CN108574595B - Method and device for improving service reliability and user experience - Google Patents
Method and device for improving service reliability and user experience Download PDFInfo
- Publication number
- CN108574595B CN108574595B CN201710149711.9A CN201710149711A CN108574595B CN 108574595 B CN108574595 B CN 108574595B CN 201710149711 A CN201710149711 A CN 201710149711A CN 108574595 B CN108574595 B CN 108574595B
- Authority
- CN
- China
- Prior art keywords
- reply
- probe
- machine learning
- threads
- data packets
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000000523 sample Substances 0.000 claims abstract description 96
- 238000010801 machine learning Methods 0.000 claims abstract description 69
- 238000000691 measurement method Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 10
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 22
- 230000005540 biological transmission Effects 0.000 description 11
- 230000004044 response Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000006855 networking Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
- H04L41/0836—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability to enhance reliability, e.g. reduce downtime
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5061—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the interaction between service providers and their network customers, e.g. customer relationship management
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Electrically Operated Instructional Devices (AREA)
- Debugging And Monitoring (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention provides a method and a device for improving service reliability and user experience, which are beneficial to adaptively adjusting the reply waiting time and the instruction sending number according to different service equipment and network conditions so as to utilize the performances of the service equipment and the network to the maximum extent, thereby improving the user experience while ensuring the service reliability. The method comprises a machine learning phase and an actual operation phase, and is characterized in that: in the machine learning stage, a client sends probe data packets to a service device in a plurality of threads, records the feedback time from sending the probe data packets to receiving the probe replies for the threads receiving the probe replies to the probe data packets, and sets the shortest feedback time for the threads receiving the probe replies as a reply waiting time; in the actual operation phase, the client waits for a reply to an instruction issued on a single thread according to the reply latency.
Description
Technical Field
The present invention relates to the field of computer and software technologies, and in particular, to a method and an apparatus for improving service reliability and user experience.
Background
In the field of networking services, not all operations from clients are completely successful in a network environment due to network or service equipment failure, etc. To ensure that the service device successfully receives and executes the operation instructions, the client typically needs to send the same instructions multiple times for one operation. The network environment and the adopted service equipment are different, and the reliability can be improved although the number of the sent instructions is too large, the user experience is also reduced; while too few instructions are sent, which may improve user experience by reducing reply latency, reliability is also reduced. The response speed of the service device to the operation from the client is determined by the software and hardware performance of the service device. Generally, the longer the reply latency, the less likely the operation will fail, i.e., the higher the service reliability; conversely, the shorter the reply latency, the greater the likelihood of operation failure, i.e., the lower the reliability of service. But longer reply latency will result in a poorer user experience. This results in the user experience and service reliability being a contradiction: that is, improving user experience may sacrifice service reliability, and improving service reliability may sacrifice user experience.
Therefore, in the prior art, the current technical solution mostly adopts UDP (User Datagram Protocol) communication in network communication. To improve service reliability, the client typically sends multiple identical instructions in a single thread and waits for a certain time for a feedback result of the corresponding service device. Fig. 1 is a schematic diagram of a client sending an instruction to communicate with a service device one by one in the prior art. As shown in fig. 1, the client transmits the instructions a plurality of times, waits for a reply from the service device for a short time after each instruction transmission, and if no reply is received, continues to transmit the instructions until a predetermined number of instructions are transmitted. Fig. 2 is a schematic diagram of a client continuously sending a command to communicate with a service device in the prior art. As shown in fig. 2, the client transmits a predetermined number of instructions consecutively a plurality of times, and then waits for a reply from the service device for a long time after all the instructions are transmitted. The technical solutions in fig. 1 and 2 both consider the unreliability of UDP transmission in a network environment and send the instruction multiple times; the reply waiting time is set in consideration of a certain time required for the service apparatus to perform the operation.
From the above prior art, it can be seen that the prior art has the following disadvantages: the reply waiting time and the instruction sending number are fixed values, but the software and hardware performances of the service devices are different, and the adopted network performances are different, so that the fixed reply waiting time and the instruction sending number cannot completely adapt to different service devices and network conditions, and the optimal values of the reply waiting time and the instruction sending number under different service devices and network performances cannot be known, so that the reply waiting time and the instruction sending number cannot be adjusted according to the optimal value under the current condition to improve the user experience while ensuring the service reliability.
Disclosure of Invention
In view of this, the present invention provides a method and an apparatus for improving service reliability and user experience based on machine learning, which are helpful to adaptively adjust the reply waiting time and the number of sent instructions according to different service devices and network conditions, so as to utilize the performance of the service devices and the network to the maximum extent, thereby improving the user experience while ensuring the service reliability. The method is suitable for the field of networking services such as intelligent home, and the client side obtains the reply waiting time and the instruction sending number associated with the communication of the intelligent home equipment in a self-adaptive manner through the machine learning strategy so as to improve the user experience while ensuring the service reliability
To achieve the above object, according to one aspect of the present invention, a method for improving service reliability and user experience is provided.
The method for improving the service reliability and the user experience comprises a machine learning stage and an actual operation stage, and is characterized in that: in the machine learning stage, a client sends probe data packets to a service device in a plurality of threads, records the feedback time from sending the probe data packets to receiving the probe replies for the threads receiving the probe replies to the probe data packets, and sets the shortest feedback time for the threads receiving the probe replies as a reply waiting time; in the actual operation phase, the client waits for a reply to an instruction issued on a single thread according to the reply latency.
Optionally, the method further comprises: in the machine learning stage, the client continuously sends probe data packets with different quantities to the service device in a plurality of threads, records the number of the probe data packets in the threads for the threads receiving the probe replies, and sets the minimum number of the probe data packets in the threads receiving the probe replies as an instruction sending number; in the actual operation stage, the client sends the instructions of the instruction sending number to the service equipment in a single thread.
Optionally, characterized by: if the client does not receive a reply to an instruction issued on a single one of the threads during the actual operation phase, the client re-enters the machine learning phase to determine the reply latency.
Optionally, characterized by: if the client does not receive a reply to an instruction issued on a single one of the threads during the actual operation phase, the client re-enters the machine learning phase to determine the instruction issue number.
Optionally, characterized by: determining that a network or the service device failure occurred during the machine learning phase if the probe reply is not received within a predetermined time for all of the probe packets issued.
Optionally, characterized by: and determining the reply waiting time by adopting a halving measurement method, namely, in the machine learning phase, firstly waiting for the test reply within the preset time, if the test reply is received, trying to wait for the test reply within half of the preset time, and the like until the shortest feedback time of the thread receiving the test reply in the previous attempt is set as the reply waiting time when the test reply cannot be received.
Optionally, characterized by: determining the instruction sending number by adopting a halving measurement method, namely, in the machine learning stage, firstly setting a preset number of the trial data packets in the thread, if the trial reply is received, trying to set a preset number of half of the trial data packets in the thread, and the like, and setting the minimum trial data packet number of the thread receiving the trial reply in the previous trial as the instruction sending number until the trial reply cannot be received.
To achieve the above object, according to another aspect of the present invention, an apparatus for improving service reliability and user experience is provided.
The device for improving the service reliability and the user experience comprises a machine learning module and an actual operation module, and is characterized in that: the machine learning module sends probe data packets to a service device in a plurality of threads, records the feedback time from sending the probe data packets to receiving the probe replies for the threads receiving the probe replies aiming at the probe data packets, and sets the shortest feedback time for the threads receiving the probe replies as the reply waiting time; and the actual operation module waits for the reply of the instruction sent out on the single thread according to the reply waiting time.
Optionally, the method further comprises: the machine learning module continuously sends the probe data packets with different quantities to the service device in a plurality of threads, records the number of the probe data packets in the threads for the threads receiving the probe replies, and sets the minimum number of the probe data packets in the threads receiving the probe replies as an instruction sending number; and the actual operation module sends the instructions of the instruction sending number to the service equipment in a single thread.
Optionally, characterized by: if the real operations module does not receive a reply to an instruction issued on a single one of the threads, the real operations module notifies the machine learning module to re-determine the reply latency.
Optionally, characterized by: if the real operation module does not receive a reply to an instruction issued on a single thread, the real operation module notifies the machine learning module to re-determine the instruction issue number.
Optionally, characterized by: the machine learning module determines that a network or the service device failure has occurred if the probe reply is not received within a predetermined time for all of the probe packets issued.
Optionally, characterized by: and determining the reply waiting time by adopting a halving measurement method, namely, the machine learning module firstly waits for the tentative reply within the preset time, if the tentative reply is received, the machine learning module tries to wait for the tentative reply within half of the preset time, and so on until the tentative reply cannot be received, and setting the shortest feedback time for the thread receiving the tentative reply in the previous attempt as the reply waiting time.
Optionally, characterized by: and determining the instruction sending number by adopting a halving measurement method, namely, the machine learning module firstly sets a preset number of the tentative data packets in the thread, if the tentative reply is received, tries to set a half of the tentative data packets in the thread, and so on, and sets the minimum tentative data packet number of the thread receiving the tentative reply in the previous attempt as the instruction sending number until the tentative reply cannot be received.
According to yet another aspect of the present invention, an electronic device is provided.
An electronic device of the present invention includes: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the one processor to cause the at least one processor to perform the method for improving service reliability and user experience based on machine learning provided by the present invention.
According to still another aspect of the present invention, a readable storage medium is provided.
A non-transitory computer readable storage medium of the present invention stores computer instructions for causing the computer to perform the method of improving service reliability and user experience provided by the present invention.
According to the technical scheme of the invention, the method and the device for improving the service reliability and the user experience by using the machine learning overcome the defect that the fixed reply waiting time and the instruction sending number cannot be completely adapted to different service equipment and network conditions, are beneficial to self-adaptively adjusting the reply waiting time and the instruction sending number according to different service equipment and network conditions, and are used for utilizing the performances of the service equipment and the network to the maximum extent, thereby improving the user experience while ensuring the service reliability.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
FIG. 1 is a diagram illustrating a prior art client sending commands to communicate with a server device one by one;
FIG. 2 is a diagram illustrating a prior art client sending commands to communicate with a server device one by one;
FIG. 3A is a diagram illustrating a client performing machine learning to obtain reply latency and number of instruction sends, according to an embodiment of the invention;
FIG. 3B is a diagram illustrating how a client application replies to latency and the number of command dispatches to perform actual operations, according to an embodiment of the invention;
FIG. 4 is a schematic diagram of the run cycle of the machine learning phase and the actual operation phase of the client according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating a learning reply latency according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of determining a network or service device failure due to not learning a reply latency according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of the number of learn instruction dispatches, according to an embodiment of the invention;
FIG. 8 is a schematic diagram of determining a network or service device failure due to an absence of learning of the number of command transmissions, according to an embodiment of the present invention;
FIG. 9 is a flow chart of machine learning and actual operation of a client according to an embodiment of the present invention;
FIG. 10 is a schematic diagram of an apparatus implementing a method of improving service reliability and user experience, according to an embodiment of the invention;
fig. 11 is a schematic hardware configuration diagram of an electronic device implementing a method for improving service reliability and user experience according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
There are two main parameters related to the actual operation of the client: the number of instructions that need to be repeatedly sent to complete an operation, i.e., the instruction sending number; and waiting for the feedback time of the service equipment after the instruction is sent, namely the reply waiting time. The more the number of the transmitted instructions is, the higher the possibility of receiving feedback on the instructions is, but the longer the time for waiting for feedback is, namely, the service reliability is improved and the user experience is lost; the fewer the number of instructions sent, the less likely feedback to be received for the instructions, but the shorter the time required to wait for feedback, i.e. the improved user experience at the expense of service reliability.
In an environment providing networking services, the run-time of a client includes two types: a machine learning phase and an actual operation phase.
Fig. 3A is a diagram illustrating machine learning by a client to obtain reply latency and the number of instruction transmissions according to an embodiment of the present invention. In the machine learning stage, the client learns the reply waiting time and the instruction sending number influenced by the performance of the network and the service equipment by simultaneously enabling multithreading, wherein each thread is provided with a plurality of probe data packets for realizing probing and sampling functions. The reason why the multithreading is adopted is to improve the learning efficiency, thereby shortening the machine learning time.
Fig. 3B is a diagram illustrating the response latency and the number of command transmissions of the client application to perform actual operations according to an embodiment of the present invention. After completing the machine learning, the client applies the learned reply waiting time and the number of instruction sending to the operation of the service device, and the strategies of learning the reply waiting time and the number of instruction sending adopted by the embodiment of the invention are described in detail in fig. 5 to 8. In the actual operation phase, the client generally needs to enable a thread to send instructions to the service device, set the learned instruction sending number of instructions in the thread, and wait for a reply from the service device according to the learned reply waiting time.
For example, the client may send the learned instruction sending number of instructions to the service device one by one within a single thread in an actual operation phase, and wait for a reply from the service device within the learned reply latency after each instruction sending. For another example, the client continuously sends the learned instruction sending number of instructions to the service device within a single thread in the actual operation phase, and waits for a reply from the service device within a time appropriately increased from the learned reply waiting time after sending all the instructions.
Further, if an operation failure occurs during an actual operation, it is necessary to re-enter the machine learning phase in order to learn the latest reply latency and the number of instruction transmission and to reapply it to a new actual operation.
Alternatively, various operations on the service device may be performed as well in the machine learning phase, but the client may pay a performance penalty for performing both machine learning and actual operations.
Fig. 4 is a schematic diagram of the run cycle of the machine learning phase and the actual operation phase of the client according to an embodiment of the present invention. As shown, the running cycle of the client presents the characteristic that the machine learning process and the actual operation process alternately appear. The client does not receive the reply to the issued command at both times t2 and t4, so the client determines that the current reply waiting time and the number of command transmissions do not adapt to the performance of the current network or service device, thereby initiating relearning to learn the reply waiting time and the number of command transmissions.
Fig. 5 is a diagram illustrating probe reply latency according to an embodiment of the invention. The client sends probe packets to the service device in multiple threads, and records for each thread from the sending of a probe packet to the receipt of a probe reply to a probe packetAnd (4) feeding back time. As shown in fig. 5, no feedback from the service device is received for both thread 1 and thread 2; for thread x, the time from issuing thread x to receiving feedback for thread x is tx(ii) a For thread n, the time from issuing thread n to receiving feedback for thread n is tn. Thus, in the example of FIG. 5, the shortest feedback time for a thread that receives a probe reply is tx，txI.e. the learned reply waiting time. In addition, when learning the reply waiting time, the client sends probe data packets in each thread enough to eliminate interference of the number of the probe data packets on the learning result.
FIG. 6 is a diagram illustrating a determination of a network or service device failure due to an unexplored reply-to-reply latency, according to an embodiment of the invention. As shown in fig. 6, within a predetermined time (not shown), no feedback from the service device is received for all threads, so that it can be determined that the network or the service device has a failure, and the user can be informed that the network or the service device has a failure by prompting.
FIG. 7 is a diagram illustrating the number of probe instruction issue, according to an embodiment of the invention. Similar to fig. 5, the client sends different numbers of probe packets to the service device in succession within multiple threads, and records the number of probe packets within a thread for which a probe reply is received for a probe packet. As shown in fig. 7, no feedback from the service device is received for both thread 1 and thread 2; for thread X, recording that it contains X probe packets because of the feedback received for thread X; for thread N, it is recorded to contain N probe packets because of the feedback received for thread N. Thus, in the example of fig. 7, the minimum number of probe packets for a thread that receives a probe reply is X, which is the learned number of instruction dispatches. In addition, the time for the client to wait for the probe reply is long enough to eliminate the interference of the waiting time on the learning result when the number of the learning instructions is sent.
FIG. 8 is a diagram illustrating a determination of a network or service device failure due to a number of non-probed instruction issue, according to an embodiment of the invention. As shown in fig. 8, within a predetermined time (not shown), no matter how many data packets are contained in the thread, no feedback from the service device is received for all threads, and it may be determined that the network or the service device has a failure, and the user may be notified that the network or the service device has a failure by prompting.
In addition, the reply waiting time can be determined by adopting a halving measurement method, namely, in a machine learning stage, the trial reply is firstly waited for a preset time, if the trial reply is received, the trial reply is waited for a half of the preset time, and the like, and the shortest feedback time of the previous trial for the thread receiving the trial reply is set as the reply waiting time until the trial reply cannot be received. Similarly, the instruction sending number may be determined by a binary measure, that is, in the machine learning phase, a predetermined number of probe packets are first set in a thread, if a probe reply is received, an attempt is made to set half of the predetermined number of probe packets in the thread, and so on, until the probe reply cannot be received, the minimum number of probe packets of the thread that received the probe reply in the previous attempt is set as the instruction sending number. By adopting a halving measurement method to determine the reply waiting time and the instruction sending number, the optimal values of the reply waiting time and the instruction sending number can be determined at a higher speed.
Furthermore, to further ensure reliability of the service, in an actual operation phase, the client may wait for a reply to an instruction issued on a single thread for a time greater than the learned reply latency, for example, 1.1 times the learned reply latency. Similarly, in an actual operation phase, the client sends instructions greater than the learned instruction sending number to the service device within a single thread, for example, sending instructions 1.1 times the learned instruction sending number to the service device within a single thread.
Fig. 9 is a flowchart of the machine learning and actual operation of the client according to an embodiment of the present invention. In S91, the client first starts machine learning, learning the reply latency and the number of instruction transmission according to the policies shown in fig. 5 and 7. That is, the client sends probe packets to the service device in multiple threads, records the feedback time from sending the probe packet to receiving the probe reply for the thread that receives the probe reply for the probe packet, and sets the shortest feedback time for the thread that receives the probe reply as the reply waiting time. Further, the client side continuously sends probe data packets with different quantities to the service device in the multiple threads, records the number of the probe data packets in the threads for the threads receiving the probe responses, and sets the minimum number of the probe data packets in the threads receiving the probe responses as the instruction sending number. In S92, if the client has not received feedback for the probe packet issued within a predetermined time, resulting in a wait time or number of command sends for learning to reply, the flow proceeds to S94, i.e., the client reports to the user that there may be a network or service device failure. In S92, if the client learns the reply waiting time or the number of instruction transmission, the flow proceeds to S93, i.e., the client proceeds to the actual operation stage. In S93, the client sends an instruction to the service device within one thread, sets the instruction of the learned instruction sending number in the thread, and waits for a reply from the service device according to the learned reply waiting time. In S94, if the client does not receive a reply to an instruction issued on a single thread-i.e., an operational exception occurs, the client enters S91 and resumes the machine learning phase to determine a new reply latency and number of instruction dispatches. In S94, if an operation abnormality occurs, the client continues to be in the actual operation phase.
Fig. 10 is a schematic diagram of an apparatus implementing a method for improving service reliability and user experience according to an embodiment of the present invention. As shown in fig. 10, the client apparatus in the present embodiment includes a machine learning module and an actual operation module. The machine learning module sends a probe data packet to the service device to learn the reply waiting time and the instruction sending number, and transmits the learned reply waiting time and the instruction sending number to the actual operation module. The actual operation module applies the reply waiting time and the instruction sending number to the actual instruction sent to the service equipment, and triggers the machine learning module to relearn the reply waiting time and the instruction sending number when the operation abnormity occurs.
The invention also provides an electronic device and a non-transitory computer-readable storage medium according to embodiments of the invention.
The electronic device of the present invention includes: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the one processor, the instructions being executable by the at least one processor to cause the at least one processor to perform the method for improving service reliability and user experience based on machine learning provided by the present invention.
The non-transitory computer readable storage medium of the present invention stores computer instructions for causing a computer to perform the method for improving service reliability and user experience based on machine learning provided by the present invention.
Fig. 11 is a schematic hardware configuration diagram of an electronic device implementing a method for improving service reliability and user experience according to an embodiment of the present invention. As shown in fig. 11, the electronic apparatus includes: one or more processors 111 and memory 112, with one processor 111 being an example in fig. 11. The memory 112 is a non-transitory computer readable storage medium provided by the present invention.
The electronic device of the method for improving service reliability and user experience based on machine learning may further include: an input device 113 and an output device 114.
The processor 111, the memory 112, the input device 113, and the output device 114 may be connected by a bus or other means, and fig. 11 illustrates an example of connection by a bus.
The memory 112, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable program instructions, and modules, such as program instructions corresponding to the method of automatic call distribution in embodiments of the present invention-e.g., the machine learning module and the actual operations module shown in fig. 10. The processor 111 executes various functional applications and data processing for improving service reliability and user experience based on machine learning by running non-transitory software programs, instructions and modules stored in the memory 112, that is, the method for improving service reliability and user experience in the above method embodiment is implemented.
The memory 112 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store the learned reply latency and the number of instruction transmission, and the like. Further, the memory 112 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 112 may optionally include memory located remotely from the processor 111, which may be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 113 may receive, for example, input numeric or character information related to a predetermined time to wait for feedback in the machine learning phase. Output device 114 may include a display device such as a display screen
The one or more modules are stored in the memory 112 and, when executed by the one or more processors 111, perform a method of improving service reliability and user experience in implementing any of the method embodiments described above.
The product can execute the method provided by the embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the method provided by the embodiment of the present invention.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (14)
1. A method for improving service reliability and user experience comprises a machine learning phase and an actual operation phase, and is characterized in that:
the machine learning phase and the actual operation phase occur alternately;
in the machine learning stage, a client sends probe data packets to a service device in a plurality of threads, records the feedback time from sending the probe data packets to receiving the probe replies for the threads receiving the probe replies to the probe data packets, and sets the shortest feedback time for the threads receiving the probe replies as a reply waiting time;
in the actual operation stage, the client side waits for the reply of the instruction sent out on the single thread according to the reply waiting time;
the method further comprises the following steps: in the machine learning stage, the client continuously sends probe data packets with different quantities to the service device in a plurality of threads, records the number of the probe data packets in the threads for the threads receiving the probe replies, and sets the minimum number of the probe data packets in the threads receiving the probe replies as an instruction sending number; in the actual operation stage, the client sends the instructions of the instruction sending number to the service equipment in a single thread.
2. The method of claim 1, wherein:
if the client does not receive a reply to an instruction issued on a single one of the threads during the actual operation phase, the client re-enters the machine learning phase to determine the reply latency.
3. The method of claim 1, wherein:
if the client does not receive a reply to an instruction issued on a single one of the threads during the actual operation phase, the client re-enters the machine learning phase to determine the instruction issue number.
4. The method of claim 1, wherein:
determining that a network or the service device failure occurred during the machine learning phase if the probe reply is not received within a predetermined time for all of the probe packets issued.
5. The method of claim 4, wherein:
and determining the reply waiting time by adopting a halving measurement method, namely, in the machine learning phase, firstly waiting for the test reply within the preset time, if the test reply is received, trying to wait for the test reply within half of the preset time, and the like until the shortest feedback time of the thread receiving the test reply in the previous attempt is set as the reply waiting time when the test reply cannot be received.
6. The method of claim 4, wherein:
determining the instruction sending number by adopting a halving measurement method, namely, in the machine learning stage, firstly setting a preset number of the trial data packets in the thread, if the trial reply is received, trying to set a preset number of half of the trial data packets in the thread, and the like, and setting the minimum trial data packet number of the thread receiving the trial reply in the previous trial as the instruction sending number until the trial reply cannot be received.
7. An apparatus for improving service reliability and user experience, the apparatus comprising a machine learning module and an actual operation module, the apparatus comprising:
the machine learning module and the actual operation module are executed alternately;
the machine learning module sends probe data packets to a service device in a plurality of threads, records the feedback time from sending the probe data packets to receiving the probe replies for the threads receiving the probe replies aiming at the probe data packets, and sets the shortest feedback time for the threads receiving the probe replies as the reply waiting time;
the actual operation module waits for the reply of the instruction sent out on the single thread according to the reply waiting time;
the device further comprises: the machine learning module continuously sends the probe data packets with different quantities to the service device in a plurality of threads, records the number of the probe data packets in the threads for the threads receiving the probe replies, and sets the minimum number of the probe data packets in the threads receiving the probe replies as an instruction sending number; and the actual operation module sends the instructions of the instruction sending number to the service equipment in a single thread.
8. The apparatus of claim 7, wherein:
if the real operations module does not receive a reply to an instruction issued on a single one of the threads, the real operations module notifies the machine learning module to re-determine the reply latency.
9. The apparatus of claim 7, wherein:
if the real operation module does not receive a reply to an instruction issued on a single thread, the real operation module notifies the machine learning module to re-determine the instruction issue number.
10. The apparatus of claim 7, wherein:
the machine learning module determines that a network or the service device failure has occurred if the probe reply is not received within a predetermined time for all of the probe packets issued.
11. The apparatus of claim 10, wherein:
and determining the reply waiting time by adopting a halving measurement method, namely, the machine learning module firstly waits for the tentative reply within the preset time, if the tentative reply is received, the machine learning module tries to wait for the tentative reply within half of the preset time, and so on until the tentative reply cannot be received, and setting the shortest feedback time for the thread receiving the tentative reply in the previous attempt as the reply waiting time.
12. The apparatus of claim 10, wherein:
and determining the instruction sending number by adopting a halving measurement method, namely, the machine learning module firstly sets a preset number of the tentative data packets in the thread, if the tentative reply is received, tries to set a half of the tentative data packets in the thread, and so on, and sets the minimum tentative data packet number of the thread receiving the tentative reply in the previous attempt as the instruction sending number until the tentative reply cannot be received.
13. An electronic device, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the one processor to cause the at least one processor to perform the method of any one of claims 1 to 6.
14. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710149711.9A CN108574595B (en) | 2017-03-14 | 2017-03-14 | Method and device for improving service reliability and user experience |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710149711.9A CN108574595B (en) | 2017-03-14 | 2017-03-14 | Method and device for improving service reliability and user experience |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108574595A CN108574595A (en) | 2018-09-25 |
| CN108574595B true CN108574595B (en) | 2021-10-01 |
Family
ID=63578559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710149711.9A Active CN108574595B (en) | 2017-03-14 | 2017-03-14 | Method and device for improving service reliability and user experience |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108574595B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109921967B (en) * | 2019-03-08 | 2021-05-07 | 美智光电科技股份有限公司 | Instruction processing method and system, gateway device and computer readable storage medium |
| CN109976168B (en) * | 2019-03-08 | 2022-07-08 | 佛山市云米电器科技有限公司 | Decentralized intelligent home control method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011022920A1 (en) * | 2009-08-28 | 2011-03-03 | 中兴通讯股份有限公司 | Timing method and apparatus for hybrid automatic repeat request |
| CN103795564A (en) * | 2012-10-31 | 2014-05-14 | 三星Sds株式会社 | Terminal and method of monitoring network by using the same |
| CN105024933A (en) * | 2014-04-22 | 2015-11-04 | 腾讯科技（深圳）有限公司 | Request packet sending frequency control method and apparatus |
| CN105491594A (en) * | 2015-11-24 | 2016-04-13 | 小米科技有限责任公司 | Network connection processing method and device |
| CN106487603A (en) * | 2015-08-26 | 2017-03-08 | 阿里巴巴集团控股有限公司 | A kind of response test method and device |
-
2017
- 2017-03-14 CN CN201710149711.9A patent/CN108574595B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011022920A1 (en) * | 2009-08-28 | 2011-03-03 | 中兴通讯股份有限公司 | Timing method and apparatus for hybrid automatic repeat request |
| CN103795564A (en) * | 2012-10-31 | 2014-05-14 | 三星Sds株式会社 | Terminal and method of monitoring network by using the same |
| CN105024933A (en) * | 2014-04-22 | 2015-11-04 | 腾讯科技（深圳）有限公司 | Request packet sending frequency control method and apparatus |
| CN106487603A (en) * | 2015-08-26 | 2017-03-08 | 阿里巴巴集团控股有限公司 | A kind of response test method and device |
| CN105491594A (en) * | 2015-11-24 | 2016-04-13 | 小米科技有限责任公司 | Network connection processing method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108574595A (en) | 2018-09-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10637719B2 (en) | Bus exception handing method of robot and bus exception handling device | |
| CN104317764B (en) | A kind of real-time control method of 1553B buses asynchronous message | |
| US12057991B2 (en) | Autonomous server agents | |
| CN108574595B (en) | Method and device for improving service reliability and user experience | |
| CN107948063B (en) | Method for establishing aggregation link and access equipment | |
| CN105812406B (en) | Information transferring method and device based on WEB emulation terminal system | |
| US20220100649A1 (en) | Method for testing a microservice application | |
| AU2019267894B2 (en) | Resource determination method, receiving method, device, network device and terminal | |
| CN116414521A (en) | Virtual machine remote command interaction method, device and medium for distributed heterogeneous shooting ranges | |
| US10298687B2 (en) | Method, apparatus and computer program product for managing storage system | |
| CN101611378B (en) | Method and apparatus for providing passive automated provisioning | |
| US9189370B2 (en) | Smart terminal fuzzing apparatus and method using multi-node structure | |
| CN106506074B (en) | A kind of method and apparatus detecting optical port state | |
| EP3534657A1 (en) | Dynamic time allocation method and device | |
| US10922115B2 (en) | Commissioning of virtualized entities | |
| CN114625501A (en) | Automatic evidence obtaining scheduling system and method based on block chain | |
| US9385968B2 (en) | Methods and devices for connecting to multiple interfaces | |
| CN110908805A (en) | Information distribution method, robot and storage medium | |
| CN105991424B (en) | Network data forwarding method and device | |
| CN116208495B (en) | Network performance tuning method, BMC and server | |
| CN104796228B (en) | A kind of method, apparatus and system of information transmission | |
| US20180212747A1 (en) | Dynamic Adjustment of Wait Time Values | |
| CN113472841B (en) | Implementation method and device for terminating remote procedure call request | |
| CN114615167B (en) | Monitoring method for establishing link connection for negotiation and intelligent network equipment | |
| CN111488229B (en) | Aging processing method and device for RRU (remote radio unit) of base station |
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 |