CN108183927A - The monitoring method and system that a kind of distributed system link calls - Google Patents
The monitoring method and system that a kind of distributed system link calls Download PDFInfo
- Publication number
- CN108183927A CN108183927A CN201711175087.6A CN201711175087A CN108183927A CN 108183927 A CN108183927 A CN 108183927A CN 201711175087 A CN201711175087 A CN 201711175087A CN 108183927 A CN108183927 A CN 108183927A
- Authority
- CN
- China
- Prior art keywords
- node
- link
- identification
- monitoring
- distributed
- 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.)
- Withdrawn
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—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
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer And Data Communications (AREA)
Abstract
The present invention provides the monitoring method and system that a kind of distributed system link calls, and wherein monitoring method is included according to preset sampling rule, the distributed request that acquisition user sends；Link identification is generated, each node in the calling link of the distributed request is marked using the link identification；Node identification is generated, the hierarchical relationship of each node in the calling link is marked using the node identification；Monitoring information is obtained according to the link identification of each vertex ticks and node identification.The present invention, which realizes, is presented full link data, and each request link situation, network condition can be observed in real time, and using flexibly controllable sampling policy, can meet general distributed system, be also applied for the distributed system of high concurrent by monitoring information.
Description
Technical field
The present invention relates to distributed system technical field, the monitoring called more particularly, to distributed system link
Method and system.
Background technology
Current Internet service is usually all realized with large-scale distributed cluster.Service may be from not
Company together, R＆D team, are realized using different programming languages, operated on thousands of servers.With business development,
The distributed system of company becomes to become increasingly complex, and the primary request of user is usually handled by multiple systematic collaboration completions, and
These processing are happened on the even different clusters of different machines, when request processing is abnormal, need fast understanding
System action is pinpointed the problems, and is accurately positioned to being which link is out of joint, this is to have very much for the maintenance personnel of system
Meaning.
Current solution mainly has business to be keen to tracing scheme and Global ID according to scheme.
For business diary tracking scheme, some daily record frames of simple use, such as operation system integrate
Log4j, logback etc., software developer are come auxiliary by recording some crucial business datums in the operation flow of system
Assistant solves behavior on the line of system.Due to business diary sheerly developer coding record, it is arbitrarily scattered there are data, lack
The problem of integrality, investigation problem are got up very bothersome laborious.
For Global ID's tracking scheme, by process inlet generate one it is unrelated with business globally unique
ID is asked, in a distributed system, all operation notes of the secondary request are all associated using request ID, can be realized substantially
With business onrelevant, by asking ID quick positioning questions.But the program needs to integrate specific business diary, and ID is coupled with business
Together, high to the intrusion of operation system, autgmentability is low；And it is simple due to realizing, lack important clue, the boundary of some problems
Face is prompted and real-time intelligent early warning, and operation maintenance personnel understanding is got up relatively difficult.
Invention content
The present invention provides chain in a kind of distributed system for overcoming the above problem or solving the above problems at least partly
The monitoring method and system that road is called.
According to an aspect of the present invention, the monitoring method that a kind of distributed system link calls is provided, including：
According to preset sampling rule, the distributed request that acquisition user sends；
Link identification is generated, each node of the call chain road of the distributed request is carried out using the link identification
Label；
Node identification is generated, using the node identification to the hierarchical relationship of each node in the calling link into rower
Note；
Monitoring information is obtained according to the link identification of each vertex ticks and node identification.
Preferably, described the step of monitoring information is obtained according to the link identification and the node identification, including：
By the global core data for burying a middleware and acquiring link and calling, the core data is corresponded to including each node
Link identification, node identification and call type, by the core data asynchronous write of acquisition to local trace files；
According to the hierarchical relationship of node each in same link, the core data in local trace files is stored to Hbase
In database, monitoring information is obtained；Wherein, row major key of the link identification as Hbase databases, by node identification and node
Row cluster of the combination that call type is formed as Hbase databases.
Preferably, it described the step of monitoring information is obtained according to the link identification and the node identification, also wraps later
It includes：
The monitoring information is preserved, and the monitoring information is sent to front-end interface.
Preferably, the node identification identifies the hierarchical relationship of node by the way of multistage serial number.
Preferably, the core data further includes：Calling moment of node, call take, nodename, hierarchical relationship and
Exception information.
According to another aspect of the present invention, the querying method that a kind of distributed system link calls also is provided, including：
Monitoring information is obtained according to the link identification of each vertex ticks and node identification；
Abnormal distributed request is selected to retrieve the chain of the monitoring objective in Hbase databases as monitoring objective
Line obtains corresponding row data；
It to all nodes in the row data, is ranked up according to node identification, obtains top-down calling link；
The corresponding core data of each node identification is matched in the corresponding node called in link, is retrieved from bottom to top
Each node, using the first node for non-conformance description occur as the source node for leading to the distributed request exception.
According to another aspect of the present invention, the monitoring system that a kind of distributed system link calls also is provided, including：
Acquisition module is asked, for according to preset sampling rule, the distributed request of acquisition user's transmission；
Link indicia module, for generating link identification, using the link identification to the calling of the distributed request
Each node of chain road is marked；
Vertex ticks module, for generating node identification, using the node identification to each node in the calling link
Hierarchical relationship be marked；
Monitoring information obtains module, and monitoring information is obtained for the link identification according to each vertex ticks and node identification.
Preferably, the monitoring system further includes：
The monitoring information for preserving the monitoring information, and is sent to front-end interface by monitoring module.
According to another aspect of the present invention, the inquiry system that a kind of distributed system link calls also is provided, including：
Monitoring module obtains monitoring information for the link identification according to each node and node identification；
Row data obtaining module, for abnormal distributed request to be selected to be examined in Hbase databases as monitoring objective
The link identification of Suo Suoshu monitoring objectives obtains corresponding row data；
Link obtains module, for all nodes in the row data, being ranked up according to node identification, obtained from
Calling link under above；
Source node obtains module, for the corresponding core data of each node identification to be matched to the correspondence called in link
In node, each node is retrieved from bottom to top, using the first node for non-conformance description occur as causing the distributed request different
Normal source node.
The monitoring method and system that distributed system link proposed by the present invention calls are adopted by preset using rule
Collect the distributed request that user sends, general distributed system can be met by having, and be also applied for the distributed system of high concurrent
The advantageous effect of system；Link identification is generated, the calling link of the distributed request is marked using the link identification；
Node identification is generated, the hierarchical relationship of each node in the calling link is marked using the node identification；Based on tune
The full dose abundant data arrived with link acquisition, the link relationship between intuitive presentation project, based on this, can carry out the fast of problem
Speed positioning and the quick investigation of bottleneck node, code performance issue, network latency problems etc. cover all at one glance.
Description of the drawings
Fig. 1 is the flow diagram of monitoring method called according to the distributed system link of the embodiment of the present invention；
Fig. 2 for according to the distributed request of the embodiment of the present invention formed calling link transmittance process figure；
Fig. 3 is the flow diagram according to the generation node identification of the embodiment of the present invention；
Fig. 4 is to bury a stage schematic diagram when being called according to two nodes of the embodiment of the present invention.
Fig. 5 is to bury collecting flowchart schematic diagram of the middleware to distributed request according to the embodiment of the present invention；
Fig. 6 is the flow diagram of querying method called according to the distributed system link of the embodiment of the present invention；
Fig. 7 is the functional block diagram of monitoring system called according to the distributed system link of the embodiment of the present invention；
Fig. 8 is the functional block diagram of inquiry system called according to the distributed system link of the embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawings and examples, the specific embodiment of the present invention is described in further detail.Implement below
Example is used to illustrate the present invention, but be not limited to the scope of the present invention.
The present invention provides the monitoring method and system that a kind of distributed system link calls.In order to make it easy to understand, herein
First the related notion that may relate in the present embodiment and subsequent embodiment is explained：
Distributed system, the software systems being built upon on network.In a distributed system, the meter of one group of independence
What calculation machine was presented to user is a unified entirety, just looks like as being a system.System possesses the physics of many general
And logical resource, task can be dynamically distributed, the physics and logical resource of dispersion realize information exchange by computer network.
Manage the distributed operating system of computer resource in system in a manner of global there are one.In general, for user, distribution
Only there are one model or patterns for formula system.There is one layer of software middleware (middleware) to be responsible for realizing this on operating system
A model.
The common ground of distributed system and computer network system is：Most distributed systems are built upon computer network
On, so distributed system and computer network are being physically essentially identical.They difference lies in：Distribution
The design philosophy and network operating system of formula operating system are different, which dictates that they are in structure, working method and function
On it is also different.The network operating system requirement network user have to be understood that Internet resources when using Internet resources first, and network is used
Family must be known by situations such as function of each computer is with configuration, software resource, network file structure in network, in a network such as
When fruit user will read a shared file, user must be known by this file and be placed under which catalogue of which platform computer；
Distributed operating system is to manage system resource with global mode, it can be the arbitrary schedule network resources of user, and adjust
It is " transparent " to spend journey.When user submits an operation, distributed operating system can select in systems as needed
The operation of user is submitted to the processing routine by most suitable processor, and after processor fulfils assignment, result is transmitted to use
Family.In this process, user is not aware that the presence of multiple processors, this system is like a processor one
Sample.
It buries a little, plays a simple analogy, be exactly to increase monitoring point in the application, be clear how to use from backstage in this way
What does in this monitoring point in family.Data bury be a little data analysis preparation, only burying to have put can just obtain
Data are analyzed.
Middleware：(English name：Middleware) it is to provide the software connected between system software and application software, with
It is a kind of independent convenient for the communication between each component of software, particularly application software for the logic of the concentration of system software
System software or service routine, Distributed Application software shared resource between different technologies by this software.Middleware
On the operating system of client/server, computer resource and network communication are managed.The system being connected, even if it
Have different interfaces, but by middleware between each other remain to exchange information.Performing a critical path of middleware is
Information is transmitted.By middleware, application program may operate in multi-platform.
Hbase databases：It is a high reliability, high-performance, towards row, telescopic distributed memory system, utilizes
HBase technologies can erect large-scale structure storage cluster on cheap PCServer.Hbase is one and is stored towards row
Distributed memory system, its advantage is that the operation of high performance concurrent reading and writing can be realized, while Hbase can also to data into
The transparent cutting of row, allows for storage and is provided with horizontal extension in itself in this way.
There are following two main concepts inside Hbase databases：Row major key (English name：Row key) and row cluster
(English name：Column Family).
Column family are pre-defined before system startup, each Column Family can root
There are multiple column according to " qualifier ".If there are one User tables in system, if according to traditional RDBMS, User tables
In row be fixed, for example schema defines the attributes such as name, age, sex, the attribute of User is that dynamic is not increased.
But in Hbase databases, User tables can be defined, then define info row clusters, the data of User can be divided into：info:
Name=zhangsan, info:Age=30, info:Sex=male etc., if wanting to increase other attribute again later, in this way
Easily only need info:NewProperty can.
Row key be RDBMS in some row major key, but because Hbase databases not supports conditions inquiry with
And the inquiries such as Order by, therefore the design of Row key will be designed according to the query demand of your system.It is newest inquiring
During some Feed, the Row key of Feed can have following three parts to form<userId><timestamp><feedId>, this
The Start Rowkey can be specified to be when inquiring Feed most only since sample<userId><0><0>, End Rowkey are<
userId><Long.MAX_VALUE><Long.MAX_VALUE>It queried, simultaneously as the record in Hbase databases is
It sorts according to rowkey, allowing for inquiry in this way becomes very fast.
The advantages of Hbase databases, is：1st, row can dynamically increase, and be classified as sky and just do not store data, and saving is deposited
Store up space；2nd, automatic segmentation data so that data storage has horizontal scalability automatically；3rd, Hbase can provide height simultaneously
Send out the support of read-write operation.
Elasticsearch is a search server based on Lucene.It provides a distributed multi-user energy
The full-text search engine of power, based on RESTful web interfaces.
It is the flow of the monitoring method signal that a kind of distributed system link of the embodiment of the present invention calls with reference to Fig. 1
Figure, This embodiment describes the present invention it is a kind of it is monitoring distributed ask the method for link data that is generated in calling process,
This method includes：
101st, according to preset sampling rule, the distributed request that acquisition user sends.
It should be noted that low overhead is a crucial design object of the embodiment of the present invention, sampling can reduce open
Pin, but simple sampling may lead to sampled result without representativeness.The embodiment of the present invention by provide it is a variety of sampling rule come
Meet performance and representative both sides demand.
Default sample rule：By selecting several distributed requests in every 100 distributed requests into line trace, this
Kind sampling rule is for the very high service of request amount, it is ensured that traces into valuable information；
But for those not high services of load, it is too low to may result in sample frequency, so as to omit important information,
Second of sampling rule is thus provided, i.e. operation system or service flexibly selects different samplings according to actual loading situation
Strategy, for example to loading low operation system, carry out whole samplings.To loading high operation system selection low probability sampling plan
Slightly.Certain class data is important, matching sampling policy can be used, to important data sampling.
The third is using rule, in the case of request appearance exception, time-out etc. are extreme, it is believed that the data such as exception, time-out
It is very valuable, therefore our decisions are fully sampled in extreme circumstances.
4th kind of sampling rule：For under debugging, test scene, an embodiment of the present invention provides can be aobvious to certain primary request
Show specified sampling.It samples so just not only flexible but also highly controllable.The main function of being to provide similar probe, by once please to certain
It asks, specific specified, this request is wanted to sample, and effect is only limitted to this request.In this request, it can neglect automatically
Self-defined sampling policy and default policy etc., other requests are still influenced by self-defined sampling policy or default policy.
Other sampling rules：By optimizing message, caching and storage organization, batch asynchronous write etc., for optimizing message
For, refer to definition simplify data format, to link identification, node represent etc. core datas short nomenclature is used to be named with
And long message is cut, as storehouse waits for extremely.For optimization caching and storage organization, including defining internal storage data queue knot
Structure, gathered data real time propelling movement to queue heads are periodically written to from queue batch pulling data in trace files, with this substantially
Spend lifting system performance.
102nd, link identification is generated, using the link identification to each node of the call chain road of the distributed request
It is marked.In embodiments of the present invention, link identification (English name：Trace Id) it is globally unique, for identifying primary point
Cloth is asked, and can be transmitted in the entire calling network of distributed request.Different distributed requests corresponds to different chain road signs
Know.
In embodiments of the present invention, link identification be request perform previous existence into, by intercepting each request entrance,
Then by link identification create-rule, (create-rule is string encoding, coded portion：Request method such as HTTP/DUBBO-
Timestamp date -10 random factors of Hour Minute Second), generate link identification.
The transmission of link identification is to be called to call to next stage from higher level to transmit, the mode of transmission and specific requesting party
Formula is related, if HTTP call, higher level call to next stage call when, link identification is transmitted to by http header modes
Next stage；If DUBBO is called, link data is transmitted to by next stage by RMI modes.
Fig. 2 shows the calling link that the distributed request of the embodiment of the present invention is formed, user sends out from client in figure
A distributed request is sent, arrow line represents the call relation of application, the corresponding Trace Id of the distributed request are pass-through to
Using A (namely node A), call relation occurs using A and application B, C and D later, Trace Id are pass-through to application using A
B, C and D calls database (English full name using B and C：Database, English abbreviation：DB it), while by Trace Id is pass-through to
Using C call relation also occurs with application D and caching for DB, therefore Trace Id are pass-through to using D and caching, application by application C
Trace Id are pass-through to file and caching by D with file and caching there is also call relation, therefore using D, it should be noted that
File represents multiple files in Fig. 2, therefore using two Trace Id of transparent transmission between D and file in figure.
103rd, node identification is generated, the hierarchical relationship of each node in the calling link is carried out using the node identification
Label.
For a calling link, a TraceId can be generated first to mark, then to repeatedly being adjusted present in request
It is marked with relationship, by each invocation flags into a Span node (i.e. node), increases generation SpanId numbers by natural number
Word, and this Span node is marked with SpanId, SpanId finally is added to this calls in link.In link is called
It needs to distinguish the hierarchical relationship between each node, node identification (English name is based in the embodiment of the present invention：Span Id) to section
The hierarchical relationship of point is marked.
For once asking, can there are multiple call relations, label to be marked for call relation, for each
The mark of call relation is all marked when calling.And the label of node is all by collected operation system
Title marks.
Fig. 3 shows the flow diagram of the generation node identification of the embodiment of the present invention.Invoke is represented to call and be closed in figure
System, it is clear that the calling between two applications be it is mutual, using both can be used as ask transmitting terminal can also be used as request
Server-side, since A being applied directly to receive the distributed request that user issues, the level using A is up to span0, application
The spanId of A is 0, is the next stage using A using B, and therefore, the level using B is span1, and the spanid using B is 0.1,
It is the next stage using B using C, spanid 0.1.1.
104th, monitoring information is obtained according to the link identification of each vertex ticks and node identification.
As shown in the above, the node of each call chain road all has unique node mark in the embodiment of the present invention
Know, therefore monitoring information is obtained by node identification and link identification, allocating time, the IP of each node can also be recorded
Time-consuming, exception information, self-defining data of operation system etc. are called in location.The monitoring information that the embodiment of the present invention obtains, by right
Primary request is restored, and is formed a complete calling chain graph, is helped quickly to position problem, after analysis-reduction
Calling chain graph, can quickly be navigated in interface problem appearance service node title, machine IP and questions and prospect.
By the gathered data for analyzing each node in call chain road, additionally it is possible to see that the processing of each node expends duration, judgement is
No is bottleneck node.
On the basis of above-described embodiment, the step of monitoring information is obtained according to the link identification and the node identification
Suddenly, including：
By the global core data for burying a middleware and acquiring link and calling, the core data is corresponded to including each node
Link identification, node identification and call type, by the core data asynchronous write of acquisition to local trace files.
In the specific implementation, the global middleware that buries is used to perform following steps：A, acquisition traceID, spanID etc.
Core data, b, by the above-mentioned data asynchronous write of acquisition to local trace files, c, pass through rsyslog programs and send acquisition
To above-mentioned data to message queue cluster in.
The embodiment of the present invention invades application developer near-zero and accesses, corresponding by being provided for different heterogeneous systems
It buries a middleware (such as JAVA, PHP language middleware etc.) and completes link trace.A middleware is being buried, is generating unification
The core datas such as TraceId, and collect such as call type, machine IP, service name, take, hierarchical relationship, exception information,
Self-defining data of operation system etc..
It should be noted that a common technology of burying is will to bury the data generated in real time to collect and be pushed to distal end, it is interior
It is big to deposit expense height, network overhead.And in embodiments of the present invention, a data for acquisition are buried, it can be automatically first according to essence in acquisition
Simple form optimizes, and internal storage data queue is written later, and batch write-in trace files, whole process greatly drop followed by
The low expense of memory, while also there is no network overheads.Suitable for the system of various performance requirements harshness.
In a calling process, there are 4 to bury a stage, referring to Fig. 4, when Fig. 4 shows that two nodes are called
4 bury a stage, respectively：
1st, client sends (English name：Client Send) stage：Client is initiated to bury a little during request, needs to transmit
Some parameters, such as TraceId, SpanId etc.；
2nd, server-side receives (English name：Server Receive) stage：Server-side receives to bury during request a little, backfill one
A little parameters, such as TraceId, SpanId etc.；
3rd, server-side sends (English name：Server Send) stage：Server-side is buried a little when returning to request, will bury points
According to asynchronous write local trace files；
4th, client receives (English name：Client Receive) stage：Client receiving is buried a little when returning the result, will
Bury point data asynchronous write local trace files.
Fig. 5, which is shown, buries collecting flowchart of the middleware to a distributed request, as described in Figure, the distributed request
TraceId=12345, the distributed request need call 4 nodes, i.e., using A, using B, using C and using D, wherein, should
It is 0 with the spanId of A, belongs to same level using B and using C, but since in calling process, there are sequencings, it should
It is 0.1 with the spanId of B, the spanId using C is 0.2, is next level using B using D, is using the spanId of D
0.0.1.It is 0ms at the time of receiving user terminal request using A, the response moment is 69ms, i.e. taking for the distributed request is
69ms.It is directed toward for the SS using A, SS=40ms, referred at the time of server-side is sent for 40ms, that is to say, that please with application B
It asks and is sent by user terminal, successively by A-B-D-B-A-C-A.
From the graph in 5, the secondary distributed request, tracking system can just collect following data：
{ TraceId=12345, spanId=0.0.1, methodD, SR=20, SS=30, duration=10, e
Ndpoint=application D ... }
{ TraceId=12345, spanId=0.0.1, methodD, CS=15, CR=35, duration=20, e
Ndpoint=application D ... }
{ TraceId=12345, spanId=0.1, methodB, SR=10, SS=40, duration=30, end
Point=application B ... }
{ TraceId=12345, spanId=0.1, methodB, CS=0, CR=45, duration=45, end
Point=application A ... }
{ TraceId=12345, spanId=0.2, methodC, SR=52, SS=62, duration=10, end
Point=application C ... }
{ TraceId=12345, spanId=0.2, methodC, CS=47, CR=67, duration=20, en
Dpoint=application A ... }
{ TraceId=12345, spanId=0.2, methodA, SR=0, SS=67, duration=20, endp
Oint=application A ... }
According to the hierarchical relationship of node each in same link, the core data in local trace files is stored to Hbase
In database, monitoring information is obtained；Wherein, row major key of the link identification as Hbase databases, by node identification and node
Row cluster of the combination that call type is formed as Hbase databases.
It should be noted that the monitoring of the embodiment of the present invention mainly by Hbase databases and Elasticsearch into
Row, the RowKey (row major key) using TraceId as Hbase, the call type (English name of SpanId+ nodes：Type,
Wherein when Type is c, represent client, when Type is s, represent server-side) row cluster as Hbase databases.In this way,
In hbase databases, ask just to correspond to a line record each time, traceId is adopted as RowKey, row content for span etc.
The data of collection, a natural whole call chain will condense together, and improve search efficiency.
Table 1 shows the data structure in Hbase databases in the embodiment of the present invention.
As shown in table 1,0.1c, 0.1s, 0.1.1c are that spanId and Type are spliced into, that is to say, that 0.1,0.1.1 be
SpanId, c：Node is as client, s：Node is as server-side.0.1c：0th node transmit a request to as client
1 node, 0.1s：1st node receives request as the server-side of the 0th node.Due to the generation of each distributed request
Rowkeys of the traceId as hbase databases, so seeming that first row is why traceId has in some tables
Span, and do not have in some tablesThere is span to show there is primary call, for example there are span by 0.2c, illustrate the 0th node conduct
Client has invoked the 2nd node again, and similarly there are span by 0.2s, illustrate that the 2nd node has responded the 0th as server-side
The primary calling of a node.
It is described that monitoring information is obtained according to the link identification and the node identification on the basis of the various embodiments described above
The step of, it further includes later：
The monitoring information is preserved, and the monitoring information is sent to front-end interface.
On the basis of the various embodiments described above, the level that the node identification identifies node by the way of multistage serial number is closed
System.By taking the collecting flowchart of the distributed request of Fig. 5 descriptions as an example, the spanId using A is 0, belongs to same using B and using C
Level, but since in calling process, there are sequencings, the spanId of application B is 0.1, is using the spanId of C
0.2, it is next level using B using D, the spanId using D is 0.0.1.
On the basis of the various embodiments described above, the core data further includes：The calling moment of node calls time-consuming, section
Point title, hierarchical relationship and exception information.
On the basis of the various embodiments described above, the embodiment of the present invention also provides a kind of looking into for distributed system link calling
Inquiry method, referring to Fig. 6, including：
601st, monitoring information is obtained according to the monitoring method of the various embodiments described above.
602nd, abnormal distributed request is selected to retrieve the monitoring objective in Hbase databases as monitoring objective
Link identification, obtain corresponding row data.
It should be noted that when distributed request occurs abnormal, the primary request where the problem can be seen first
TraceId usually by mail or SMS alarm mode, sends problematic request traceId, is asked by Rowkey=in real time
The traceId once asked where inscribing, finds corresponding row data and all extracts complete line data in memory.
603rd, it to all nodes in the row data, is ranked up according to node identification, obtains top-down calling
Link.
Specifically, in memory in one's own profession content all span nodes analyze, and according to the label 0 of SpanId,
0.1st, 0.1.1,0.2 ... .. are ranked up, and form a complete top-down calling chain graph, and shown in UI interfaces
The calling chain graph.
604th, the corresponding core data of each node identification is matched in the corresponding node called in link, from bottom to top
Each node is retrieved, using the first node for non-conformance description occur as the source node for leading to the distributed request exception.
Specifically, each spanId is marked other data in corresponding span be shown to calls the corresponding of chain graph to save
Point in.Data in each Span include the total data for calling acquisition every time, as spanId, service node IP, send with
Response time point, handling result and handling duration, network transmission duration, non-conformance description etc..Chain graph is called in observation from bottom to top
Each node, the 1st node for non-conformance description occur must be this source node for going wrong of request.
On the basis of the various embodiments described above, the present invention also provides the monitoring systems that a kind of distributed system link calls
System, referring to Fig. 7, including：
Acquisition module 701 is asked, for according to preset sampling rule, the distributed request of acquisition user's transmission.
It should be noted that low overhead is a crucial design object of the embodiment of the present invention, sampling can reduce open
Pin, but simple sampling may lead to sampled result without representativeness.The request acquisition module 701 of the embodiment of the present invention is by carrying
Meet performance and representative both sides demand for a variety of sampling rules.
Default sample rule：By selecting several distributed requests in every 100 distributed requests into line trace, this
Kind sampling rule is for the very high service of request amount, it is ensured that traces into valuable information；
But for those not high services of load, it is too low to may result in sample frequency, so as to omit important information,
Second of sampling rule is thus provided, i.e. operation system or service flexibly selects different samplings according to actual loading situation
Strategy, for example to loading low operation system, carry out whole samplings.To loading high operation system selection low probability sampling plan
Slightly.Certain class data is important, matching sampling policy can be used, to important data sampling.
The third is using rule, in the case of request appearance exception, time-out etc. are extreme, it is believed that the data such as exception, time-out
It is very valuable, therefore our decisions are fully sampled in extreme circumstances.
4th kind of sampling rule：For under debugging, test scene, an embodiment of the present invention provides can be aobvious to certain primary request
Show specified sampling.It samples so just not only flexible but also highly controllable.The main function of being to provide similar probe, by once please to certain
It asks, specific specified, this request is wanted to sample, and effect is only limitted to this request.In this request, it can neglect automatically
Self-defined sampling policy and default policy etc., other requests are still influenced by self-defined sampling policy or default policy.
Other sampling rules：By optimizing message, caching and storage organization, batch asynchronous write etc., for optimizing message
For, refer to definition simplify data format, to link identification, node represent etc. core datas short nomenclature is used to be named with
And long message is cut, as storehouse waits for extremely.For optimization caching and storage organization, including defining internal storage data queue knot
Structure, gathered data real time propelling movement to queue heads are periodically written to from queue batch pulling data in trace files, with this substantially
Spend lifting system performance.
Link indicia module 702, for generating link identification, using the link identification to the tune of the distributed request
Is marked with each node in link
In embodiments of the present invention, the link identification that link indicia module 702 generates is globally unique, for identifying primary point
Cloth is asked, and can be transmitted in the entire calling network of distributed request.Different distributed requests corresponds to different chain road signs
Know.
Vertex ticks module 703, for generating node identification, using each node identification to each node in the calling link
Hierarchical relationship be marked.
For a calling link, a TraceId can be generated first to mark, then to repeatedly being adjusted present in request
It is marked with relationship, by each invocation flags into a Span node (i.e. node), increases generation SpanId numbers by natural number
Word, and this Span node is marked with SpanId, SpanId finally is added to this calls in link.In link is called
It needs to distinguish the hierarchical relationship between each node, being based on vertex ticks module 703 in the embodiment of the present invention generates node identification (English
Literary fame claims：Span Id) hierarchical relationship of node is marked.
Monitoring information obtains module 704, and monitoring letter is obtained for the link identification according to each vertex ticks and node identification
Breath.As shown in the above, the node of each call chain road all has unique node identification in the embodiment of the present invention, because
This monitoring information obtains module 704 and obtains monitoring information by node identification and link identification, can also record each node
Allocating time, calls time-consuming, exception information, self-defining data of operation system etc. at IP address.
On the basis of the various embodiments described above, the monitoring system of the embodiment of the present invention further includes：
The monitoring information for preserving the monitoring information, and is sent to front-end interface by monitoring module.
On the basis of the various embodiments described above, the present invention also provides the inquiry systems that a kind of distributed system link calls
System, referring to Fig. 8, including：
Monitoring module 801, for obtaining monitoring information according to the monitoring method of above-described embodiment.
Row data obtaining module 802, for selecting abnormal distributed request as monitoring objective, in Hbase databases
The link identification of the middle retrieval monitoring objective, obtains corresponding row data.
Link obtains module 803, for all nodes in the row data, being ranked up, obtaining according to node identification
Obtain top-down calling link.
Source node obtains module 804, is called in link for the corresponding core data of each node identification to be matched to
In corresponding node, retrieve each node from bottom to top, using the first node for non-conformance description occur as cause it is described it is distributed please
Seek abnormal source node.
The apparatus embodiments described above are merely exemplary, wherein the unit illustrated as separating component can
To be or may not be physically separate, the component shown as unit may or may not be physics list
Member, you can be located at a place or can also be distributed in multiple network element.It can be selected according to the actual needs
In some or all of module realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying creativeness
Labour in the case of, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on such understanding, on
Technical solution is stated substantially in other words to embody the part that the prior art contributes in the form of software product, it should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and (can be personal computer, server or the network equipment etc.) so that computer equipment is used to perform each implementation
Method described in certain parts of example or embodiment.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
The present invention is described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that：It still may be used
To modify to the technical solution recorded in foregoing embodiments or carry out equivalent replacement to which part technical characteristic；
And these modification or replace, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (9)
1. a kind of monitoring method that distributed system link calls, which is characterized in that including：
According to preset sampling rule, the distributed request that acquisition user sends；
Link identification is generated, using the link identification to each node called in link of the distributed request into rower
Note；
Node identification is generated, the hierarchical relationship of each node in the calling link is marked using the node identification；
Monitoring information is obtained according to the link identification of each vertex ticks and node identification.
2. the monitoring method that distributed system link as described in claim 1 calls, which is characterized in that described in the basis
The step of link identification and the node identification obtain monitoring information, including：
A core data for middleware acquisition link calling is buried by global, by the core data asynchronous write of acquisition extremely
Local trace files, the core data include the corresponding link identification of each node, node identification and call type；
According to the hierarchical relationship of node each in same link, the core data in local trace files is stored to Hbase data
In library, monitoring information is obtained；Wherein, row major key of the link identification as Hbase databases, by the calling of node identification and node
Row cluster of the combination that type is formed as Hbase databases.
3. the monitoring method that the distributed system link as described in claim 1 or 3 calls, which is characterized in that the basis
The step of link identification and the node identification obtain monitoring information, further includes later：
The monitoring information is preserved, and the monitoring information is sent to front-end interface.
4. the monitoring method that the distributed system link as described in claims 1 or 2 or 3 calls, which is characterized in that the section
Point identification identifies the hierarchical relationship of node by the way of multistage serial number.
5. the monitoring method that distributed system link as claimed in claim 2 calls, which is characterized in that the core data
It further includes：The calling moment of node calls time-consuming, nodename, hierarchical relationship and exception information.
6. a kind of querying method that distributed system link calls, which is characterized in that including：
Monitoring information is obtained according to the link identification of each vertex ticks and node identification；
Abnormal distributed request is selected to retrieve the chain road sign of the monitoring objective in Hbase databases as monitoring objective
Know, obtain corresponding row data；
It to all nodes in the row data, is ranked up according to node identification, obtains top-down calling link；
The corresponding core data of each node identification is matched in the corresponding node called in link, retrieval is each from bottom to top
Node, using the first node for non-conformance description occur as the source node for leading to the distributed request exception.
7. a kind of monitoring system that distributed system link calls, which is characterized in that including：
Acquisition module is asked, for according to preset sampling rule, the distributed request of acquisition user's transmission；
Link indicia module, for generating link identification, using the link identification to the calling link of the distributed request
In each node be marked；
Vertex ticks module, for generating node identification, using the node identification to the layer of each node in the calling link
Grade relationship is marked；
Monitoring information obtains module, and monitoring information is obtained for the link identification according to each vertex ticks and node identification.
8. monitoring system as claimed in claim 7, which is characterized in that further include：
The monitoring information for preserving the monitoring information, and is sent to front-end interface by monitoring module.
9. a kind of inquiry system that distributed system link calls, which is characterized in that including：
Monitoring module obtains monitoring information for the link identification according to each node and node identification；
Row data obtaining module, for abnormal distributed request to be selected to retrieve institute in Hbase databases as monitoring objective
The link identification of monitoring objective is stated, obtains corresponding row data；
Link obtains module, for all nodes in the row data, being ranked up according to node identification, obtained from upper
Under calling link；
Source node obtains module, for the corresponding core data of each node identification to be matched to the corresponding node called in link
In, each node is retrieved from bottom to top, using the first node for non-conformance description occur as causing the distributed request abnormal
Source node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711175087.6A CN108183927A (en) | 2017-11-22 | 2017-11-22 | The monitoring method and system that a kind of distributed system link calls |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711175087.6A CN108183927A (en) | 2017-11-22 | 2017-11-22 | The monitoring method and system that a kind of distributed system link calls |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108183927A true CN108183927A (en) | 2018-06-19 |
Family
ID=62545151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711175087.6A Withdrawn CN108183927A (en) | 2017-11-22 | 2017-11-22 | The monitoring method and system that a kind of distributed system link calls |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108183927A (en) |
Cited By (48)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108984404A (en) * | 2018-07-11 | 2018-12-11 | 江苏满运软件科技有限公司 | A kind of exception information processing method and system, a kind of computer product |
| CN109104302A (en) * | 2018-07-18 | 2018-12-28 | 杭州鑫合汇互联网金融服务有限公司 | A kind of full link tracing monitoring method |
| CN109271453A (en) * | 2018-10-22 | 2019-01-25 | 阿里巴巴集团控股有限公司 | A kind of determination method and apparatus of database volume |
| CN109359094A (en) * | 2018-08-03 | 2019-02-19 | 挖财网络技术有限公司 | A kind of full link tracing method and device of distributed system journal |
| CN109522263A (en) * | 2018-11-16 | 2019-03-26 | 郑州云海信息技术有限公司 | A kind of I2C link monitoring system |
| CN109768962A (en) * | 2018-12-13 | 2019-05-17 | 平安科技（深圳）有限公司 | Firewall strategy-generating method, device, computer equipment and storage medium |
| CN109873717A (en) * | 2019-01-18 | 2019-06-11 | 深圳壹账通智能科技有限公司 | Monitoring method, device, computer equipment and storage medium |
| CN109885458A (en) * | 2019-01-16 | 2019-06-14 | 深圳壹账通智能科技有限公司 | Monitoring method, device, computer system and storage medium based on CAT |
| CN109923847A (en) * | 2018-08-29 | 2019-06-21 | 华为技术有限公司 | Call discovery method, apparatus, equipment and the storage medium of link |
| CN109981352A (en) * | 2019-03-06 | 2019-07-05 | 深圳微品致远信息科技有限公司 | A kind of end-to-end application monitoring method, system and storage medium based on detachable distributed system |
| CN110096374A (en) * | 2019-03-20 | 2019-08-06 | 西安电子科技大学 | Communication middleware control system and method between internal multi-class computing unit |
| CN110262889A (en) * | 2019-06-27 | 2019-09-20 | 深圳前海微众银行股份有限公司 | A kind of link tracing method and device |
| CN110351131A (en) * | 2019-06-28 | 2019-10-18 | 北京奇才天下科技有限公司 | It is a kind of for the monitoring method of distributed link, device and electronic equipment |
| CN110401579A (en) * | 2019-06-18 | 2019-11-01 | 平安科技（深圳）有限公司 | The full link data method of sampling, device, equipment and storage medium based on hash table |
| CN110717032A (en) * | 2019-08-26 | 2020-01-21 | 四川新网银行股份有限公司 | Method for generating and displaying multi-system method call link diagram |
| CN110737592A (en) * | 2019-09-16 | 2020-01-31 | 平安科技（深圳）有限公司 | link abnormity identification method, server and computer readable storage medium |
| CN110928934A (en) * | 2018-09-19 | 2020-03-27 | 北京京东尚科信息技术有限公司 | Data processing method and device for business analysis |
| CN111078551A (en) * | 2019-12-13 | 2020-04-28 | 浙江诺诺网络科技有限公司 | Full link testing method, device and system and computer readable storage medium |
| CN111078435A (en) * | 2019-12-18 | 2020-04-28 | 支付宝(杭州)信息技术有限公司 | Service processing method and device and electronic equipment |
| CN111124819A (en) * | 2019-12-10 | 2020-05-08 | 中国建设银行股份有限公司 | Method and device for monitoring full link |
| CN111159590A (en) * | 2019-12-31 | 2020-05-15 | 亚信科技（中国）有限公司 | Serial connection method and device based on front-end and back-end service call links |
| CN111221659A (en) * | 2018-11-23 | 2020-06-02 | 北京图森智途科技有限公司 | Subscription performance tracking system of multi-robot operating system environment |
| CN111224843A (en) * | 2019-12-31 | 2020-06-02 | 北京百度网讯科技有限公司 | Resource link monitoring method, device, equipment and storage medium |
| CN111385122A (en) * | 2018-12-29 | 2020-07-07 | 广州市百果园信息技术有限公司 | Distributed system link tracking method and device, computer equipment and storage medium |
| WO2020147336A1 (en) * | 2019-01-18 | 2020-07-23 | 苏宁易购集团股份有限公司 | Micro-service full-link monitoring system and method |
| CN111770022A (en) * | 2020-06-28 | 2020-10-13 | 中国平安财产保险股份有限公司 | Link monitoring-based capacity expansion method, system, equipment and computer storage medium |
| CN111782534A (en) * | 2020-07-03 | 2020-10-16 | 普信恒业科技发展（北京）有限公司 | Business process monitoring method, device and system |
| CN112134739A (en) * | 2020-09-24 | 2020-12-25 | 平安科技（深圳）有限公司 | Link state tracking method and device, electronic equipment and computer storage medium |
| CN112131013A (en) * | 2019-06-25 | 2020-12-25 | 顺丰科技有限公司 | Monitoring method, device, equipment and storage medium of call chain in distributed system |
| CN112348208A (en) * | 2020-10-13 | 2021-02-09 | 贵州电网有限责任公司 | System and method for monitoring metering and collecting key service full link |
| CN112395180A (en) * | 2020-12-10 | 2021-02-23 | 南威软件股份有限公司 | Method for realizing service log full link tracking |
| CN112486786A (en) * | 2020-11-12 | 2021-03-12 | 贝壳技术有限公司 | Calling link tracking method and device |
| CN112925721A (en) * | 2021-03-29 | 2021-06-08 | 建信金融科技有限责任公司 | Distributed system testing method and device |
| CN112965973A (en) * | 2021-02-10 | 2021-06-15 | 中国工商银行股份有限公司 | Distributed database monitoring method and device based on full link monitoring |
| CN113114533A (en) * | 2021-04-08 | 2021-07-13 | 中国工商银行股份有限公司 | Network time consumption display method and device for distributed service invocation |
| CN113407511A (en) * | 2021-06-23 | 2021-09-17 | 中移（杭州）信息技术有限公司 | Log aggregation method, log aggregation equipment and computer program product |
| CN113452607A (en) * | 2020-03-24 | 2021-09-28 | 华为技术有限公司 | Distributed link acquisition method and device, computing equipment and storage medium |
| CN113472850A (en) * | 2021-05-31 | 2021-10-01 | 北京达佳互联信息技术有限公司 | Link data acquisition method, device, system, electronic equipment and storage medium |
| CN113537824A (en) * | 2021-08-03 | 2021-10-22 | 南方电网数字电网研究院有限公司 | Power grid data full link monitoring system |
| CN113760677A (en) * | 2021-02-02 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Abnormal link analysis method, device, equipment and storage medium |
| CN114143221A (en) * | 2021-11-29 | 2022-03-04 | 重庆富民银行股份有限公司 | Service asset management method, platform and storage medium based on full link monitoring |
| CN114331209A (en) * | 2022-01-10 | 2022-04-12 | 北京蓝莓时节科技有限公司 | Recommendation strategy data tracking method, device, system and storage medium |
| CN114756469A (en) * | 2022-04-24 | 2022-07-15 | 阿里巴巴（中国）有限公司 | Data relation analysis method and device and electronic equipment |
| CN115022213A (en) * | 2022-06-30 | 2022-09-06 | 福州畅昕信息科技有限公司 | Method for identifying request abnormity and storage medium |
| CN115473839A (en) * | 2021-06-11 | 2022-12-13 | 北京字跳网络技术有限公司 | Data processing method, device and equipment based on buried point and storage medium |
| CN116170344A (en) * | 2022-12-26 | 2023-05-26 | 海尔优家智能科技（北京）有限公司 | Message monitoring method and device, storage medium and electronic device |
| CN117170985A (en) * | 2023-11-02 | 2023-12-05 | 武汉大学 | Distributed monitoring method and system for open geographic information network service |
| CN110262889B (en) * | 2019-06-27 | 2024-10-29 | 深圳前海微众银行股份有限公司 | Link tracking method and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105224445A (en) * | 2015-10-28 | 2016-01-06 | 北京汇商融通信息技术有限公司 | Distributed tracking system |
| CN106326017A (en) * | 2016-08-17 | 2017-01-11 | 广东亿迅科技有限公司 | Realization system for call chain based on Annotation application |
| CN106487596A (en) * | 2016-10-26 | 2017-03-08 | 宜人恒业科技发展（北京）有限公司 | Distributed Services follow the tracks of implementation method |
| CN106502874A (en) * | 2016-10-26 | 2017-03-15 | 南京途牛科技有限公司 | A kind of call chain tracking |
| CN106503276A (en) * | 2017-01-06 | 2017-03-15 | 山东浪潮云服务信息科技有限公司 | A kind of method and apparatus of the time series databases for real-time monitoring system |
| CN106603270A (en) * | 2016-11-10 | 2017-04-26 | 广州酷狗计算机科技有限公司 | Tracking method and device for invoked link in distributed system |
| CN106790718A (en) * | 2017-03-16 | 2017-05-31 | 北京搜狐新媒体信息技术有限公司 | Service call link analysis method and system |
| CN107135276A (en) * | 2017-06-28 | 2017-09-05 | 北京中电普华信息技术有限公司 | A kind of full link monitoring methods, devices and systems under micro services framework |
-
2017
- 2017-11-22 CN CN201711175087.6A patent/CN108183927A/en not_active Withdrawn
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105224445A (en) * | 2015-10-28 | 2016-01-06 | 北京汇商融通信息技术有限公司 | Distributed tracking system |
| CN106326017A (en) * | 2016-08-17 | 2017-01-11 | 广东亿迅科技有限公司 | Realization system for call chain based on Annotation application |
| CN106487596A (en) * | 2016-10-26 | 2017-03-08 | 宜人恒业科技发展（北京）有限公司 | Distributed Services follow the tracks of implementation method |
| CN106502874A (en) * | 2016-10-26 | 2017-03-15 | 南京途牛科技有限公司 | A kind of call chain tracking |
| CN106603270A (en) * | 2016-11-10 | 2017-04-26 | 广州酷狗计算机科技有限公司 | Tracking method and device for invoked link in distributed system |
| CN106503276A (en) * | 2017-01-06 | 2017-03-15 | 山东浪潮云服务信息科技有限公司 | A kind of method and apparatus of the time series databases for real-time monitoring system |
| CN106790718A (en) * | 2017-03-16 | 2017-05-31 | 北京搜狐新媒体信息技术有限公司 | Service call link analysis method and system |
| CN107135276A (en) * | 2017-06-28 | 2017-09-05 | 北京中电普华信息技术有限公司 | A kind of full link monitoring methods, devices and systems under micro services framework |
Cited By (69)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108984404A (en) * | 2018-07-11 | 2018-12-11 | 江苏满运软件科技有限公司 | A kind of exception information processing method and system, a kind of computer product |
| CN109104302A (en) * | 2018-07-18 | 2018-12-28 | 杭州鑫合汇互联网金融服务有限公司 | A kind of full link tracing monitoring method |
| CN109359094A (en) * | 2018-08-03 | 2019-02-19 | 挖财网络技术有限公司 | A kind of full link tracing method and device of distributed system journal |
| CN109359094B (en) * | 2018-08-03 | 2021-04-16 | 挖财网络技术有限公司 | Distributed system log full-link tracking method and device |
| CN109923847A (en) * | 2018-08-29 | 2019-06-21 | 华为技术有限公司 | Call discovery method, apparatus, equipment and the storage medium of link |
| CN110928934A (en) * | 2018-09-19 | 2020-03-27 | 北京京东尚科信息技术有限公司 | Data processing method and device for business analysis |
| CN109271453A (en) * | 2018-10-22 | 2019-01-25 | 阿里巴巴集团控股有限公司 | A kind of determination method and apparatus of database volume |
| CN109271453B (en) * | 2018-10-22 | 2021-08-27 | 创新先进技术有限公司 | Method and device for determining database capacity |
| CN109522263B (en) * | 2018-11-16 | 2021-10-29 | 郑州云海信息技术有限公司 | I2C link monitoring system |
| CN109522263A (en) * | 2018-11-16 | 2019-03-26 | 郑州云海信息技术有限公司 | A kind of I2C link monitoring system |
| CN111221659A (en) * | 2018-11-23 | 2020-06-02 | 北京图森智途科技有限公司 | Subscription performance tracking system of multi-robot operating system environment |
| CN111221659B (en) * | 2018-11-23 | 2023-10-03 | 北京图森智途科技有限公司 | Subscription performance tracking system of multi-robot operating system environment |
| CN109768962B (en) * | 2018-12-13 | 2022-04-12 | 平安科技（深圳）有限公司 | Firewall strategy generation method and device, computer equipment and storage medium |
| CN109768962A (en) * | 2018-12-13 | 2019-05-17 | 平安科技（深圳）有限公司 | Firewall strategy-generating method, device, computer equipment and storage medium |
| CN111385122A (en) * | 2018-12-29 | 2020-07-07 | 广州市百果园信息技术有限公司 | Distributed system link tracking method and device, computer equipment and storage medium |
| CN111385122B (en) * | 2018-12-29 | 2023-06-16 | 广州市百果园信息技术有限公司 | Distributed system link tracking method, device, computer equipment and storage medium |
| CN109885458A (en) * | 2019-01-16 | 2019-06-14 | 深圳壹账通智能科技有限公司 | Monitoring method, device, computer system and storage medium based on CAT |
| WO2020147419A1 (en) * | 2019-01-18 | 2020-07-23 | 深圳壹账通智能科技有限公司 | Monitoring method and apparatus, computer device and storage medium |
| CN109873717A (en) * | 2019-01-18 | 2019-06-11 | 深圳壹账通智能科技有限公司 | Monitoring method, device, computer equipment and storage medium |
| WO2020147336A1 (en) * | 2019-01-18 | 2020-07-23 | 苏宁易购集团股份有限公司 | Micro-service full-link monitoring system and method |
| CN111464373A (en) * | 2019-01-18 | 2020-07-28 | 苏宁易购集团股份有限公司 | Micro-service full-link monitoring system and method |
| CN109981352A (en) * | 2019-03-06 | 2019-07-05 | 深圳微品致远信息科技有限公司 | A kind of end-to-end application monitoring method, system and storage medium based on detachable distributed system |
| CN110096374A (en) * | 2019-03-20 | 2019-08-06 | 西安电子科技大学 | Communication middleware control system and method between internal multi-class computing unit |
| CN110096374B (en) * | 2019-03-20 | 2023-03-14 | 西安电子科技大学 | System and method for controlling communication middleware between internal multi-class computing units |
| WO2020252896A1 (en) * | 2019-06-18 | 2020-12-24 | 平安科技（深圳）有限公司 | Hash table-based full-link data sampling method, apparatus, and device, and storage medium |
| CN110401579A (en) * | 2019-06-18 | 2019-11-01 | 平安科技（深圳）有限公司 | The full link data method of sampling, device, equipment and storage medium based on hash table |
| CN110401579B (en) * | 2019-06-18 | 2022-08-23 | 平安科技（深圳）有限公司 | Full link data sampling method, device and equipment based on hash table and storage medium |
| CN112131013A (en) * | 2019-06-25 | 2020-12-25 | 顺丰科技有限公司 | Monitoring method, device, equipment and storage medium of call chain in distributed system |
| CN110262889B (en) * | 2019-06-27 | 2024-10-29 | 深圳前海微众银行股份有限公司 | Link tracking method and device |
| CN110262889A (en) * | 2019-06-27 | 2019-09-20 | 深圳前海微众银行股份有限公司 | A kind of link tracing method and device |
| CN110351131A (en) * | 2019-06-28 | 2019-10-18 | 北京奇才天下科技有限公司 | It is a kind of for the monitoring method of distributed link, device and electronic equipment |
| CN110351131B (en) * | 2019-06-28 | 2022-04-05 | 北京奇才天下科技有限公司 | Monitoring method and device for distributed link and electronic equipment |
| CN110717032A (en) * | 2019-08-26 | 2020-01-21 | 四川新网银行股份有限公司 | Method for generating and displaying multi-system method call link diagram |
| CN110737592B (en) * | 2019-09-16 | 2024-01-30 | 平安科技（深圳）有限公司 | Link abnormality identification method, server and computer readable storage medium |
| CN110737592A (en) * | 2019-09-16 | 2020-01-31 | 平安科技（深圳）有限公司 | link abnormity identification method, server and computer readable storage medium |
| CN111124819A (en) * | 2019-12-10 | 2020-05-08 | 中国建设银行股份有限公司 | Method and device for monitoring full link |
| CN111124819B (en) * | 2019-12-10 | 2023-09-22 | 中国建设银行股份有限公司 | Method and device for full link monitoring |
| CN111078551B (en) * | 2019-12-13 | 2023-09-22 | 浙江诺诺网络科技有限公司 | Full-link testing method, device and system and computer readable storage medium |
| CN111078551A (en) * | 2019-12-13 | 2020-04-28 | 浙江诺诺网络科技有限公司 | Full link testing method, device and system and computer readable storage medium |
| CN111078435A (en) * | 2019-12-18 | 2020-04-28 | 支付宝(杭州)信息技术有限公司 | Service processing method and device and electronic equipment |
| CN111159590A (en) * | 2019-12-31 | 2020-05-15 | 亚信科技（中国）有限公司 | Serial connection method and device based on front-end and back-end service call links |
| CN111224843A (en) * | 2019-12-31 | 2020-06-02 | 北京百度网讯科技有限公司 | Resource link monitoring method, device, equipment and storage medium |
| CN113452607A (en) * | 2020-03-24 | 2021-09-28 | 华为技术有限公司 | Distributed link acquisition method and device, computing equipment and storage medium |
| CN111770022A (en) * | 2020-06-28 | 2020-10-13 | 中国平安财产保险股份有限公司 | Link monitoring-based capacity expansion method, system, equipment and computer storage medium |
| CN111782534A (en) * | 2020-07-03 | 2020-10-16 | 普信恒业科技发展（北京）有限公司 | Business process monitoring method, device and system |
| CN112134739A (en) * | 2020-09-24 | 2020-12-25 | 平安科技（深圳）有限公司 | Link state tracking method and device, electronic equipment and computer storage medium |
| CN112348208A (en) * | 2020-10-13 | 2021-02-09 | 贵州电网有限责任公司 | System and method for monitoring metering and collecting key service full link |
| CN112486786A (en) * | 2020-11-12 | 2021-03-12 | 贝壳技术有限公司 | Calling link tracking method and device |
| CN112486786B (en) * | 2020-11-12 | 2022-08-09 | 贝壳技术有限公司 | Calling link tracking method and device |
| CN112395180A (en) * | 2020-12-10 | 2021-02-23 | 南威软件股份有限公司 | Method for realizing service log full link tracking |
| CN113760677A (en) * | 2021-02-02 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Abnormal link analysis method, device, equipment and storage medium |
| CN112965973A (en) * | 2021-02-10 | 2021-06-15 | 中国工商银行股份有限公司 | Distributed database monitoring method and device based on full link monitoring |
| CN112925721A (en) * | 2021-03-29 | 2021-06-08 | 建信金融科技有限责任公司 | Distributed system testing method and device |
| CN113114533A (en) * | 2021-04-08 | 2021-07-13 | 中国工商银行股份有限公司 | Network time consumption display method and device for distributed service invocation |
| CN113472850B (en) * | 2021-05-31 | 2023-05-09 | 北京达佳互联信息技术有限公司 | Link data acquisition method, device, system, electronic equipment and storage medium |
| CN113472850A (en) * | 2021-05-31 | 2021-10-01 | 北京达佳互联信息技术有限公司 | Link data acquisition method, device, system, electronic equipment and storage medium |
| CN115473839A (en) * | 2021-06-11 | 2022-12-13 | 北京字跳网络技术有限公司 | Data processing method, device and equipment based on buried point and storage medium |
| CN115473839B (en) * | 2021-06-11 | 2024-03-05 | 北京字跳网络技术有限公司 | Buried point-based data processing method, device, equipment and storage medium |
| CN113407511A (en) * | 2021-06-23 | 2021-09-17 | 中移（杭州）信息技术有限公司 | Log aggregation method, log aggregation equipment and computer program product |
| CN113537824A (en) * | 2021-08-03 | 2021-10-22 | 南方电网数字电网研究院有限公司 | Power grid data full link monitoring system |
| CN114143221A (en) * | 2021-11-29 | 2022-03-04 | 重庆富民银行股份有限公司 | Service asset management method, platform and storage medium based on full link monitoring |
| CN114331209A (en) * | 2022-01-10 | 2022-04-12 | 北京蓝莓时节科技有限公司 | Recommendation strategy data tracking method, device, system and storage medium |
| CN114756469A (en) * | 2022-04-24 | 2022-07-15 | 阿里巴巴（中国）有限公司 | Data relation analysis method and device and electronic equipment |
| CN114756469B (en) * | 2022-04-24 | 2024-07-16 | 阿里巴巴（中国）有限公司 | Data relationship analysis method and device and electronic equipment |
| CN115022213A (en) * | 2022-06-30 | 2022-09-06 | 福州畅昕信息科技有限公司 | Method for identifying request abnormity and storage medium |
| CN115022213B (en) * | 2022-06-30 | 2024-04-05 | 福州畅昕信息科技有限公司 | Method for identifying request abnormality and storage medium |
| CN116170344A (en) * | 2022-12-26 | 2023-05-26 | 海尔优家智能科技（北京）有限公司 | Message monitoring method and device, storage medium and electronic device |
| CN117170985B (en) * | 2023-11-02 | 2024-01-12 | 武汉大学 | Distributed monitoring method and system for open geographic information network service |
| CN117170985A (en) * | 2023-11-02 | 2023-12-05 | 武汉大学 | Distributed monitoring method and system for open geographic information network service |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108183927A (en) | The monitoring method and system that a kind of distributed system link calls | |
| US12067007B1 (en) | Analyzing a pipelined search to determine data on which to execute the pipelined search | |
| US11741089B1 (en) | Interactive location queries for raw machine data | |
| US11061918B2 (en) | Locating and categorizing data using inverted indexes | |
| US10205643B2 (en) | Systems and methods for monitoring and analyzing performance in a computer system with severity-state sorting | |
| US10469344B2 (en) | Systems and methods for monitoring and analyzing performance in a computer system with state distribution ring | |
| US11403333B2 (en) | User interface search tool for identifying and summarizing data | |
| US20200081891A1 (en) | Multi-phased execution of a search query | |
| US10528607B2 (en) | Syntax templates for coding | |
| CN106326381B (en) | HBase data retrieval method based on MapDB building | |
| US20200372007A1 (en) | Trace and span sampling and analysis for instrumented software | |
| US20170193016A1 (en) | Generation of a data model | |
| US20050234894A1 (en) | Techniques for maintaining collections of generated web forms that are hyperlinked by subject | |
| US11494395B2 (en) | Creating dashboards for viewing data in a data storage system based on natural language requests | |
| US11106713B2 (en) | Sampling data using inverted indexes in response to grouping selection | |
| US20170109639A1 (en) | General Model for Linking Between Nonconsecutively Performed Steps in Business Processes | |
| US10901811B2 (en) | Creating alerts associated with a data storage system based on natural language requests | |
| US10216862B1 (en) | Predictive estimation for ingestion, performance and utilization in a data indexing and query system | |
| CN111125068A (en) | Metadata management method and system | |
| US20190034499A1 (en) | Navigating hierarchical components based on an expansion recommendation machine learning model | |
| CN106096644B (en) | Method for measuring similarity and device based on probability workflow | |
| US20180018372A1 (en) | Semantic tagging of nodes | |
| US20030018584A1 (en) | System and method for analyzing transaction data | |
| US10409866B1 (en) | Systems and methods for occupation normalization at a job aggregator | |
| CN102902705B (en) | Ambiguity in location data |
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 | ||
| CB02 | Change of applicant information |
Address after: 100085 Floor 102-1, Building No. 35, West Second Banner Road, Haidian District, Beijing Applicant after: Seashell Housing (Beijing) Technology Co., Ltd. Address before: 100085 Floor 102-1, Building No. 35, West Second Banner Road, Haidian District, Beijing Applicant before: Chain home network (Beijing) Technology Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180619 |
|
| WW01 | Invention patent application withdrawn after publication |