CN107239973B - System monitoring and early warning method - Google Patents
System monitoring and early warning method Download PDFInfo
- Publication number
- CN107239973B CN107239973B CN201710418411.6A CN201710418411A CN107239973B CN 107239973 B CN107239973 B CN 107239973B CN 201710418411 A CN201710418411 A CN 201710418411A CN 107239973 B CN107239973 B CN 107239973B
- Authority
- CN
- China
- Prior art keywords
- transaction
- time
- monitored
- concurrency
- unit time
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
Abstract
In order to solve the defects of the prior art, the application provides a system monitoring and early warning method, which includes the steps of carrying out statistics on transaction data of a plurality of previous unit times to obtain a front transaction concurrency amount, a monitored transaction concurrency amount and an average response time, carrying out mathematical fitting to obtain an influence relation R1 of the front transaction concurrency amount on the monitored transaction concurrency amount of the next unit time and an influence relation R2 of the monitored transaction concurrency amount of the unit time on the average response time of the monitored transaction, estimating the monitored transaction concurrency amount of the next unit time according to R1 and the front transaction concurrency amount of the unit time, estimating the monitored transaction response time of the next unit time according to R2 and the monitored transaction concurrency amount of the next unit time, and further prejudging whether early warning and emergency operation are needed or not.
Description
Technical Field
The application relates to the technical field of computers, in particular to the technical field of internet, and particularly relates to a system monitoring method.
Background
With the development of mobile internet, users have higher and higher dependence on internet applications, do not want the system to operate without problems, and put higher demands on the use experience, wherein the most important is the system response time.
For large-scale internet platforms, especially for e-commerce, user behaviors are influenced by time and date such as day and night, working day and weekend, legal holidays (clearness festival, morning festival, labor festival, mid-autumn festival, national day and spring festival), emerging holidays (valentine's day, christmas day, twenty-first class) and the like, and thus unpredictability is provided. From the development of internet e-commerce, the existing platform has a certain support degree for massive access and is vulnerable to transaction-type access, for example, during 11 sections, the reason that the final transaction success is affected by waiting, failure and the like always occurs in the payment link.
Therefore, weak links always exist for internet application, the weak links are the most important links for a platform, and how to monitor the weak links and predict and alarm is particularly important, so that sufficient time can be provided for background operation and maintenance personnel, the background software and hardware resource configuration can be adjusted in time, the experience of a user is not influenced, and final transaction can be achieved.
The invention provides a method for monitoring key transactions of a system and pre-judging the running conditions of the key transactions, thereby realizing monitoring and early warning of the key transactions.
Disclosure of Invention
The present application aims to provide a system monitoring and early warning method to solve the technical problems mentioned in the background section above.
The application provides a system monitoring and early warning method, which comprises the following steps:
analyzing the monitored transaction, and listing the prepositive transaction of the monitored transaction; setting a pre-judgment starting threshold value and an early warning threshold value of transaction response time for the monitored transaction;
counting the collected operation data of the prepositive transaction and the monitored transaction according to unit time, wherein the operation data comprises concurrency and average transaction response time;
when the average transaction response time of the monitored transaction in unit time exceeds the pre-judgment starting threshold value, starting pre-judgment on the response time of the monitored transaction, wherein the pre-judgment adopts the following method;
carrying out mathematical fitting on the concurrency of the prepositive transaction of a plurality of continuous unit times before the current time and the monitored transaction concurrency of the next unit time to obtain an influence relation R1 of the prepositive transaction concurrency of the unit time on the monitored transaction concurrency of the next unit time;
carrying out mathematical fitting on the concurrency of the monitored transaction in a plurality of continuous unit times before the current time and the average response time of the monitored transaction to obtain an influence relation R2 of the monitored transaction concurrency in unit time to the average response of the monitored transaction in the unit time;
simulating the concurrency of the monitored transaction in the next unit time according to the preposed transaction concurrency of the current unit time and the influence relation R1, and estimating the average transaction response time of the monitored transaction in the next unit time according to the simulated concurrency of the monitored transaction and the influence relation R2;
if the average response time of the monitored transaction in the next unit time is estimated to exceed the early warning threshold value, early warning is carried out;
preferably, the collected operation data of the preposition transaction and the monitored transaction is counted once every 10-20 minutes (wherein 10 minutes is optimal), namely 10 minutes is adopted in unit time, and the concurrence amount and average response time of 10 minutes content transaction are counted once every 10 minutes;
preferably, when the pre-judgment is carried out, the statistical data of the first 6-10 unit times are adopted for carrying out mathematical fitting, and the data of the first 6 unit times are preferably adopted for carrying out mathematical fitting;
preferably, linear fitting is adopted to obtain an influence relation R1 of the pre-transaction concurrency amount per unit time to the monitored transaction concurrency amount per next unit time;
preferably, a polynomial fitting is adopted to obtain a relation R2 of the monitored transaction concurrency amount per unit time to the transaction response time;
in addition, for the monitored transaction possibly comprising more than one front transaction, fitting calculation needs to be carried out respectively for blind transaction paths to predict that the concurrency amount of the monitored transaction is overlapped.
Because the behaviors of the crowd have strong tendency in a short time (especially the internet often brings massive consistent behaviors due to emergencies), the experience value calculated by long-term historical data is greatly strengthened by adopting timely data to carry out prejudgment. The method can pre-judge the transaction response time in real time, and valuable time is won for emergency treatment. Moreover, the existing cloud computing technology enables the system deployment to be adjusted very quickly, and the adjustment of the system can be completed even if only 10 minutes or even minutes are won to avoid loss.
Drawings
Figure 1 illustrates the process of the invention in the form of a flow chart.
Detailed Description
For an e-commerce transaction system, the most important is known to be payment transaction, and the payment transaction is from the first two preceding transactions, and after browsing, the order can be directly placed or the order can be placed for the goods in the shopping cart.
We set a prejudice initiation threshold ZTY and an early warning threshold ZTJ for payment response time.
To enable monitoring of payment transactions we collect data every 10 minutes, the commodity view amount LL, the payment amount Z L, and the payment evaluation response time ZT. so that we obtain a set of data every 10 minutes
(LL, Z L, ZT), monitoring over time we can obtain a series of data sets.
When the monitoring finds that ZT is greater than ZTY, the prejudgment of the response time of the payment transaction is started;
the last six groups of data were analyzed:
(LL6、ZL6、ZT6)、
(LL5、ZL5、ZT5)、
(LL4、ZL4、ZT4)、
(LL3、ZL3、ZT3)、
(LL2、ZL2、ZT2)、
(LL1、ZL1、ZT1),
smaller numbers indicate more recent data;
firstly, mathematically fitting (LL 6, Z L5), (L0L 5, Z L4), (L14, Z L3), (LL 3, Z L2), (LL 2, ZL1) to obtain an influence relation R1 of the commodity browsing amount on the payment transaction amount in 10 minutes, wherein linear fitting is adopted, so that the payment transaction amount ZL0 in the next 10 minutes can be predicted to be R1 (LL 1);
then, mathematical fitting is carried out on (Z L6, ZT6), (Z L5, ZT5), (Z L4, ZT4), (Z L3, ZT3), (Z L2, ZT2), (Z L1, ZT1) to obtain the relation R2 of the payment transaction amount to the payment transaction response time, and quadratic polynomial fitting is adopted here, so that the response time of the next 10 minutes of payment transaction can be estimated
ZT0=R2(ZL0);
If ZT0> ZTJ, adopt the predetermined emergency response method, for example limit the number of people paid, avoid many and concurrency bring deadlock cause more serious consequence, or adjust and deploy and increase the processing capacity in time through the cloud computing platform; if ZT0< ZTJ, we only need to continue monitoring and prediction.
Embodiments for multiple front end transactions
Further, we need to consider the influence of the commodity browsing and the shopping cart browsing on the payment transaction, and then every 10 minutes we need to count 5 sets of data collected, which are the commodity browsing amount LL, the shopping cart browsing amount GW, the payment transaction amount Z L _ LL from the commodity browsing, the payment transaction amount Z L _ GW from the shopping cart browsing, and the average payment transaction response time ZT.
Similarly, when the monitoring finds ZT > ZTY, the prejudgment of the response time of the payment transaction is started;
the last six groups of data were analyzed:
(LL6、GW6、ZL_LL6、ZL_GW6、ZT6)、
(LL5、GW5、ZL_LL5、ZL_GW5、ZT5)、
(LL4、GW4、ZL_LL4、ZL_GW4、ZT4)、
(LL3、GW3、ZL_LL3、ZL_GW3、ZT3)、
(LL2、GW2、ZL_LL2、ZL_GW2、ZT2)、
(LL1、GW1、ZL_LL1、ZL_GW1、ZT1)、
smaller numbers indicate more recent data;
firstly, mathematically fitting (LL 6, Z L _ L0L 35), (L15, Z L6 _ L24), (L44, Z L _ L53), (L73, Z L _ L82), (L92 and ZL _ LL 1) to obtain an influence relation R _ LL 1 of the commodity browsing amount on the payment transaction amount brought by the commodity browsing in 10 minutes, wherein linear fitting is adopted, and then the payment transaction amount ZL _ LL 0 in the next 10 minutes can be predicted to be R _ LL 1 (LL 1);
then mathematically fitting (GW 6, Z L _ GW5), (GW 5, Z L _ GW4), (GW 4, Z L _ GW3), (GW 3, Z L _ GW2), (GW2, Z L _ GW1) to obtain the influence relation R _ GW1 of the shopping cart browsing amount on the payment transaction amount brought by the shopping cart browsing in 10 minutes, wherein linear fitting is adopted, so that the payment transaction amount Z L _ GW 0 in the next 10 minutes can be judged in advance to be R _ GW 1(GW 1);
then pair ((Z L _ LL 6+ Z L _ GW6), ZT6), ((Z L _ LL 5+ Z L _ GW5), ZT5),
(Z L _ LL 4+ Z LL 1_ GW4), ZT4), ((Z LL 2_ LL 03+ Z LL 4_ GW3), ZT3), ((Z L _ LL 32+ Z L _ GW2), ZT2), ((Z L _ LL 1+ Z L _ GW1), ZT1) are mathematically fitted to obtain a relation R2 of the total payment transaction amount to the payment transaction response time, where a quadratic polynomial fit is used, that can be estimated, and the payment transaction response time ZT0 of the next 10 minutes is R2((Z L _ LL 0+ Z L _ GW 0));
if ZT0> ZTJ, adopt the predetermined emergency response method, for example limit the number of people paid, avoid many and concurrency bring deadlock cause more serious consequence, or adjust and deploy and increase the processing capacity in time through the cloud computing platform; if ZT0< ZTJ, we only need to continue monitoring and prediction.
Similarly, if the number of the preposed transactions of the core transaction is more than two, the core transaction amount of the next unit time of different transaction paths is respectively estimated according to different transaction paths and then accumulated, the estimation of the transaction amount of the core transaction can be realized, and then the estimation of the transaction response time is carried out according to the estimated transaction amount.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (5)
1. A system monitoring and early warning method is characterized in that:
analyzing the monitored transaction, and listing the prepositive transaction of the monitored transaction; setting a pre-judgment starting threshold value and an early warning threshold value of transaction response time for the monitored transaction;
counting the collected operation data of the prepositive transaction and the monitored transaction according to unit time, wherein the operation data comprises concurrency and average transaction response time;
when the average transaction response time of the monitored transaction in unit time exceeds the pre-judgment starting threshold value, starting pre-judgment on the response time of the monitored transaction, wherein the pre-judgment adopts the following method;
carrying out mathematical fitting on the concurrency of the prepositive transaction of a plurality of continuous unit times before the current time and the monitored transaction concurrency of the next unit time to obtain an influence relation R1 of the prepositive transaction concurrency of the unit time on the monitored transaction concurrency of the next unit time;
carrying out mathematical fitting on the concurrency of the monitored transaction in a plurality of continuous unit times before the current time and the average response time of the monitored transaction to obtain an influence relation R2 of the monitored transaction concurrency in unit time to the average response of the monitored transaction in the unit time;
simulating the concurrency of the monitored transaction in the next unit time according to the preposed transaction concurrency of the current unit time and the influence relation R1, and estimating the average transaction response time of the monitored transaction in the next unit time according to the simulated concurrency of the monitored transaction and the influence relation R2;
and if the average response time of the monitored transaction in the next unit time is estimated to exceed the early warning threshold value, early warning is carried out.
2. The system monitoring and forewarning method of claim 1, wherein the unit time is 10 minutes to 20 minutes.
3. The system monitoring and forewarning method of claim 1, wherein the plurality of consecutive units of time is 6-10 consecutive units of time.
4. A system monitoring and pre-warning method according to any one of claims 1-3, wherein the influence relationship R1 is obtained by linear fitting of the amount of concurrence of the preceding transaction for a plurality of consecutive units of time before the current time and the amount of concurrence of the monitored transaction for the next unit of time.
5. A system monitoring and pre-warning method as claimed in any one of claims 1-3, wherein the influence relationship R2 is obtained by performing a polynomial fit on the concurrency of the monitored transaction for a plurality of consecutive units of time before the current time and the average response time of the monitored transaction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710418411.6A CN107239973B (en) | 2017-06-06 | 2017-06-06 | System monitoring and early warning method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710418411.6A CN107239973B (en) | 2017-06-06 | 2017-06-06 | System monitoring and early warning method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107239973A CN107239973A (en) | 2017-10-10 |
CN107239973B true CN107239973B (en) | 2020-07-28 |
Family
ID=59984925
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710418411.6A Active CN107239973B (en) | 2017-06-06 | 2017-06-06 | System monitoring and early warning method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107239973B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108182134A (en) * | 2017-12-29 | 2018-06-19 | 顺丰科技有限公司 | A kind of general-purpose interface monitoring method, device and equipment, storage medium |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102411757A (en) * | 2011-08-05 | 2012-04-11 | 中国工商银行股份有限公司 | Method and system for forecasting capacity of large host central processing unit (CPU) |
CN103237003A (en) * | 2013-01-17 | 2013-08-07 | 北京印天网真科技有限公司 | Method and device for responding to high visiting flow in network |
US8615420B2 (en) * | 2006-08-22 | 2013-12-24 | Infrared Integrated Systems Limited | Automatic self-optimizing queue management system |
CN104158789A (en) * | 2013-05-13 | 2014-11-19 | 腾讯科技(深圳)有限公司 | Method and device for detecting security of payment type website |
CN105389214A (en) * | 2015-11-10 | 2016-03-09 | 中国建设银行股份有限公司 | Monitoring method and system |
CN105930245A (en) * | 2015-12-16 | 2016-09-07 | 中国银联股份有限公司 | Method and system for monitoring service state of transaction terminal |
CN105976114A (en) * | 2016-05-09 | 2016-09-28 | 中国农业银行股份有限公司 | Transaction monitoring method and system |
CN106056400A (en) * | 2016-05-19 | 2016-10-26 | 政和科技股份有限公司 | Method, apparatus and system of predicting number of new users |
CN106294746A (en) * | 2016-08-10 | 2017-01-04 | 中国银行股份有限公司 | A kind of concurrent transaction data processing method and device |
CN106651569A (en) * | 2016-12-26 | 2017-05-10 | 中国建设银行股份有限公司 | Transaction response time obtaining system and method and analysis method |
-
2017
- 2017-06-06 CN CN201710418411.6A patent/CN107239973B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8615420B2 (en) * | 2006-08-22 | 2013-12-24 | Infrared Integrated Systems Limited | Automatic self-optimizing queue management system |
CN102411757A (en) * | 2011-08-05 | 2012-04-11 | 中国工商银行股份有限公司 | Method and system for forecasting capacity of large host central processing unit (CPU) |
CN103237003A (en) * | 2013-01-17 | 2013-08-07 | 北京印天网真科技有限公司 | Method and device for responding to high visiting flow in network |
CN104158789A (en) * | 2013-05-13 | 2014-11-19 | 腾讯科技(深圳)有限公司 | Method and device for detecting security of payment type website |
CN105389214A (en) * | 2015-11-10 | 2016-03-09 | 中国建设银行股份有限公司 | Monitoring method and system |
CN105930245A (en) * | 2015-12-16 | 2016-09-07 | 中国银联股份有限公司 | Method and system for monitoring service state of transaction terminal |
CN105976114A (en) * | 2016-05-09 | 2016-09-28 | 中国农业银行股份有限公司 | Transaction monitoring method and system |
CN106056400A (en) * | 2016-05-19 | 2016-10-26 | 政和科技股份有限公司 | Method, apparatus and system of predicting number of new users |
CN106294746A (en) * | 2016-08-10 | 2017-01-04 | 中国银行股份有限公司 | A kind of concurrent transaction data processing method and device |
CN106651569A (en) * | 2016-12-26 | 2017-05-10 | 中国建设银行股份有限公司 | Transaction response time obtaining system and method and analysis method |
Also Published As
Publication number | Publication date |
---|---|
CN107239973A (en) | 2017-10-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103901305B (en) | Online early warning method for power equipment | |
CN106600463B (en) | Local shape similarity ultra-short term load prediction method and device | |
CN110839040B (en) | Flow monitoring method, model training method, device and storage medium | |
CN110019072B (en) | AB experiment method and system based on real-time data | |
CN108111348A (en) | A kind of security policy manager method and system for enterprise's cloud application | |
CN110401660B (en) | False flow identification method and device, processing equipment and storage medium | |
CN107239973B (en) | System monitoring and early warning method | |
CN112835769A (en) | Service data abnormity diagnosis method, device, equipment and storage medium | |
CN109447496A (en) | Electricity stealing determination method and terminal device | |
CN115689752A (en) | Method, device and equipment for adjusting wind control rule and storage medium | |
CN107306200A (en) | Network failure method for early warning and the gateway for network failure early warning | |
CN105938573A (en) | Actuarial early-warning system and method for medical benefits fund | |
CN112737851A (en) | Internet anti-cheating identification method and platform | |
CN106199251A (en) | A kind of distribution network failure early warning system analyzed based on adaptive modeling and method | |
CN2570789Y (en) | Device for monitoring moisture in transformer oil | |
CN108335044A (en) | A kind of power transmission and transformation equipment state evaluation method | |
CN112783637B (en) | Resource regulation and control method and device | |
CN114663232A (en) | Abnormal transaction behavior monitoring method and device and computer program product | |
CN113626964A (en) | Power grid enterprise security investment dynamic prediction optimization method and system | |
CN112633751A (en) | System for prejudging product running state based on product alarm and fault data | |
CN115296840B (en) | Security early warning method and device based on security state of associated node and electronic equipment | |
CN113541982A (en) | Network element health early warning method and device, computing equipment and computer storage medium | |
KR20140147303A (en) | Money shift monitoring method and server performing the same | |
CN116596665B (en) | Data processing method, device, equipment and storage medium | |
Mata et al. | Automated detection of load changes in large-scale networks |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20200629 Address after: Room 313-2360, 3rd floor, Shanshui building, No.3 Gulou East Street, Miyun District, Beijing Applicant after: Smart Co., Ltd. (Beijing) Technology Co., Ltd Address before: 102209 No. 6, 305-488 District, Xing Gu Economic Development Zone, Pinggu District, Beijing Applicant before: BEIJING CHUANGYUN INTELLECTUAL PROPERTY SERVICE Co.,Ltd. |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |