与应用";温彬民;《中国优秀硕士学位论文全文数据库信息科技》;20151231;I139-9 *

Also Published As

Publication number Publication date
CN107819642A (en) 2018-03-20

Similar Documents

Publication Publication Date Title
CN107819642B (en) Self-adaptive heartbeat method and system based on distribution
JP4857374B2 (en) Method for assigning tasks, data processing system, client data processing node and machine-readable storage medium
Liang et al. Scalable and crash-tolerant load balancing based on switch migration for multiple open flow controllers
CN111538587B (en) Service function chain reconfiguration method based on load balancing
CN107040284B (en) Multi-address access method for improving reliability and stability of data transmission
CN110234127B (en) SDN-based fog network task unloading method
CN110933137B (en) Data synchronization method, system, equipment and readable storage medium
WO2020010670A1 (en) Method and device for acquiring network resource, and scheduling server
CN107919994B (en) Method and server for realizing hot standby of network service dual-computer
CN107135279A (en) It is a kind of to handle the method and apparatus that request is set up in long connection
US7904534B2 (en) Network topology aware configuration of network addresses in wireless personal area networks
CN112631788B (en) Data transmission method and data transmission server
WO2018121201A1 (en) Distributed cluster service structure, node cooperation method and device, terminal and medium
CN109040184B (en) Host node election method and server
CN113992691B (en) Method, device and equipment for distributing edge computing resources and storage medium
CN102624634A (en) Method and device for controlling network bandwidth on client side
CN114338267A (en) Maintenance method, device, equipment, bus network and medium for multiple management nodes
JP2003143163A (en) Method for sharing band width, network architecture of shared band width and network device
CN114168312A (en) Distributed cluster load balancing method and device and storage medium
CN111131081B (en) Method and device for supporting high-performance one-way transmission of multiple processes
CN109451039B (en) Network information transmission processing method
CN106453656B (en) Cluster host selection method and device
CN112533304B (en) Ad hoc network management method, device, system, electronic equipment and storage medium
CN111918338B (en) Wireless cooperative agent method, device and network system
CN113765795A (en) Networking method, device, system and storage medium

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