CN104537091A - Networked relational data query method based on hierarchical identification routing - Google Patents
Networked relational data query method based on hierarchical identification routing Download PDFInfo
- Publication number
- CN104537091A CN104537091A CN201510003869.6A CN201510003869A CN104537091A CN 104537091 A CN104537091 A CN 104537091A CN 201510003869 A CN201510003869 A CN 201510003869A CN 104537091 A CN104537091 A CN 104537091A
- Authority
- CN
- China
- Prior art keywords
- inquiry
- data
- query
- routing
- name
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/288—Entity relationship models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a networked relational data query method based on hierarchical identification routing. The method comprises the steps that (1) hierarchical names of data entities are constructed in local relational databases; (2) the hierarchical names are converted into query groups formed by hierarchical routing factors and query constraint factors according to information to be queried by a user, and then the query groups are sent to a routing node connected with a client side; (3) the node matches and forwards all the received query groups; (4) information of the query groups is converted into an SQL supporting the local relational databases to carry out local query; (5) a local query result and the hierarchical routing factors of an original query request are packaged into data groups, and the data groups are sent back to interfaces of the corresponding query groups; (6) routing nodes cache and aggregate the received data groups and then send the data groups back along original paths. By means of the method, SQL query of all relational databases connected through an NDN can be achieved.
Description
Technical field
The present invention relates to relational database, self-organizing network environment, named data network field, relate in particular to a kind of networking relation data querying method based on level identities route.
Background technology
All there is the complex query of DYNAMIC DISTRIBUTION data and the difficulty of acquisition in self-organization P2P, radio sensing network WSN, mobile Internet etc., because correct, the effective routing problem of inquiry request under dynamic and distributed environment should be considered, also to consider the dynamic aggregation of distributed result and reliably return problem.Wherein dynamic routing can affect the recall ratio of Query Result, precision ratio and data transmission efficiency.In addition, Query Result dynamic aggregation and reliably return availability, the reliability that can affect Query Result.And the personalized acquisition capability of the form that under dynamic and distributed environment, inquiry system is supported to DYNAMIC DISTRIBUTION data has material impact.For ease of better open inquiry with share, structuring or semi-structured description are carried out to the data object of DYNAMIC DISTRIBUTION, set up the metadata format of specification and structuring or semi-structured data storehouse and there is important effect.
In order to effectively obtain the DYNAMIC DISTRIBUTION content blocks in internetwork environment, CCN/NDN network (i.e. Content-Centric Network/Named Data Networking) proposes a kind of method for routing (we are referred to as level identities route) obtaining DYNAMIC DISTRIBUTION content blocks based on layer naming mark.In this method for routing, the name of route not only identifies level belonging to content and name, the more important thing is that replacement IP address becomes the elementary cell in route processing.The route implementing mechanism of its core sets up the forwarding tree of form for (stratification prefix-interface) in a network according to the stratification name from all data sources.Therefore, this routing mechanism and machine location (comprising IP address) have nothing to do, and can realize the asynchronous multicast of user's request, and the transmission of its data is more effective, because the path that its advertising of route meeting prioritizing selection is the fastest.
But, in the face of extensive, dynamically and the relation data of distribution or other structural data or semi-structured data, complex query should adapt to route change and data mobile, and various forms of inquiry should be able to the route of intelligence in a uniform manner.And system should not need the distribution and variation situation knowing data in advance during structure query statement; For reducing the computation burden of data source and improving search efficiency, when inquiry is forwarded to related data sources by communication network, this inquiry should not need to check the distribution characteristics of data and carry out query decomposition, namely the distribution of large-scale data and dynamic change situation can to inquiring user and upper layer application transparent.The current querying method achieving clean culture or multicast pattern, mainly based on address maps mode, is difficult to obtain in time the data of extensive, distribution and reaches higher data transmission efficiency when the memory address change of network topology or data.
Summary of the invention
For the above-mentioned technical matters that prior art exists, data that are extensive, that distribute can be obtained in time when the object of this invention is to provide the change of a kind of memory address when network topology or data and there is the networking relation data querying method based on level identities route of higher data transmission efficiency.
For achieving the above object, a kind of networking relation data querying method based on level identities route of the present invention, comprises the following steps:
(1) relation of inclusion in each local relational database between identification data entity, construct the stratification name of each data entity, and by the name of each stratification or prefix advertisement in the forwarding information table (referred to as fib table) in each routing node of named data network (referred to as NDN network);
(2) data category inquired about according to user for each inquiry request is converted into other constraint condition the inquiry packets (referred to as NDNQL) be made up of Tiered routing Summing Factor inquiry constraint factor with reference to the stratification name in forwarding information table, be sent to client's side link routing node;
(3) each routing node is to each inquiry packets received, the matched routings algorithm of named data network is used the Tiered routing factor in inquiry packets and each in forwarding information table to be carried out mating and forwarding, the information of the inquiry packets that registration forwards from each interface in the request scratch list (referred to as PIT table) of named data network;
(4) after the information of inquiry packets arrives relevant database node, the information of inquiry packets is converted to the SQL query statement supporting local relational database, carries out local search;
(5) local search result (referred to as LRS) adds that the Tiered routing factor of former inquiry request is packaged into packet (referred to as NDNRS), be then recycled to receive corresponding inquiry packets interface on;
(6) each routing node carries out buffer memory, polymerization and former road to the packet received and returns (user obtains result referred to as RS).
Described inquiry packets comprises following items: stratification name or prefix, inquiry constraint condition, return form, other condition, security parameter, not tuple, wherein stratification name or prefix, the form of returning can be more than one.
Described packet comprises following items: stratification name or prefix, return results, security parameter and timestamp, and wherein returning results can the form tissue of one dimension, two-dimensional array or chained list.
SQL query statement destructing in described step (4) is the inquiry packets be made up of Tiered routing Summing Factor inquiry constraint factor.
What the Tiered routing factor in described step (4) and each in forwarding information table were carried out mates is longest prefix match, by inquiry packets through repeatedly routing forwarding, delivers in relevant local data storehouse.
The present invention can solve the SQL query problem of all relational databases connected by NDN network, consolidation form is adopted during use, use an inquiry packets can be routed to all Relational database nodes, each database node can be easy to this inquiry packets is converted to the SQL statement supporting local search.
When one aspect of the present invention can solve and carry out relation data inquiry under NDN network environment, returning and aggregation problem of high usage route problem in DYNAMIC DISTRIBUTION network environment of the SQL query request of user and Query Result thereof; On the other hand, only need to extract the stratification taxonomy and nomenclature information of each data entity and be communicated to routing node, just can be easy to that each local data base is issued and share content, as long as user can find the content of issue according to content distributed multiclass classification (or adding name information).
The present invention (as non-structural P 2 P network, WSN and other self-organizing network) can provide than existing querying method as better inquiry recall rate and the degree of accuracy such as probabilistic query method, the querying method based on mark-address maps, the querying method based on semanteme-address maps under the network environment of DYNAMIC DISTRIBUTION.Query object of the present invention is not limited to relation data, structural data or semi-structured data in DYNAMIC DISTRIBUTION network, can also be all digitizing objects that can carry out stratification taxonomy and nomenclature to it.
Accompanying drawing explanation
Fig. 1 is the data structure diagram of inquiry packets in the present invention;
Fig. 2 is the data structure diagram of packet in the present invention;
Fig. 3 is the SQL query system diagram based on NDN network in the present invention;
Fig. 4 is the process flow diagram extracting stratification name information in the present invention from relational database;
Fig. 5 is the map example figure of SQL query statement to inquiry packets of single concept (table) in the present invention;
Fig. 6 is the map example figure of the relevant connection (join) of two concepts (table) in the present invention SQL statement of inquiring about to inquiry packets;
Fig. 7 is the map example figure of the relevant associating (union) of two concepts (table) in the present invention SQL statement of inquiring about to inquiry packets;
Fig. 8 is that the relevant of two concepts (table) in the present invention intersects the map example figure of (intersect) SQL statement of inquiring about to inquiry packets;
Fig. 9 is the map example figure of relevant nested (nested) of two concepts (table) in the present invention SQL statement of inquiring about to inquiry packets;
Figure 10 is the query processing process flow diagram of the inquiry packets of single concept (table) in the present invention;
Figure 11 is the query processing process flow diagram of join, union, intersect, except inquiry packets of multiple concept (table) in the present invention;
Figure 12 is the query processing process flow diagram of the nested inquiry packets of multiple concept (table) in the present invention;
Figure 13 is the step-by-step polymerization computing process flow diagram of aggregate query avg in the present invention.
Embodiment
The following examples just for describing the present invention in detail, and limit the protection domain of invention never in any form.
In relational database, metadata mainly defines in table form, is generally span and other constraint of statement row.Each form (sometimes referred to as a relation) comprises the one or more data attribute shown with list.Often row comprises a unique data entity.In relational database, syndeton between table (as associating of major key and external key) not only specifies the Data relationship between form, also often specifies the hierarchical relationship between kind belonging to data or classification.Even and if these hierarchical information General Requirements data also can be existed by vertical division in each burst.Therefore, in a given application, no matter the relational database of foundation is centralized or distributed database, in fact all can comprise following several main information:
(1) things (or concept): as books, people, commodity, website, geography information etc.;
(2) classification foundation or taxonomic hierarchies: limiting the classification range wanting query object, is generally obtain by drawing with different levels attribute;
(3) instance name: the given title of user, may bear the same name, and can pass through classification, property value is distinguished;
(4) other attribute-value.
Usually, the meaning represented by each clause in the SQL statement of " Select-From-Where " version is:
" Select list< attribute-name > ": this clause states the information that will return, mainly each property value of certain class object;
" From table< table name > ": this clause specifies the concept belonging to information (concept definition one class object, substantially equivalent table name) that will search to gather category;
" Where<expression list> ": this clause provides the constraint condition of Query Result, as limited the value range (as time, author etc.) of affiliated classification, partial name (uncertain part asterisk wildcard represents) and other attribute;
" Group by<name> ": this clause to return data classify (group) integrate;
" Having< conditional expression > ": to returning according to condition expression formula of dividing into groups, this clause requires that carrying out filtration calculates (generally needing to be deferred to last aggregation node to carry out).
Such as, store a lot of e-book in a NDN network, then for inquiry request " book that network engineering specialty is published for nearly 3 years ", the SQL expression-form of its inquiry is:
Query=Select* (title, author, publishing house, Publication Year)
From books table
Where subject=computing machine and specialty=network engineering and (2013-Year (publication time)) <=3
Above-mentioned main information can be extracted from this SQL query, as:
As shown in Figure 1, be the data structure of inquiry packets used in this invention.The data structure of inquiry packets of the present invention can comprise following items: stratification name or prefix, and inquiry constraint condition, returns form, security parameter, other condition, not tuple.Wherein stratification name or prefix determine according to the list item in user's request and routing node FIB, which specify the hierarchical classification belonging to the data entity involved by inquiry, even specifies its name or partial name (as represented by asterisk wildcard).Inquiry constraint condition refers to that the scope limited in stratification name or prefix retrains further and returns results the required condition met, as specified the value range of other attribute demand fulfillment further.Can specify in multiple table in a SQL query and inquire about, therefore can have many levels name or prefix in an inquiry packets, show that Query Result can from many levels classification.The form of returning indicates some property value returned results by data entity and forms, or divides into groups by certain condition, or some statistical value etc.Constraint condition and return form also can be more than one, the form returned results with the constraint condition indicated when inquiring about different table and needing.Security parameter is for carrying user identity, digest information etc., and to judge authority that user accesses, inquiry packets is not tampered, forgery etc.Other condition is as the type in specific data source or the restriction forwarding jumping figure, and whether appointment is polymerized returns results, etc.Tuple can not be random number or timestamp, can distinguish to make routing node or data server the new and old inquiry packets received.
As shown in Figure 2, be the data structure of packet used in this invention.The data structure of packet of the present invention can comprise following items: stratification name or prefix, return results, security parameter and timestamp.Stratification name in packet or prefix are consistent with the stratification name in corresponding inquiry packets or prefix.Return results is meet querying condition in relational database and according to the content returning the substantive requirements of form and return.Returning results can with the form tissue of one dimension, two-dimensional array or chained list.Security parameter is for the identification of data set provider identity, and verifies to return results whether be tampered and forge.The time of return that timestamp essential record returns results, whether routing node or user can be used for judged result expired.
In order to realize the query processing of above-mentioned inquiry packets in NDN network, the primary structure that its system realizes as shown in Figure 3.Based on this system architecture, roughly realization approach is:
First the stratigraphic classification (hierarchicaltaxonomy) of some data entity in relational database is extracted, and the name of the upper solid data of splicing, then this layer naming information is diffused in routing node as routing index, sets up the routing index (being stored in FIB) of named data network.Because the data acquisition request of CCN/NDN can be routed in correct data source by using hierarchical identifier, then inquiry request also can be routed in correct data source by hierarchical identifier.
Concrete inquiry request describing method obtains the definition of asking to named data suitably to expand.Its core concept is: based on FIB and user's request, therefrom extract associated level identities information, and carry other property value constraint condition, formation can be routed the inquiry packets (NDNQL) of forwarding in NDN, then by the longest prefix match (LPM) with routing index, by its Multicast Routing in correct data source, also the main information of NDNQL and the routing interface of process are recorded in PIT table in routing forwarding process; When NDNQL arrives after in data source, be converted into the inquiry of SQL form, then carry out the inquiry of localized relation data, obtain the relevant tuple and property value of meeting consumers' demand.Return results to be encapsulated as packet (being referred to as NDNRS) that applicable former road returns and to utilize the PIT of NDN to show information and former for NDNRS road is returned.Finally, from NDNRS, all relevant tuple and property value etc. of meeting consumers' demand is extracted by client.
Query processing process is substantially identical with the processing procedure that distributed relational database is inquired about, and key step comprises: localization inquiry, result aggregator and returning.But consider the dynamic change of data, and not by IP address route and locator data, therefore concrete Query Processing Algorithm is completely different.
Fig. 4 shows the flow process extracting stratification name information from relational database.Table definition in the database various attribute informations of data object, comprising the data entity attribute information with multilayer classification, and also have inclusive relation between these stratification classifications.Therefore, if can extract as the index in level identities route using the stratigraphic classification information of these attributes, then the relation data inquiry in named data network can be realized.Detailed process is: the stratigraphic classification (hierarchicaltaxonomy) first extracting some data entity in relational database, and the name of the upper solid data of splicing, then this layer naming information is diffused in routing node as routing index, sets up the routing index (FIB) of named data network.Around stratification identification index, associated stratification classification information is extracted from user's request, looked into data class level even data entity title is comprised in making to call request, namely level identities is comprised, and carry other property value constraint condition, form the inquiry packets (NDNQL) that can be routed forwarding in NDN.
The inquiry packets relating to single concept only has a kind of stratification name or prefix.It generally only relates to a table in basic SQL query statement.Fig. 5 show relate to single concept (table) SQL query statement to the map example of inquiry packets.Such as: the aggregate query gathering the sales situation of each commodity (ProductID) according to region (AreaID).Its SQL query statement is:
Select AreaID,ProductID,Sum(Total)
From CW_Orderdetail
Where ostate＝1
Group By AreaID,ProductID With cube
This SQL query statement can be exchanged into the structuralized query of NDN, and its 3 main category informations are respectively:
When relating to multiple concept (i.e. multiple table) in a structure query, this multiple concept may have also may not have direct this locality (local) association.When not having directly local association, then need inquiry request to be sent in the data source relevant to these concepts (table) and carry out local queries, then returning results of each distribution is carried out being polymerized and returning to user.Therefore need the level identities using each concept in routing forwarding, inquiry request is routed in the data source at each concept place.Fig. 6 shows SQL statement that the relevant connection (join) that relates to two concepts (table) the inquires about map example to inquiry packets.Connection inquiring (join) as following SQL form:
Select customer-name,borrower.loan-number,amount
From borrower,loan
Where borrower.loan-number＝loan.loan-number
This SQL query statement can be converted to the structured form being suitable for level identities route:
| Type | Information |
| Stratification name 1 | The level identities of " borrower " |
| Return form 1 | customer-name,loan-number |
| Stratification name 2 | The level identities of " loan " |
| Return form 2 | loan-number,amount |
| Constraint condition (aggregation) | borrower.loan-number＝loan.loan-number |
Fig. 7 shows SQL statement that the relevant associating (union) that relates to two concepts (table) the inquires about map example to inquiry packets.Inquire about as also (union) for following SQL form (need to carry out local union query processing on a single node and carry out overall union query processing in whole NDN network):
(Select customer-name From depositor)union(Selectcustomer-name From borrower)
This SQL query statement can be converted to the structured form being suitable for level identities route, for:
| Type | Information |
| Stratification name 1 | The level identities of " depositor " |
| Return form 1 | customer-name |
| Stratification name 2 | The level identities of " borrower " |
| Return form 2 | customer-name |
| Constraint condition | The union process of local or the overall situation |
Fig. 8 shows and relates to the relevant of two concepts (table) and intersect the map example of (intersect) SQL statement of inquiring about to inquiry packets.As the friendship (Intersect) for following SQL form is inquired about: (need statement only to carry out local Intersect query processing on a single node or carry out global I ntersect query processing in whole NDN network, be expressed as Intersect
lor Intersect
g)
(Select customer-name From depositor)Intersect
G(Selectcustomer-name From borrower)
This SQL query statement can be converted to the structured form being suitable for level identities route, for:
| Type | Information |
| Stratification name 1 | The level identities of " depositor " |
| Return form 1 | customer-name |
| Stratification name 2 | The level identities of " borrower " |
| Return form 2 | customer-name |
| Constraint condition | Intersect G |
The SQL statement of inquiring about for relevant different (except) that relate to two concepts (table) is to the map example of inquiry packets, and its transition form is similar to Fig. 8 equally.As the difference (Except) for following SQL form is inquired about: (need statement only to carry out local Except query processing on a single node or carry out overall Except query processing in whole NDN network, be expressed as Except
lor Except
g)
(Select customer-name From depositor)Except
G(Selectcustomer-name From borrower)
This SQL query statement can be converted to the structured form being suitable for level identities route, for:
| Type | Information |
| Stratification name 1 | The level identities of " depositor " |
| Return form 1 | customer-name |
| Stratification name 2 | The level identities of " borrower " |
| Return form 2 | customer-name |
| Constraint condition | Except G |
Fig. 9 shows SQL statement that relevant nested (nested) that relate to two concepts (table) the inquire about map example to inquiry packets.Nested query as following SQL form:
Select CustomerID From Sales.Customer Where TerritoryID＝(Select TerritoryID From Sales.SalesPerson Where SalesPersonID＝276)
This SQL query statement can be converted to the structured form being suitable for level identities route, for:
| Type | Information |
| Stratification name (outward) | The level identities of " Sales.Customer " |
| Return form (outward) | CustomerID |
| Stratification name (interior) | The level identities of " Sales.SalesPerson " |
| Return form (interior) | TerritoryID |
| Constraint condition (outward) | TerritoryID=returns results (interior) |
| Constraint condition (interior) | SalesPersonID＝276 |
Figure 10 shows the query processing flow process of the inquiry packets of single concept (table).Query processing process is substantially identical with the processing procedure that distributed relational database is inquired about, and key step comprises: localization inquiry, result aggregator and returning.Stratification name in inquiry packets or prefix are by the longest prefix match (LPM) with routing index, with Multicast Routing in correct data source, also the main information of NDNQL and the routing interface of process are recorded in PIT table in routing forwarding process; Then in data source, be converted into the inquiry of SQL form, complete the inquiry of localized relation data, obtain the relevant tuple and property value of meeting consumers' demand.Return results to be encapsulated as packet (NDNRS) that applicable former road returns and to utilize the PIT of NDN to show information and former for NDNRS road is returned.
Relate to the relation data inquiry of single concept if projection is (as π
a1 ..., an(R)) and select (as
) etc. operation.For this kind of inquiry, before given how to define its route Summing Factor constraint condition (mainly showing the value of alternative condition or the demand fulfillment given attribute).For the distributed implementation of this kind of inquiry, need to carry out self-organization process in a network, mainly comprise two parts, respectively: the route processing process of self-organization and the query processing process of self-organization.
(1) self-organization route processing
In order to the self-organization route processing of implementing structured inquiry, we adopt following steps: first the prefix of the level identities in inquiry packets and FIB is carried out longest prefix match, then carry out forward process according to the interface of FIB coupling.Above-mentioned process is repeated, until inquiry packets be forwarded in each related data sources at each routing node receiving inquiry packets.In addition, the Query Result that each data source nodes obtains after local search then need according to inquiry packets the path Yuan Lu of process return, or multiple returning results returns through the polymerization Hou Zaiyuan road of aggregation node.
(2) self-organization query processing
The polymerization process that self-organization query processing mainly comprises local search process and returns results.Local search process mainly first converts inquiry packets to SQL form in the data source nodes respectively receiving inquiry, then performs local search in the local database.
The polymerization process returned results mainly to return results each inquiry about certain at aggregation node and is polymerized.According to the feature returned results, polymerization process is mainly divided into two classes:
One is attribute polymerization (being vertically polymerized): respectively returning results namely for the vertical division had to a certain degree, is vertically polymerized, but will removes the property value repeated in each tuple according to its data attribute;
Another kind is tuple polymerization (level polymerization): namely for obtaining the tuple subset satisfied condition in each related data sources, carrying out level polymerization, but want the tuple of eliminate redundancy in aggregation node to each tuple subset.
Figure 11 shows the inquiry packets treatment scheme of the operations such as join, union, intersect, the except relating to multiple concept (table).The groundwork of such query processing first needs inquiry packets to be sent in the data source relevant to these concepts (table) carry out SQL conversion and complete local queries, then the grouping of the return data of each distribution carried out being polymerized and returning to user.Therefore need the level identities using each concept in routing forwarding, inquiry packets is routed in the data source at each concept place.Specifically for the distributed treatment of the complex queries such as aforesaid join, union, intersect, except, if multiple concept is distributed in same data source, then the process of inquiry packets is similar to the query processing of single concept, and difference is local search mainly.If multiple concept is distributed in different pieces of information source, then need the distributed approach considering form of specifically inquiring about.We are for equivalent method of attachment below, introduce the distributed treatment conventional method of this type of complex query.
Equivalent connection (
) relate generally to two concepts, might as well T and S be set to.Then the equivalent result connected is combination public attribute name being worth equal all tuples in T and S.Such as, search student name that same teacher attends class from student's table, teacher's table, be name, course numbering and teacher's name:
Select Sname,S.Dname,S.CNO,Tname
From Student AS S,Teacher AS T
Where S.CNO＝T.CNO
Order By Sname
For in the distributed treatment process of this inquiry, first need consideration two kinds of situations:
(1) relation data of S and T is at same source node: route can select arbitrarily a route factor to carry out mating and forwarding, and the query processing of local search and centralized relational database is similar;
(2) relation data of S and T is at not source node.Here the main query processing considering this situation, main method is the distributed semi method of attachment by improving.
Method 1 half-connection inquiry implementation method:
For also, hand over and the distributed ad-hoc process of difference operation, also to consider source node or aggregation node by each words and expressions return results carry out corresponding also, hand over or difference operation.
Figure 12 shows the treatment scheme of nested (nested) inquiry packets relating to multiple concept (table).The inquiry request sent as user relates to multiple concept (table), then in nested query grouping, necessarily there is many levels mark or prefix, then concrete processing procedure is:
(1) set i as innermost layer inquiry, an inquiry packets is used as in the inquiry of i layer;
(2) whether more than onely judge that i layer inquires about the concept number related to, if more than one, then by inquiry packets treatment scheme process this layer of inquiry relating to multiple concept, otherwise by relating to inquiry packets treatment scheme process this layer of inquiry of single concept;
(3) when the Query Result receiving i layer, first judge whether i layer is outermost layer inquiry, if it is turns (5), otherwise turns (4);
(4) according to the Query Result of i layer, rewrite the inquiry of i+1 layer, and be used as an inquiry packets, then i is increased 1 certainly, turn (2);
(5) by the Query Result loopback of i layer to user.
Sql like language provides 5 kinds of predefine aggregate functions, is mean value (avg), minimum value (min), maximal value (max), summation (sum) and counting (count) respectively.Under DYNAMIC DISTRIBUTION network environment, for our method that recurrence can be adopted repeatedly to assemble of computing of aggregate function, as Figure 13 shows the step-by-step polymerization computing flow process of aggregate query avg.For avg aggregate function, can be repeatedly again averaged value computing at the mean value operation result of aggregation node (mainly the routing node of NDN) by local.Just for consider during mean value computing for the local mean values that experience operation times is different be again averaged value computing time, then need the weight (the raw data number according to comprising is determined) that consideration is respective.
Claims (5)
1., based on a networking relation data querying method for level identities route, it is characterized in that, comprise the following steps:
(1) relation of inclusion in each local relational database between identification data entity, constructs the stratification name of each data entity, and by the name of each stratification or prefix advertisement in the forwarding information table in each routing node of named data network;
(2) data category inquired about according to user for each inquiry request is converted into other constraint condition the inquiry packets be made up of Tiered routing Summing Factor inquiry constraint factor with reference to the stratification name in forwarding information table, be sent to client's side link routing node;
(3) each routing node is to each inquiry packets received, use the matched routings algorithm of named data network the Tiered routing factor in inquiry packets and each in forwarding information table to be carried out mating and forwarding, in the request scratch list of named data network, register the information of the inquiry packets forwarded from each interface;
(4) after the information of inquiry packets arrives relevant database node, the information of inquiry packets is converted to the SQL query statement supporting local relational database, carries out local search;
(5) local search result adds that the Tiered routing factor of former inquiry request is packaged into packet, be then recycled to receive corresponding inquiry packets interface on;
(6) each routing node carries out buffer memory, polymerization and former road to the packet received and returns.
2. as claimed in claim 1 based on the networking relation data querying method of level identities route, it is characterized in that, described inquiry packets comprises following items: stratification name or prefix, inquiry constraint condition, return form, other condition, security parameter, not tuple, wherein stratification name or prefix, the form of returning can be more than one.
3. as claimed in claim 1 based on the networking relation data querying method of level identities route, it is characterized in that, described packet comprises following items: stratification name or prefix, return results, security parameter and timestamp, wherein returning results can the form tissue of one dimension, two-dimensional array or chained list.
4. as claimed in claim 1 based on the networking relation data querying method of level identities route, it is characterized in that, the SQL query statement destructing in described step (4) is the inquiry packets be made up of Tiered routing Summing Factor inquiry constraint factor.
5. as claimed in claim 1 based on the networking relation data querying method of level identities route, it is characterized in that, what the Tiered routing factor in described step (4) and each in forwarding information table were carried out mates is longest prefix match, by inquiry packets through repeatedly routing forwarding, deliver in relevant local data storehouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510003869.6A CN104537091B (en) | 2015-01-06 | 2015-01-06 | A kind of networking relation data querying method based on level identities routing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510003869.6A CN104537091B (en) | 2015-01-06 | 2015-01-06 | A kind of networking relation data querying method based on level identities routing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104537091A true CN104537091A (en) | 2015-04-22 |
| CN104537091B CN104537091B (en) | 2018-08-10 |
Family
ID=52852619
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510003869.6A Active CN104537091B (en) | 2015-01-06 | 2015-01-06 | A kind of networking relation data querying method based on level identities routing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104537091B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104754065A (en) * | 2015-04-28 | 2015-07-01 | 湖南科技大学 | Dynamic distribution Web resource management method and system based on content center network |
| CN107870978A (en) * | 2017-09-30 | 2018-04-03 | 上海壹账通金融科技有限公司 | Relational network topological method and application server |
| CN109247065A (en) * | 2016-06-10 | 2019-01-18 | 国际商业机器公司 | Enable the lasting stream identifier of different application |
| CN110417668A (en) * | 2019-07-24 | 2019-11-05 | 新华三大数据技术有限公司 | Message transmitting method and the network equipment |
| CN110572455A (en) * | 2019-09-06 | 2019-12-13 | 赛尔网络有限公司 | resource access method, device, node, equipment and medium |
| CN110598072A (en) * | 2019-09-24 | 2019-12-20 | 恩亿科(北京)数据科技有限公司 | Feature data aggregation method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120317643A1 (en) * | 2011-06-09 | 2012-12-13 | Samsung Electronics Co., Ltd. | Apparatus and method preventing overflow of pending interest table in name based network system |
| CN103873371A (en) * | 2014-02-21 | 2014-06-18 | 北京邮电大学 | Name routing fast matching search method and device |
-
2015
- 2015-01-06 CN CN201510003869.6A patent/CN104537091B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120317643A1 (en) * | 2011-06-09 | 2012-12-13 | Samsung Electronics Co., Ltd. | Apparatus and method preventing overflow of pending interest table in name based network system |
| CN103873371A (en) * | 2014-02-21 | 2014-06-18 | 北京邮电大学 | Name routing fast matching search method and device |
Non-Patent Citations (1)
| Title |
|---|
| 廖祝华 等: "互联网络中语义媒体查询的关系路由方法", 《软件学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104754065A (en) * | 2015-04-28 | 2015-07-01 | 湖南科技大学 | Dynamic distribution Web resource management method and system based on content center network |
| CN104754065B (en) * | 2015-04-28 | 2018-01-16 | 湖南科技大学 | DYNAMIC DISTRIBUTION web resource management method and system based on content center network |
| CN109247065A (en) * | 2016-06-10 | 2019-01-18 | 国际商业机器公司 | Enable the lasting stream identifier of different application |
| CN109247065B (en) * | 2016-06-10 | 2021-08-24 | 国际商业机器公司 | Continuous flow identifier enabling different applications |
| CN107870978A (en) * | 2017-09-30 | 2018-04-03 | 上海壹账通金融科技有限公司 | Relational network topological method and application server |
| CN110417668A (en) * | 2019-07-24 | 2019-11-05 | 新华三大数据技术有限公司 | Message transmitting method and the network equipment |
| CN110417668B (en) * | 2019-07-24 | 2021-11-02 | 新华三大数据技术有限公司 | Message transmission method and network equipment |
| CN110572455A (en) * | 2019-09-06 | 2019-12-13 | 赛尔网络有限公司 | resource access method, device, node, equipment and medium |
| CN110572455B (en) * | 2019-09-06 | 2022-05-13 | 赛尔网络有限公司 | Resource access method, device, node, equipment and medium |
| CN110598072A (en) * | 2019-09-24 | 2019-12-20 | 恩亿科(北京)数据科技有限公司 | Feature data aggregation method and device |
| CN110598072B (en) * | 2019-09-24 | 2022-03-01 | 恩亿科(北京)数据科技有限公司 | Feature data aggregation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104537091B (en) | 2018-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104537091A (en) | Networked relational data query method based on hierarchical identification routing | |
| CN101436192B (en) | Method and apparatus for optimizing inquiry aiming at vertical storage type database | |
| CN101873335B (en) | Distributed type searching method of cross-domain semantic Web service | |
| CN105930446B (en) | A kind of telecom client label generating method based on Hadoop distributed computing technology | |
| US20100131564A1 (en) | Index data structure for a peer-to-peer network | |
| WO2008014062A2 (en) | System and apparatus for optimally trading off the replication overhead and consistency level in distributed applications | |
| CN102708130A (en) | Scalable engine that computes user micro-segments for offer matching | |
| CN102567932A (en) | Method, device and system for managing comprehensive resources | |
| CN103425726A (en) | Processing queries in a computer network using business intelligence tools | |
| CN104091228A (en) | Systems for resource management, resource registering, resource inquiry and resource semantic corpus management of internet of things | |
| CN101667191B (en) | Information sharing method based on P2P | |
| CN102591960A (en) | Agricultural economy electronic map data service interface method | |
| CN102378407B (en) | Object name resolution system and method in internet of things | |
| CN102571427B (en) | The name of CIM resource and analytic method in a kind of distributed system | |
| Sladić et al. | The use of ontologies in cadastral systems | |
| CN105868414A (en) | Clustering separation distributive indexing method | |
| CN104765763B (en) | A kind of semantic matching method of the Heterogeneous Spatial Information classification of service based on concept lattice | |
| CN113792068A (en) | Method and device for organizing and retrieving multi-level multi-topic spatial data | |
| CN101442466B (en) | Superpose network and implementing method | |
| CN101813485B (en) | Electronic map based on geographic information data and navigation method thereof | |
| CN116467026A (en) | Cloud desktop data secure sharing and tracing method and system based on blockchain | |
| CN102929914A (en) | Mobile map service searching method based on P2P (point to point) node scheduling | |
| Hoang et al. | Collective cubing platform towards definition and analysis of warehouse cubes | |
| Ganti et al. | MP-trie: Fast spatial queries on moving objects | |
| CN108833466B (en) | System and method for publishing/subscribing traffic network space text |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |