CN112308328B - Top-Down network measurement system-oriented parallel measurement task optimization method and system - Google Patents
Top-Down network measurement system-oriented parallel measurement task optimization method and system Download PDFInfo
- Publication number
- CN112308328B CN112308328B CN202011236613.7A CN202011236613A CN112308328B CN 112308328 B CN112308328 B CN 112308328B CN 202011236613 A CN202011236613 A CN 202011236613A CN 112308328 B CN112308328 B CN 112308328B
- Authority
- CN
- China
- Prior art keywords
- query
- measurement
- source
- queries
- processing flow
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/12—Network monitoring probes
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Signal Processing (AREA)
- Entrepreneurship & Innovation (AREA)
- Computer Networks & Wireless Communication (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multi Processors (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention provides a method and a system for optimizing parallel measurement tasks of a Top-Down network measurement system, which aim at the problem of expansibility of the Top-Down network measurement system when the parallel measurement tasks are executed, and provides an optimization technology of multiplexing query results.
Description
Technical Field
The invention belongs to the technical field of computer network measurement, and particularly relates to a method and a system for optimizing parallel measurement tasks of a Top-Down network measurement system.
Background
Many network management tasks such as traffic engineering and network security require analysis of statistical information of network traffic, and thus rely on network measurements to provide real-time and accurate collection of statistical information of network traffic.
Existing network measurement systems can be divided into two main categories:
one type is a network measurement system of from-Up, which configures measurement rules to collect data according to measurement functions (such as port mirroring) that can be supported on each network device, and an upper layer measurement analysis task can only perform analysis processing according to the data that can be provided by the measurement rules, so as to obtain desired information. The method has the defects that as the configuration of the measurement rules is separated from the requirement of an upper-layer measurement task, the data collected by the data plane may not only contain a large amount of invalid data, but also have data omission, and the situation can lead to high data processing cost and low accuracy of the upper-layer measurement task, for example, a user wants to measure against SYN flooding attack, and network equipment only supports sFlow sampling, so that sFlow can collect a large amount of non-SYN useless data packets, the analysis difficulty of the measurement task is increased, and most SYN data packets are missed in the sampling process, so that the measurement result accuracy is low.
The other type is a Top-Down network measurement system, the measurement system directly takes the intention of an upper layer measurement task as input, translates the measurement intention into measurement rules executable by bottom layer data plane equipment through a series of compiling and optimizing, and controls the data plane to collect data according to the measurement rules, so as to accurately provide data for the upper layer measurement task. Top-Down network measurement system architecture such as
As shown in FIG. 1, the intent of a measurement task is expressed as a query through a declarative interface, compiled into executable rules by a compilation module, and distributed to network devices. The Query (Query) in the measurement system refers to a processing flow sequence formed by a basic processing flow predefined by the system, for example, in some Top-Down network measurement systems, data flow operations such as Filter and Map are used as the basic processing flow, and the Query is expressed as a data flow operation sequence.
Compared with two types of measuring systems, the Top-Down network measuring system can effectively avoid the problems of large and incomplete acquired information, high measuring expense and limited applicable scene of the Bottom-Up network measuring system.
The Top-Down network measurement system of the prior art also has problems. The Top-Down network measurement system collects according to the user measurement tasks as required, and when a plurality of parallel measurement tasks exist, data collection can be independently carried out for each measurement task. These tasks, although initiated by different users, are likely to be the same tasks, which can result in repeated measurements. In addition, the process of collecting data may be partially identical for different measurement tasks, which may also result in repeated measurements. Therefore, in a large network, for a large number of different measurement requirements from different users (such as network administrators, application developers, end users) and different analysis applications (such as attack detection, attack-defense efficacy evaluation, network performance analysis, etc.), independently performing data collection for each measurement requirement can cause a linear increase in measurement expense along with the number of measurement tasks, affecting the expansibility of the measurement system.
Disclosure of Invention
In order to solve the problems in the prior art, a method for optimizing parallel measurement tasks of a Top-Down network measurement system is provided, which comprises the following steps:
and 4, compiling the new query to generate rules or codes for deployment to the network equipment.
Preferably, the step 2 includes:
in step 22, when the sources of the two queries are the same and the common prefix exists in the basic process flow sequences of the two queries, the common prefix containing the basic process flow sequences is extracted as a single query.
Preferably, the step 2 further includes:
step 21, when the sources of the two queries are the same and the same basic processing flow exists, the same basic processing flow is moved to a common prefix without affecting the query function.
Preferably, the step 3 includes: and modifying the processing flow of each query to respectively utilize the output of the common processing flow as a source for serial connection to form a new query corresponding to the measurement task.
Preferably, the step 4 includes:
step 41, for each current query to be compiled, when the source of the current query is the output of another query, the source query is compiled first, and then the current query is compiled.
Preferably, the step 4 includes:
at step 42, code for examining the source query of the current query is generated for executing the code of the current query when the source query has completed.
Preferably, the step 4 includes:
step 43, compile each basic process flow constituting the current query into rules executable by the network device.
According to another aspect of the present invention, there is provided an optimization system for parallel measurement tasks of a Top-Down network measurement system, including: the system comprises a declaration type interface, an optimization module and a compiling module;
wherein the declarative interface is configured to express each measurement task as a query, each of the queries being composed of one or more basic process flows and including source attributes for specifying data to be measured of the query;
the optimizing module is connected with the declaration interface, extracts common processing flows in a plurality of queries based on the source attribute of the queries, wherein the common processing flows at least comprise a basic processing flow, and optimizes each query to form a new query by multiplexing the common processing flows;
the compiling module is connected with the optimizing module and used for compiling the new query and generating rules or codes for deployment to network equipment.
The present invention also provides a computer readable storage medium having stored thereon a computer program, wherein the program when executed by a processor performs the steps of any of the methods described above.
The invention also provides a computer device comprising a memory and a processor, on which memory a computer program is stored which can be run on the processor, characterized in that the processor implements the steps of any of the methods described above when executing the program.
The invention has the following characteristics and beneficial effects: by expanding the Top-Down measurement system, the system can optimize the query input by a user through a declaration interface, and unnecessary calculation can be reduced by multiplexing intermediate results when the system is executed on equipment, so that the problem of expansibility faced by the Top-Down network measurement system under the condition of a large number of parallel measurement tasks is solved well.
Drawings
Fig. 1 shows the architecture of a prior art Top-Down network measurement system.
Fig. 2 shows an example of a query expressed through a declarative interface in a prior art Top-Down network measurement system.
Fig. 3 shows an example of input and output of a prior art Top-Down network measurement system query.
Fig. 4 shows an example of a repeated process flow for different measurement intents of a prior art Top-Down network measurement system within the same time window.
FIG. 5 illustrates an example of the expression of a declarative interface for measurement intent after expansion according to one embodiment of the present invention.
FIG. 6 illustrates an example of a flow prior to query optimization based on exchange of a basic operational flow order, according to one embodiment of the invention.
FIG. 7 illustrates an example of a flow after query optimization exchanging a basic operational flow order in accordance with one embodiment of the present invention.
FIG. 8 illustrates an example of code after compilation by a compilation module, in accordance with an embodiment of the present invention.
Fig. 9 shows a flow of an implementation of a method for optimizing parallel measurement tasks for a Top-Down network measurement system according to an embodiment of the present invention.
Detailed Description
The invention will now be described with reference to the drawings and specific embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the prior art, top-Down network measurement system architecture such as
The working process of the device is as shown in fig. 1: the different measurement intents from the user or application are represented as three measurement tasks, respectively: measurement task 1, measurement task 2, measurement task 3, the three measurement tasks are expressed as three queries through the declarative interface: query 1, query 2 and query 3 are compiled into executable rules through a compiling module, and then the executable rules are deployed on network equipment such as a switch, a router and the like for data acquisition. The declarative interface is a query interface provided by the Top-Down network measurement system, taking the Sonata system as an example, the query interface of the system can express measurement tasks as short codes through five data operations of filt er and map, reduce, join, distinct, and two examples of querying by using the query interface provided by Sonata are given in FIG. 2.
The problems with prior art Top-Down network measurement systems can also be seen in fig. 1. The query 1 comprises a process flow 1, a process flow 2 and a process flow 3, the query 2 comprises a process flow 1, a process flow 2 and a process flow 4, and the query 3 comprises a process flow 1, a process flow 2, a process flow 1, a process flow 5 and a process flow 6, wherein the process flow 1, the process flow 2, the process flow 3, the process flow 5 and the process flow 6 respectively correspond to one basic process flow, such as filter, map, reduce in a Sonata system. It can be seen that the three queries, although generally completing different tasks, contain a portion of the same basic process flow, where query 1 and query 2 contain the same basic process flow 1 and basic process flow 2, and the three queries also contain a common basic process flow 1. These common basic process flows are not reused, but are performed multiple times, respectively, resulting in repeated measurements.
The Top-Down network measurement system in the prior art cannot solve the problem of repeated measurement, and the specific reason is that the Top-Down network measurement system compiles the query into an executable rule of the network device of the data plane, and the network device collects and submits the data in units of time windows. According to the input, the queries are classified into two types, one type is that only all data to be measured is taken as input (for example, all data packets entering the network device), and the other type is that all data to be measured and the output of the last time window of the query are taken as input.
Fig. 3 shows the input and output of a prior art Top-Down network measurement system query.
Three measurement tasks are included in fig. 3: measurement task 1, measurement task 2, measurement task 3, three measurement tasks correspond to three queries: query 1, query 2, query 3. The input and output of the three queries in time window 1 and time window 2 are different, wherein the input of query 1 and query 3 in time window 1 and time window 2 are both data to be measured, and the output of time window 1 is not input as time window 2, that is, the measurement result of the last time window is not used for the measurement of the next time window. The query 3 differs in that its input in time window 2 includes both the data to be measured and its output in time window 1, i.e. the measurement of query 3 in time window 2 takes advantage of the measurement results of the query in time window 1.
The two types of queries have a common problem that each query independently completes a corresponding measurement task, and the output of one query cannot be used as the input of the other query, that is, one query cannot utilize the measurement result of the other query. It is therefore difficult to optimize the repeated processing flow of data based on these queries for the purpose of different measurement tasks. Fig. 4 shows an example of a prior art Top-Down network measurement system in which different measurement intents have repeated processing flows within the same time window. Wherein measurement task 1 and measurement task 2 correspond to query 1 and query 2, and within time window 1 measurement task 1 and measurement task 2 contain the same basic process flow 1 and basic process flow 2, but the system can only repeatedly perform these two basic process flows for the data to be measured. Therefore, as the network scale expands and the number of different measurement demands in the network increases, the Top-Down measurement system faces an expansibility problem: the system would consume a lot of computing resources and memory space of the network device for each measurement intention to independently perform data collection, and the expense of network measurement would increase linearly with the number of measurement tasks.
In order to solve the above-mentioned problems, according to one embodiment of the present invention, there is provided a method for optimizing parallel measurement tasks for a Top-Down network measurement system, including the following steps.
And step 1, expanding a declaration interface of the Top-Down network measurement system.
As described above, in the Top-Down network measurement system of the prior art, the input of a certain query can only be the data to be measured and the output of the query in the previous time window, and different queries are completely independent, and one query cannot use the output result of another query, so it is difficult to provide optimization for the intention of the parallel measurement task. According to one embodiment of the invention, the declarative interface of the Top-Down network measurement system is first extended, maintaining the attribute "Source" for each query. The "source" refers to the data to be measured specified by the system or the user in the query, and defaults to all the data to be measured on the network device, and can also specify the output of other queries as the input of the query. As shown in fig. 5, query 1 is a case where the user does not particularly declare the source, and it regards data to be measured in the network device as its own data to be measured, and the sources of query 2 and query 3 are query 1, because these two queries take the output of query 1 of the current time window as the data to be measured.
By extension as above, a user or system can concatenate different queries such that multiple concatenated queries together express the intent of a measurement task, i.e., the measurement intent may be made up of one or more query requests. The two measurement intents can avoid repeated calculations in the data acquisition process by multiplexing the results of certain queries. FIG. 5 shows the declarative interface representation of measurement intent after expansion, containing three queries: query 1, query 2, query 3. The "source" attribute of query 1 is "raw data to be processed", meaning that its input is data to be measured in the network device, and its processing flow includes basic processing flow 1 and basic processing flow 2, whose outputs are input for query 2 and query 3. The "source" attribute of query 2 is "query 1," meaning that its input is the output result of query 1, which includes only basic process flow 3. The "source" attribute of query 3 is also "query 1," meaning that its input is the output result of query 1, which includes only process flow 4. Query 1 and query 2 are concatenated together to represent one measurement task, query 1 and query 3 are concatenated together to represent another measurement task, so that the two measurement tasks multiplex the output of query 1, thereby avoiding repeated execution of query 1 processing flow on the network device.
The technical effect of the adoption of the step 1 is that the query is not independent by the expanded declarative interface, and different measurement tasks can be optimized by multiplexing the query results, which is the basis for optimizing the parallel multi-measurement tasks.
And 2, optimizing the parallel input intention.
After expanding the declaration interface, the intentions of different measurement tasks can be optimized by multiplexing some query results, and because the measurement tasks in a large network come from a large number of different users, the parallel measurement tasks have large intentions, and the optimization of the query by relying on manpower alone is not feasible, so that the measurement system is required to automatically optimize the query of the measurement tasks input by all users. The query input by the user through the declarative interface is composed of basic processing flows, and according to one embodiment of the invention, the intent optimization processing method for parallel input is based on the basic processing flows, and the steps are as follows:
if the data to be processed (i.e., the source of the query) of the two queries are identical and they have the same basic process flow, then consider that a pair of identical basic process flows is moved to a common prefix if it can be moved to a portion of the common prefix of the two query basic process flow sequences without affecting the query function, step 21. According to one embodiment of the present invention, as shown in fig. 6, measurement task 1 and measurement task 2 correspond to query 1 and query 2, and both queries contain the same basic process flow 1 and basic process flow 4, where basic process flow 1 is originally a common prefix for both queries, and no movement is required. In query 1, basic process flow 4 is moved to before basic process flow 2, and is located after basic process flow 1, provided that the order in which basic process flow 4 and basic process flow 2 are exchanged does not affect the functionality of query 1. In query 2, process flow 4 is moved to process flow 3 before and after process flow 1, provided that the order of exchanging process flow 4 and process flow 3 also does not affect the function of query 2. Thus, process flow 1 and process flow 4 together form a common prefix for query 1 and query 2, the result of which is shown in FIG. 7.
If the data to be processed (i.e., the source of the query) of the two queries are identical and their basic process flow sequences have a common prefix, then the common prefix of the basic process flow sequences may be extracted as a single query, multiplexing the results of the query, step 22. For example, query 1 and query 2 in FIG. 4 have the same inputs and the same basic process flow 1 and process flow 2, that is, have a common basic process flow sequence, i.e., common prefix < basic process flow 1, basic process flow 2>, whereupon the system automatically extracts the basic process flow sequence as a single query node, resulting in the optimization result shown in FIG. 5. FIG. 5 shows the declarative interface representation of measurement intent after expansion, containing three queries: query 1, query 2, query 3. The "source" attribute of query 1 is "raw data to be processed", meaning that its input is data to be measured in the network device, and its processing flow includes basic processing flow 1 and basic processing flow 2, whose outputs are input for query 2 and query 3. The "source" attribute of query 2 is "query 1," meaning that its input is the output result of query 1, which includes only basic process flow 3. The "source" attribute of query 3 is also "query 1", meaning that its input is the output result of query 1, which includes only basic process flow 4. Query 1 and query 2 are concatenated together to represent one measurement task, query 1 and query 3 are concatenated together to represent another measurement task, so that the two measurement tasks multiplex the output of query 1, thereby avoiding repeated execution of query 1 processing flow on the network device.
Two different queries can be optimized by steps 21 and 22 so that the intent of each measurement task can be expressed as one or more queries, thus optimizing the queries can accomplish optimization of parallel measurement tasks. After receiving the query input by the user through the declaration interface, the Top-Down network measurement system needs to optimize all the queries according to the method, and then hands the optimized query to the compiling module.
The technical effect of the step 2 is that the query result is multiplexed by extracting the common prefix of the basic processing flow of the parallel measurement task as a single query, thereby realizing the automatic optimization of the intention of the parallel measurement task.
And 3, compiling the parallel measurement task optimized in the step 2.
In order to optimize the intention of the parallel measurement task, step 1 expands the declarative interface of the Top-Down network measurement system, maintains the attribute of "source" for each query, redefines the data to be measured of the query as the output of other queries, and the expanded attribute needs to be supported by the compiling module of the Top-Down network measurement system to be correctly compiled into the device for execution, thus the compiling module of the Top-Down network measurement system needs to be designed. The Top-Down network measurement system may compile and deliver queries to switches, end hosts, or controllers in the network, where the worst programmability is the switch, how to compile queries into rules on programmable switches, and implementing the designation of the "source" of queries is a major consideration in designing compilation modules.
According to one embodiment of the invention, for each query entered into the compilation module, the compilation module compiles by:
step 31, the source of the query is checked, if the source is a default value, then step 32 is directly entered, if the source is the output of another query, then it is checked whether the source query has been compiled into code, if the source query has not been compiled, then it is necessary to compile the source query first, and then continue compiling the current query. This is to ensure that when the code of any one query is executed, the code of its source has been executed, thereby knowing the output of the source query as input to the current query.
Step 32, a piece of code is generated for checking whether the source query of the current query is completed, and the code of the current query is executed only after the source query is completed.
Each basic process flow constituting the current query is compiled 33 into code executable by the network device, which according to one embodiment of the invention comprises a programmable switch and a programmable router.
The design ensures that when data acquisition is performed on the equipment, the data packet for inputting the query can be specified according to the source of the query, so that the acquisition process on the equipment is consistent with the query. Taking the query in fig. 5 as an example, the code after compiling using the extended compiling module is shown in fig. 8, where, as the source of query 2 and query 3, the code of query 1 needs to precede query 2 and query 3, and before executing the codes of query 2 and query 3, it is determined whether the data packet already meets the requirement of query 1, that is, it is determined that query 1 has been completed and the data packet is output.
The technical effect of the step 3 is that the query after the optimization of the parallel measurement task intention can be correctly executed on the network equipment to realize the data acquisition meeting the user input measurement intention.
Fig. 9 shows an example of an implementation flow of a method for optimizing parallel measurement tasks for a Top-Down network measurement system according to the present invention. The system comprises three measurement tasks, namely a measurement task 1, a measurement task 2 and a measurement task 3. The three tasks are expressed as three queries, query 1, query 2, query 3, respectively, through the declarative interface. Each inquiry comprises a respective basic processing flow, and the inquiry 1 comprises a basic processing flow 1, a basic processing flow 2 and a basic processing flow 3; the query 2 comprises a basic processing flow 1, a basic processing flow 2 and a basic processing flow 4; query 3 contains basic process flow 1, basic process flow 5, basic process flow 6. The three queries are optimized by the optimization module based on the public processing flow of the invention, the processing flow of each query is modified, and the output of the public processing flow is used as a source for serial connection to form a new query corresponding to the measurement task. After optimization, 5 queries, query a, query b, query c, query d, query e, were generated. The optimized query contains source attributes for specifying the input of the query. Specifically, the source of query a is all the data to be tested, which contains basic process flow 1; the source of query b is the output of query a, so it contains not only basic process flow 2, but also basic process flow 1 of query a; the source of the query c is the output of the query b, which not only comprises the basic processing flow 3, but also comprises the basic processing flow 1 and the basic processing flow 2 of the query b, namely the query c completes the measurement task of the query 1; the source of the query d is the output of the query b, which not only comprises the basic processing flow 4, but also comprises the basic processing flow 1 and the basic processing flow 2 of the query b, namely the query d completes the measurement task of the query 2; the source of query e is the output of query a, which includes not only basic process flow 5 and basic process flow 6, but also basic process flow 1 of query a, i.e., query e completes the measurement task of query 3 described above. It can be seen that compared with the prior art of fig. 1, the present invention avoids that different queries repeatedly perform the same process flow by referencing the output results of the common process flow.
The invention also provides an optimization system of parallel measurement tasks facing to the Top-Down network measurement system, comprising: the system comprises a declaration type interface, an optimization module and a compiling module;
wherein the declarative interface is configured to express each measurement task as a query, each of the queries being composed of one or more basic process flows and including source attributes for specifying data to be measured of the query;
the optimizing module is connected with the declaration interface, extracts common processing flows in a plurality of queries based on the source attribute of the queries, wherein the common processing flows at least comprise a basic processing flow, and optimizes each query to form a new query by multiplexing the common processing flows;
the compiling module is connected with the optimizing module and used for compiling the new query and generating rules or codes for deployment to network equipment.
In general, in the practical application process, the invention needs to expand the declaration interface and compile the module of the Top-Down network measurement system, and adds an optimizing module for optimizing the query according to the public basic processing flow between the declaration interface and the compile module. The user expresses the intentions of various measurement tasks as queries through a declaration interface, and the Top-Down network measurement system firstly gives the queries to an optimization module to be optimized uniformly, and the queries obtained after the optimization are compiled to each device by an compiling module to be executed. The invention can effectively avoid the problem that some data processing flows of the Top-Down network measurement system are repeatedly performed for many times under the condition of parallel multi-measurement intention, reduces the expenditure of the parallel multi-measurement intention, and enhances the expandability of the system.
It should be noted and appreciated that various modifications and improvements of the invention described in detail above can be made without departing from the spirit and scope of the invention as claimed in the appended claims. Accordingly, the scope of the claimed subject matter is not limited by any particular exemplary teachings presented.
Claims (8)
1. A method for optimizing parallel measurement tasks of a Top-Down network measurement system comprises the following steps:
step 1, expressing each measurement task as a query, wherein each query consists of one or more basic processing flows and comprises source attributes for specifying data to be measured of the query;
step 2, extracting a plurality of public processing flows in the query based on the source attribute of the query, wherein the public processing flows at least comprise a basic processing flow;
wherein, the step 2 further comprises:
step 21, when the sources of two queries are the same and the same basic processing flow exists, the same basic processing flow is moved to a common prefix on the premise of not affecting the query function; and
step 22, when the sources of the two queries are the same and the common prefix exists in the basic processing flow sequences of the two queries, extracting the common prefix containing the basic processing flow sequences as a single query;
step 3, optimizing each query to form a new query by multiplexing the public processing flow;
and 4, compiling the new query to generate rules or codes for deployment to the network equipment.
2. The method of claim 1, said step 3 comprising: and modifying the processing flow of each query to respectively utilize the output of the common processing flow as a source for serial connection to form a new query corresponding to the measurement task.
3. The method of claim 1, said step 4 comprising:
step 41, for each current query to be compiled, when the source of the current query is the output of another query, the source query is compiled first, and then the current query is compiled.
4. The method of claim 1, said step 4 comprising:
at step 42, code for examining the source query of the current query is generated for executing the code of the current query when the source query has completed.
5. The method of claim 1, said step 4 comprising:
step 43, compile each basic process flow constituting the current query into rules executable by the network device.
6. An optimization system for parallel measurement tasks of a Top-Down oriented network measurement system for a method according to any of claims 1-5, comprising: the system comprises a declaration type interface, an optimization module and a compiling module;
wherein the declarative interface is configured to express each measurement task as a query, each of the queries being composed of one or more basic process flows and including source attributes for specifying data to be measured of the query;
the optimizing module is connected with the declaration interface, extracts common processing flows in a plurality of queries based on the source attribute of the queries, wherein the common processing flows at least comprise a basic processing flow, and optimizes each query to form a new query by multiplexing the common processing flows;
the compiling module is connected with the optimizing module and used for compiling the new query and generating rules or codes for deployment to network equipment.
7. A computer readable storage medium having stored thereon a computer program, wherein the program when executed by a processor realizes the steps of the method according to any of claims 1 to 5.
8. A computer device comprising a memory and a processor, on which memory a computer program is stored which can be run on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 5 when the program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011236613.7A CN112308328B (en) | 2020-11-09 | 2020-11-09 | Top-Down network measurement system-oriented parallel measurement task optimization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011236613.7A CN112308328B (en) | 2020-11-09 | 2020-11-09 | Top-Down network measurement system-oriented parallel measurement task optimization method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112308328A CN112308328A (en) | 2021-02-02 |
| CN112308328B true CN112308328B (en) | 2023-06-06 |
Family
ID=74325179
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011236613.7A Active CN112308328B (en) | 2020-11-09 | 2020-11-09 | Top-Down network measurement system-oriented parallel measurement task optimization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112308328B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547757A (en) * | 2010-12-31 | 2012-07-04 | 北京亿阳信通软件研究院有限公司 | Network optimization analysis system and method |
| CN105701128A (en) * | 2014-11-28 | 2016-06-22 | 华为技术有限公司 | Query statement optimization method and apparatus |
| CN107368517A (en) * | 2017-06-02 | 2017-11-21 | 上海恺英网络科技有限公司 | A kind of method and apparatus of high amount of traffic inquiry |
| CN108920110A (en) * | 2018-07-25 | 2018-11-30 | 湖南机电职业技术学院 | A kind of parallel processing big data storage system and method calculating mode based on memory |
-
2020
- 2020-11-09 CN CN202011236613.7A patent/CN112308328B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547757A (en) * | 2010-12-31 | 2012-07-04 | 北京亿阳信通软件研究院有限公司 | Network optimization analysis system and method |
| CN105701128A (en) * | 2014-11-28 | 2016-06-22 | 华为技术有限公司 | Query statement optimization method and apparatus |
| CN107368517A (en) * | 2017-06-02 | 2017-11-21 | 上海恺英网络科技有限公司 | A kind of method and apparatus of high amount of traffic inquiry |
| CN108920110A (en) * | 2018-07-25 | 2018-11-30 | 湖南机电职业技术学院 | A kind of parallel processing big data storage system and method calculating mode based on memory |
Non-Patent Citations (1)
| Title |
|---|
| 软件定义网络的测量方法研究;戴冕;程光;周余阳;;软件学报(06);第 283-304页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112308328A (en) | 2021-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10044575B1 (en) | System for organizing and fast searching of massive amounts of data | |
| Gao et al. | Continuous pattern detection over billion-edge graph using distributed framework | |
| US20170102986A1 (en) | Automatic correlation of dynamic system events within computing devices | |
| US20170031742A1 (en) | Causal engine and correlation engine based log analyzer | |
| CN107832407B (en) | Information processing method and device for generating knowledge graph and readable storage medium | |
| CN102938708B (en) | Based on alarm correlation analysis system and the analytical method thereof of alarm propagation pattern | |
| CN103188112B (en) | Network flow detection method and device | |
| CN102929613B (en) | The tuning apparatus and method of operating system | |
| KR101965277B1 (en) | System and method for analysis of hypergraph data and computer program for the same | |
| CN105022691A (en) | Highly automatic software testing method based on UML diagram | |
| JP6060312B2 (en) | Data organization and high-speed search | |
| WO2018095539A1 (en) | Efficient data propagation in a computer network | |
| CN112308328B (en) | Top-Down network measurement system-oriented parallel measurement task optimization method and system | |
| CN104836686B (en) | A kind of method and device for the configuration file generating server | |
| CN111628888B (en) | Fault diagnosis method, device, equipment and computer storage medium | |
| CN102789420A (en) | Dynamic slicing system based on execution tract of program | |
| CN116301755A (en) | Automatic batch flow data marking framework construction method based on directed calculation graph | |
| CN113162801B (en) | Alarm analysis method, device and storage medium | |
| CN115629883A (en) | Resource prediction method, resource prediction device, computer equipment and storage medium | |
| CN114817300A (en) | Log query method based on SQL (structured query language) statements and application thereof | |
| CN104516916A (en) | Method and device for analyzing network report incidence relation | |
| Showerman et al. | Large scale system monitoring and analysis on blue waters using ovis | |
| CN108780452A (en) | A kind of storing process processing method and processing device | |
| CN104699599A (en) | Interprocedural static slice extracting method based on idUCf quinary structure | |
| CN114844784B (en) | Topology modeling method of reconfigurable scanning network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |