CN109857743A - The construction method and device querying method and system of symmetrical canonical multi-dimensional indexing platform - Google Patents
The construction method and device querying method and system of symmetrical canonical multi-dimensional indexing platform Download PDFInfo
- Publication number
- CN109857743A CN109857743A CN201910111126.9A CN201910111126A CN109857743A CN 109857743 A CN109857743 A CN 109857743A CN 201910111126 A CN201910111126 A CN 201910111126A CN 109857743 A CN109857743 A CN 109857743A
- Authority
- CN
- China
- Prior art keywords
- query
- client node
- index
- tree
- node
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000010276 construction Methods 0.000 title claims abstract description 23
- 238000012423 maintenance Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims description 11
- 238000010586 diagram Methods 0.000 description 33
- 238000003860 storage Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000013507 mapping Methods 0.000 description 7
- 241000589968 Borrelia Species 0.000 description 6
- 208000016604 Lyme disease Diseases 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 5
- 238000013500 data storage Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention provides a kind of construction method of symmetrical canonical multi-dimensional indexing platform and devices, a kind of symmetrical canonical multi-dimensional indexing querying method and system, wherein the construction method of symmetrical canonical multi-dimensional indexing platform includes: at least one service end node of setting and at least one client node；The partial indexes based on k-d tree are constructed on each client node；Bohr's Cayley graph topology is added in selected part node in the partial indexes of client node k-d tree, constructs the global index based on Bohr's Cayley graph.It is high to solve prior art complexity, the problem of index maintenance cost is big, complex query inefficiency.
Description
Technical Field
The invention relates to the technical field of multidimensional indexing methods for mass information, in particular to a method and a device for constructing a symmetrical regular multidimensional indexing platform, and a method and a system for querying a symmetrical regular multidimensional index.
Background
In a cloud computing environment, a data storage mode of a data center network is a master-slave mode (master-slave), and the data is stored in a key value pair mode, so that the key value can be quickly queried. However, for complex queries, such as similarity query, multi-attribute query, interval query, K-nearest neighbor query, and queries including multiple similarity predictions, which all involve non-key-value attributes, global scanning needs to be performed in a mass data space, so that researchers provide a large number of multidimensional indexing methods and techniques for increasing the query speed of a cloud space. There are mainly three types of multidimensional indexing methods.
(1) A two-level index mode. The indexing method establishes a local index on a node which actually stores data, and establishes a global index on an index service node. The mainstream method for establishing the global index is to use a peer-to-peer network to construct a virtual overlay network, such as Chord, CAN, patrry, etc. And during query, the local index is positioned through the global index, and then the query is carried out on the data storage node. However, this two-tier index scheme requires all client nodes of the data center to be acquired to build the virtual overlay network. Meanwhile, as the query dimension increases, the performance is sharply reduced, and dimension disaster occurs.
(2) Two levels of index patterns. The two-level index mode takes the non-key-value attribute as the index of the second level, constructs a second-level index table, and gives the mapping of the non-key-value attribute and the key-value attribute belonging to the second-level index in the second-level index table. During query, key value attributes are searched in the secondary index table, and then the key value attributes are searched in the key value list. In the method, under the condition of excessive non-key value attributes, the storage cost is huge, the index updating complexity is high, and meanwhile, the high-dimensional query efficiency of multiple key values is low.
(3) Multidimensional spatial index pattern. The indexing method divides the data space into multiple dimensions with equal value, and then maps the multiple dimensions into one-dimensional linear indexes. The method has the consistency problem during space division, and when the data size is increased and the distribution is not uniform, the depth of the query tree is increased, so that the multidimensional query efficiency is influenced.
Under the existing cloud platform, a multidimensional index structure of a data center is limited by a master-slave (master-slave) data storage mode of the existing cloud platform, although various technical methods can better realize complex query under the requirement of small-scale query, when query dimensionality is increased, the query performance of multidimensional index is rapidly reduced, or dimensionality disaster occurs, or the complexity of index maintenance is too high.
In order to solve the problem of dimension disaster, a dimension reduction method based on similarity between data is provided. The dimension reduction method maps data from high dimension to low dimension or one dimension according to the similarity of original high dimension data, such as an identity method, a VA-File method, a Multi-Dimensional Scaling method (Multi-Dimensional Scaling), a random neighbor embedding method and a local linear embedding method.
Although the multidimensional indexing method is mature and the similarity-based dimension reduction method has been studied in a large quantity, the existing technology has high computational complexity and high index maintenance cost. Patent (CN102831225A) "multidimensional index structure in cloud environment, its construction method and similarity query method", using VA-file to perform quantization compression, inevitably missing part of data, patent (CN 105357247a) "multidimensional attribute cloud resource interval search method based on hierarchical cloud peer-to-peer network" and patent (CN 103678520a) "multidimensional interval query method based on cloud computing and its system" both use Chord ring topology to construct global index structure, and when the number of nodes increases and the index amount increases, query efficiency and time will increase sharply.
Disclosure of Invention
The invention aims to provide a method and a device for constructing a symmetrical regular multi-dimensional index platform, a method and a system for inquiring the symmetrical regular multi-dimensional index, which solve one of the problems of high complexity, high index maintenance cost and low complex query efficiency in the prior art or at least partially solve any one of the problems.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
one aspect of the present invention provides a method for constructing a symmetric canonical multidimensional index platform, including: setting at least one service end node and at least one client node; constructing a local index based on a k-d tree on each client node; and selecting partial nodes from the local indexes of the k-d tree of the client node to add into the bokelay graph topology, and constructing a global index based on the bokelay graph.
Each service end node is set to issue data query information and maintain data index; setting each client node to store mass data, receiving a query request, processing the query request and returning a query result, and maintaining a local index and a global index.
The invention also provides a query method of the symmetrical regular multidimensional index, which comprises the following steps: constructing a symmetrical regular multidimensional indexing platform by using the construction method of the symmetrical regular multidimensional indexing platform; a query client node acquires a query request, and acquires a global index routing algorithm of the query request by using a Bolkeclai graph topology according to the query request; the query client node forwards the query request to each client node by using a global index routing algorithm of the query request; and each client node respectively queries the local index of the k-d tree to obtain k-d tree nodes conforming to the query request, and determines a query result based on the k-d tree nodes conforming to the query request.
The global index routing algorithm for acquiring the query request by the query client node and utilizing the boekeley graph topology according to the query request comprises the following steps: constructing a breadth-first search tree taking a query client node as a root, and searching a target client node; if the target client node is inquired, stopping searching and acquiring a global index routing algorithm of the inquiry request; if the target client node is not inquired, searching the intermediate client node, determining the shortest path from the intermediate client node to the target node, and acquiring the global index routing algorithm of the inquiry request.
The client node respectively queries the local indexes of the k-d trees to obtain k-d tree nodes conforming to the query request, and the determining of the query result based on the k-d tree nodes conforming to the query request comprises the following steps: each client node judges whether the client node is a target client node; if yes, local indexes of the k-d tree are inquired in the cache of the user, k-d tree nodes meeting the conditions are returned, and a corresponding rectangular area is found out based on the index method of the k-d tree and is used as an inquiry result.
The invention also provides a device for constructing a symmetrical regular multidimensional index platform, which comprises: a setup module for setting up at least one server node and at least one client node; the building module is used for building a local index based on the k-d tree on each client node; and selecting partial nodes from the local indexes of the k-d tree of the client nodes to add into the bokelay graph topology, and constructing a global index based on the bokelay graph.
The setting module is specifically used for setting each service end node to issue data query information and maintain data indexes; setting each client node to store mass data, receiving a query request, processing the query request and returning a query result, and maintaining a local index and a global index.
In another aspect, the present invention provides a query system for symmetric canonical multidimensional indexes, including: a symmetric regular multidimensional indexing platform built by the building device of the symmetric regular multidimensional indexing platform according to claim 6 or 7: the query client node is used for acquiring a query request and acquiring a global index routing algorithm of the query request by using a Bolkyiley graph topology according to the query request; forwarding the query request to each client node by using a global index routing algorithm of the query request; each client node is used for respectively querying the local indexes of the k-d tree to obtain k-d tree nodes meeting the query request, and determining a query result based on the k-d tree nodes meeting the query request.
The query client node is specifically used for constructing a breadth-first search tree taking the query client node as a root and searching a target client node; if the target client node is inquired, stopping searching and acquiring a global index routing algorithm of the inquiry request; if the target client node is not inquired, searching the intermediate client node, determining the shortest path from the intermediate client node to the target node, and acquiring the global index routing algorithm of the inquiry request.
Each client node is specifically used for judging whether the client node is a target client node; if yes, local indexes of the k-d tree are inquired in the cache of the user, k-d tree nodes meeting the conditions are returned, and a corresponding rectangular area is found out based on the index method of the k-d tree and is used as an inquiry result.
Therefore, according to the construction method and device of the symmetrical regular multidimensional index platform, the symmetrical regular multidimensional index query method and system provided by the embodiment of the invention, because the symmetrical small-world topology Bolkele graph is adopted to construct the global topology, both index construction and route query can be directly obtained by a deterministic method, and the method and device have the characteristics of simplicity, feasibility and strong expandability. In addition, because the Borrelia diagram is more efficient than other Karelia diagrams which are in small-world topologies, the multidimensional index constructed on the Borrelia diagram is more efficient than the Chord structure query which is in ring topologies.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
Fig. 1 is a flowchart of a method for constructing a symmetric canonical multidimensional index platform according to an embodiment of the present invention;
FIG. 2 is a block diagram of a symmetric canonical multidimensional indexing platform system provided in an embodiment of the present invention;
FIG. 3 is a diagram of a local index k-d tree construction according to an embodiment of the present invention;
FIG. 4 is a block diagram of a hypercube structure of a mapping function from a local index k-d tree to a global index according to an embodiment of the present invention;
FIG. 5 is a representation of a chord ring of a Bourry Kaire chart in accordance with an embodiment of the present invention;
FIG. 6 is a global index building diagram provided by an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of a device for constructing a symmetric canonical multidimensional index platform according to an embodiment of the present invention,
FIG. 8 is a flowchart of a symmetric canonical multidimensional index query method provided by an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a symmetric canonical multidimensional index query system provided in an embodiment of the present invention;
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Fig. 1 shows a flowchart of a method for constructing a symmetric canonical multidimensional indexing platform according to an embodiment of the present invention, and referring to fig. 1, the method for constructing a symmetric canonical multidimensional indexing platform according to an embodiment of the present invention is applied to a peer-to-peer cloud network, and includes:
s101, setting at least one service end node and at least one client end node;
s102, constructing a local index based on a k-d tree on each client node;
s103, selecting partial nodes from the local indexes of the k-d tree of the client nodes to add into the bokelay graph topology, and constructing a global index based on the bokelay graph.
Specifically, in the embodiment of the invention, for the construction of the symmetric regular multidimensional index platform, a two-layer index mode can be adopted to construct a multidimensional index KD-BOCA, a local index is constructed by adopting a k-d tree, and a global index uses a hypercube-based mapping function to select leaf nodes of the k-d tree to construct a Bolcey Kaire diagram topology.
As an optional implementation manner of the embodiment of the present invention, each server node is configured to perform data query information issue and data index maintenance; setting each client node to store mass data, receiving a query request, processing the query request and returning a query result, and maintaining a local index and a global index. Specifically, a data index platform based on cloud computing is constructed, two types of nodes, namely a server and a client, can be maintained on the platform, the server is responsible for issuing data query information and maintaining data indexes, the client is responsible for storing mass data, receiving query requests, processing the query requests and returning query results, and the client maintains local indexes and global indexes at the same time. Managing original data of a data center, constructing an index structure by using multi-dimensional original data in a client node for storing the original data according to a k-d tree method, and generating a local index based on the k-d tree on each client storage node; and selecting nodes from the client storage nodes which generate the k-d tree indexes according to a mapping function based on a hypercube and adding the nodes into the global index to dynamically form a structured overlay network topology based on a Bolcey graph. The indexing can be conveniently performed by utilizing the functional characteristics of each service end node and each client node.
Therefore, by using the construction method of the symmetrical regular multidimensional index platform provided by the embodiment of the invention, the global topology is constructed by adopting the symmetrical small-world topology Borkhalei graph, so that both index construction and route query can be directly obtained by a deterministic method, and the method has the characteristics of simplicity, feasibility and strong expandability.
In specific implementation, a cloud platform supporting multidimensional indexes is constructed, and the cloud platform is composed of a server node and client nodes, as shown in fig. 2, two types of indexes are maintained on each client node.
1. A local index. The client node first locally generates a k-d tree based local index. In two-dimensional space D ═ Dl,dh]For example, wherein dl(a1,a2) Two-dimensional coordinates of the low-end boundary nodes, dh(b1,b2) Is the two-dimensional coordinate of the high-end boundary node. The k-d tree is a binary tree with each binary tree node kdT [ i ]]Axis parallel rectangle D corresponding to two-dimensional spaceiI.e. root node kdT [1 ]]Corresponds to D1Each binary tree intermediate node kdT [ i ]]Corresponding to two child nodes kdT [2i]And kdT [2i +1 ]]. The index space of these two child nodes is represented by node kdT [ i ]]The corresponding index space is split. The specific splitting rule and index construction steps are as follows:
firstly, a newly added node n randomly selects a node p in a space D to start to query, and through query routing, a rectangular area which is responsible for a node b is found to contain the newly added node.
Secondly, an index is established for the newly added node n. There are two specific situations:
in the first case: b is responsible for no other nodes in the rectangular area. At this time, the node b is split into two nodes, b is a left leaf node, and the newly added node is a right leaf node.
In the second case: and b has other nodes in the rectangular region in charge of, the newly added node n is one node in the rectangular region in charge of b, and at this time, the newly added node is only required to be merged into the existing node in the region.
2. A global index. And storing a k-d tree local index in the client node, wherein the global index is created by selecting partial nodes from the local index of the k-d tree of the client node and adding the partial nodes into the BolKaili graph topology. Specifically, the method comprises the following two steps.
First, a part of nodes in the k-d tree are selected. And the leaf nodes of the k-d tree are the nodes where the data are located. Let id of leaf node be (x, y), x be the depth of k-d tree where node is located, and h be the height of k-d tree.
Wherein, yi∈{0,1},0＜i≤h。y1y2…yhThe k-d branches on the path from the root node to the leaf node are numbered. As shown in fig. 3. Node c (2,01 x), representing the rectangular area for which it is responsible, is divided only 2 times, while it indexes rectangular areas within 01. As shown in FIG. 4, a k-d tree node of height h may be represented using a hypercube (h-cube) node of dimension h. Hypercube space H (h) can be thought of as a set of nodes { c }1c2…ch:ciE.g. 0,1, node c1c2…cnAnd d1d2…dnAdjacent if and only if, except for 1 value of i, c is present for all ii＝diHypercube H (h) can be expressed in a Karley plot. H (h) Cay (G, S),
the selection of the nodes is performed according to the following mapping function.
After the nodes of the k-d tree are mapped to the hypercube space, any leaf node of the k-d tree can find other leaf nodes by using a deterministic routing method.
Secondly, a global index based on the Bolcerly map is constructed. And randomly selecting partial nodes from hypercube nodes and putting the partial nodes into a global index table to construct a Bolceila graph. Fig. 5 is a chordal graph representation of a 4 degree boehrlacre graph with 21 nodes. Boehrliner plots have a ring topology like Chord. Studies have demonstrated that the borkelli graph topology is the most dense graph with the smallest diameter at the same node scale. The bolkalie network is a special sub-cycle graph, and the topological structure of the bolkalie network is a kalie graph constructed by bol subgroups. Bohr subgroup is defined as follows:
definition of 1 order
Wherein a ∈ Zp\ {0,1}, p is prime number, m satisfies amG is called the smallest positive integer of 1(mod p)bIs a 2 x 2 matrix group GL2(Zp) Bohr subgroup of (A) and (B) is denoted as BL2(Zp) Wherein the group operation is a modulo p matrix multiplication operation.
Let GbIs a Bohr subgroup, omega is GbIs not a null subset of (a) or (b),group GbBoCa (G) of Bollkali diagram for ΩbΩ) is defined as follows. The vertex set of BoCa (G, omega) is Boer subgroup BL2(Zp) The element in (1) is equal to the node v ≠ u ∈ GbG ∈ Ω, if v ═ u × G, the edge from node v to u is the edge set in BoCa (G, S), and the clustering operation ∈ is a modulo p matrix multiplication operation.
According to definition 1, BoCa (G) is shown in Boulkalie diagrambOmega) as Bohr subgroup BL2(Zp) 2 × 2 matrix elements in (1), then let Gb＝(Zm×hZpB), wherein, ZmAnd ZpIs a two-cycle group, the integer h ═ a-1(mod p)＝am-1(mod p)。
Let ti∈Zm,yi∈ZpIs given arbitrarilyOperation ofWhere ⊙ is a modulo-m addition operation,in order to perform the addition operation modulo p,
definitions Boehai Kaili diagram BoCa (G)bΩ) is:
BoCa (G) of BoerKaili diagrambΩ) is 4, and the number of nodes is p × m. Chudnovsky et al have demonstrated as early as 1988 that the Borkierley plot is the most dense 4 regular plot when the diameter is between 7 and 13. The boehrlac chart can be represented by a chord chart or a generalized chord chart, and can also be usedAnd a subcirculation graph. The boehrlichia graph is represented by a sub-loop topology, and then the mapping of the index can be conveniently realized, and fig. 6 shows the sub-loop topology representation of the boehrlichia graph with 21 nodes and an index mapping method of the sub-loop topology in a rectangular area.
Definition 2 any given integer m, p and h, satisfying gcd (p, h) 1, hm1(mod p), let Gb＝Zm×hZpAnd t isi∈Zm，yi∈ZpIs given arbitrarilyGroup operation ° is defined as follows:wherein ⊙ is a modulo-m addition operation,in order to perform the addition operation modulo p,scale (G)bAnd (v) is a subcycling group.
Anyvenberg subgroup BL2(Zp) An element ofIn the presence of h ═ a-1(mod p)＝am-1(mod p) reactingA two-dimensional array (t, y) ∈ Z can be usedm×hZpTo represent B. This two-dimensional representation is referred to as a subcircular representation of the Bohr subgroup. Order to generate element set Its corresponding matrix is represented as
Definition of 3 Γ (Z)m×hZpAnd omega) is a Bollkeley diagram.
It can be shown that BoCa (BL) is a Bouleian diagram for a given parameter m, p and a2(Zp) Ω) and subcirculation diagram Isomorphism, wherein h ═ a-1(mod p)。
Fig. 7 is a schematic structural diagram of a device for constructing a symmetric regular multidimensional indexing platform according to an embodiment of the present invention, where the device for constructing a symmetric regular multidimensional indexing platform is applied to the method, and only the structure of the device for constructing a symmetric regular multidimensional indexing platform is briefly described below, and other matters are not considered to be the best, please refer to the related description in the method for constructing a symmetric regular multidimensional indexing platform, see fig. 7, and the device for constructing a symmetric regular multidimensional indexing platform according to an embodiment of the present invention is applied to a peer-to-peer cloud network, and includes:
a setup module 701 for setting up at least one server node and at least one client node;
a building module 702 for building a k-d tree based local index on each client node; and selecting partial nodes from the local indexes of the k-d tree of the client nodes to add into the bokelay graph topology, and constructing a global index based on the bokelay graph.
Therefore, by using the construction device of the symmetrical regular multidimensional index platform provided by the embodiment of the invention, the global topology is constructed by adopting the symmetrical small-world topology Boulele graph, so that both index construction and route query can be directly obtained by a deterministic method, and the construction device has the characteristics of simplicity, feasibility and strong expandability.
As an optional implementation manner of the embodiment of the present invention, the setting module 701 is specifically configured to set each server node to perform data query information issue and data index maintenance; setting each client node to store mass data, receiving a query request, processing the query request and returning a query result, and maintaining a local index and a global index. The indexing can be conveniently performed by utilizing the functional characteristics of each service end node and each client node.
Fig. 8 is a flowchart illustrating a query method for a symmetric canonical multidimensional index according to an embodiment of the present invention, and referring to fig. 8, the query method for a symmetric canonical multidimensional index according to an embodiment of the present invention is applied to a peer-to-peer cloud network, and includes:
s801, constructing a symmetrical regular multidimensional index platform by using the construction method of the symmetrical regular multidimensional index platform;
s802, the query client node obtains a query request, and obtains a global index routing algorithm of the query request by using a Boulkalie graph topology according to the query request;
s803, the query client node forwards the query request to each client node by using a global index routing algorithm of the query request;
s804, each client node respectively queries the local index of the k-d tree to obtain k-d tree nodes conforming to the query request, and determines the query result based on the k-d tree nodes conforming to the query request.
As an optional implementation manner of the embodiment of the present invention, S802, the global index routing algorithm for acquiring, by a query client node, a query request and acquiring, according to the query request by using a boekalei graph topology, includes: constructing a breadth-first search tree taking a query client node as a root, and searching a target client node; if the target client node is inquired, stopping searching and acquiring a global index routing algorithm of the inquiry request; if the target client node is not inquired, searching the intermediate client node, determining the shortest path from the intermediate client node to the target node, and acquiring the global index routing algorithm of the inquiry request.
As an optional implementation manner of the embodiment of the present invention, in S804, the client node queries respective k-d tree local indexes to obtain k-d tree nodes meeting the query request, and determining the query result based on the k-d tree nodes meeting the query request includes: each client node judges whether the client node is a target client node; if yes, local indexes of the k-d tree are inquired in the cache of the user, k-d tree nodes meeting the conditions are returned, and a corresponding rectangular area is found out based on the index method of the k-d tree and is used as an inquiry result.
Therefore, by using the query method of the symmetric regular multidimensional index provided by the embodiment of the invention, the global topology is constructed by adopting the symmetric small-world topology Borkhalei graph, so that both the index construction and the route query can be directly obtained by a deterministic method, and the query method has the characteristics of simplicity, feasibility and strong expandability. In addition, because the Borrelia diagram is more efficient than other Karelia diagrams which are in small-world topologies, the multidimensional index constructed on the Borrelia diagram is more efficient than the Chord structure query which is in ring topologies.
In the specific implementation, the target node (a ', b') is queried from the nodes (a, b) based on the two-level index as follows.
(1) First, the client making the query request maps the query request to one or more virtual nodes on the boeholeleigh graph node space according to the global index BoCa query by a routing algorithm from the requesting node (a, b) to the query target node (a ', b').
Step one, belonging to Z for nodes (a, b)m×ZpFirstly, a breadth optimum taking the nodes (a, b) as roots is constructedSearch the tree T (a, b) first, order
At any depth α ≦ d, if the target node (a ', b') is reached, the search is stopped.
Otherwise, find node (a ", b"), make it satisfy the following condition:
Г(a")+Г(b")＝min{(Г(p)+Г(q))|(p,q)∈Nx(d),x＝(a,b)}
wherein Г (a) is min (a, m-a), Г (b) is min (b, p-b)
And step two, using a one-step greedy algorithm to restore the shortest path from the node (a ', b ') to the node (a ', b ') with the length of (Г (a ') + Г (b '). on the shortest path from the node (a ', b ') to the node (a ', b '), the neighbor node of the node (a ', b ') is (a '))i,bi) Optionally a node (a ', b') adjacent theretoi,bi) As the first node on the path, it is continuously at node (a)i,bi) The target node (a ', b') is found by using a one-step greedy algorithm, and the shortest path from the node (a ', b') to the node (a ', b') with the length of (Г (a ') + Г (b')) is regenerated.
(2) Secondly, the query result in the step (1) is forwarded to each client node through the global index route, when the client receives the query request, whether the client is the target node (a ', b') is judged, if yes, the local index of the k-d tree is queried in the cache, and the k-d tree node meeting the condition is returned as the query result. The local index firstly uses a hypercube space to carry out route forwarding search, returns k-d tree nodes meeting conditions, and then finds out a corresponding rectangular area based on a k-d tree indexing method to serve as a final query result.
Fig. 9 shows a schematic structural diagram of a query system for a symmetric canonical multidimensional index provided in an embodiment of the present invention, where the query system for a symmetric canonical multidimensional index is applied to the method, and only a simple description is given below on a structure of the query system for a symmetric canonical multidimensional index, and other matters are not considered to be the best, please refer to the relevant description in the query method for a symmetric canonical multidimensional index, see fig. 9, and the query system for a symmetric canonical multidimensional index provided in an embodiment of the present invention is applied to a peer-to-peer cloud network, and includes:
the symmetric regular multidimensional indexing platform 90 constructed by the construction device of the symmetric regular multidimensional indexing platform comprises: the query client node 901 is configured to obtain a query request, and obtain a global index routing algorithm of the query request according to the query request by using a boekele graph topology; forwarding the query request to each client node 902 using a global index routing algorithm for the query request;
each client node 902 is configured to query the respective k-d tree local index, obtain k-d tree nodes that meet the query request, and determine a query result based on the k-d tree nodes that meet the query request.
As an optional implementation manner of the embodiment of the present invention, the query client 901 is specifically configured to construct a breadth-first search tree using the query client as a root, and search for a target client; if the target client node is inquired, stopping searching and acquiring a global index routing algorithm of the inquiry request; if the target client node is not inquired, searching the intermediate client node, determining the shortest path from the intermediate client node to the target node, and acquiring the global index routing algorithm of the inquiry request.
As an optional implementation manner of the embodiment of the present invention, each client node 902 is specifically configured to determine whether it is a target client node; if yes, local indexes of the k-d tree are inquired in the cache of the user, k-d tree nodes meeting the conditions are returned, and a corresponding rectangular area is found out based on the index method of the k-d tree and is used as an inquiry result.
Therefore, the query system of the symmetrical regular multidimensional index provided by the embodiment of the invention adopts the symmetrical small-world topology Borkhalei graph to construct the global topology, so that both index construction and route query can be directly obtained by a deterministic method, and the query system has the characteristics of simplicity, feasibility and strong expandability. In addition, because the Borrelia diagram is more efficient than other Karelia diagrams which are in small-world topologies, the multidimensional index constructed on the Borrelia diagram is more efficient than the Chord structure query which is in ring topologies.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method for constructing a symmetrical regular multidimensional index platform is characterized by comprising the following steps:
setting at least one service end node and at least one client node;
building a k-d tree based local index on each of the client nodes;
and selecting partial nodes from the local indexes of the k-d tree of the client nodes to add into the bokelay graph topology, and constructing a global index based on the bokelay graph.
2. The method of claim 1,
setting each service end node to issue data query information and maintain data index;
and setting each client node to store mass data, receiving a query request, processing the query request and returning a query result, and maintaining the local index and the global index.
3. A query method of a symmetric regular multidimensional index is characterized by comprising the following steps:
constructing a symmetrical regular multidimensional indexing platform by using the construction method of the symmetrical regular multidimensional indexing platform according to claim 1 or 2;
a query client node acquires a query request, and acquires a global index routing algorithm of the query request by utilizing the Bolkeclai graph topology according to the query request;
the query client node forwards the query request to each client node by utilizing a global index routing algorithm of the query request;
and each client node respectively inquires the local index of the k-d tree to obtain the k-d tree nodes conforming to the inquiry request, and determines the inquiry result based on the k-d tree nodes conforming to the inquiry request.
4. The method of claim 3, wherein the query client node obtains a query request, and wherein the global index routing algorithm for obtaining the query request according to the query request using the boekeley graph topology comprises:
constructing a breadth-first search tree taking the query client node as a root, and searching a target client node;
if the target client node is inquired, stopping searching, and acquiring a global index routing algorithm of the inquiry request;
if the target client node is not queried, searching an intermediate client node, determining the shortest path from the intermediate client node to the target node, and acquiring a global index routing algorithm of the query request.
5. The method of claim 4, wherein the client nodes query respective k-d tree local indexes to obtain k-d tree nodes that satisfy the query request, and wherein determining query results based on the k-d tree nodes that satisfy the query request comprises:
each client node judges whether the client node is the target client node;
if yes, inquiring local indexes of the k-d tree in a cache of the user, returning k-d tree nodes meeting conditions, and finding out corresponding rectangular areas based on an index method of the k-d tree to serve as the inquiry results.
6. A construction device of a symmetrical regular multidimensional index platform is characterized by comprising the following components:
a setup module for setting up at least one server node and at least one client node;
the building module is used for building a local index based on a k-d tree on each client node; and selecting partial nodes from the local indexes of the k-d tree of the client nodes to add into the bokelay graph topology, and constructing a global index based on the bokelay graph.
7. The apparatus according to claim 6, wherein the setting module is specifically configured to set each of the server-side nodes to perform data query information publishing and data index maintenance; and setting each client node to store mass data, receiving a query request, processing the query request and returning a query result, and maintaining the local index and the global index.
8. A query system for symmetric canonical multidimensional indexing, comprising:
the symmetric regular multidimensional indexing platform constructed by the construction device of the symmetric regular multidimensional indexing platform according to claim 6 or 7: wherein,
the query client node is used for acquiring a query request and acquiring a global index routing algorithm of the query request by utilizing the Bolkyiley graph topology according to the query request; forwarding the query request to each client node using a global index routing algorithm for the query request;
each client node is used for respectively querying the local index of the k-d tree to obtain the k-d tree nodes meeting the query request, and determining a query result based on the k-d tree nodes meeting the query request.
9. The system according to claim 8, wherein said query client node is specifically configured to construct a breadth-first search tree rooted at said query client node for finding a target client node; if the target client node is inquired, stopping searching, and acquiring a global index routing algorithm of the inquiry request; if the target client node is not queried, searching an intermediate client node, determining the shortest path from the intermediate client node to the target node, and acquiring a global index routing algorithm of the query request.
10. The system according to claim 9, wherein each of said client nodes is specifically configured to determine whether it is said target client node; if yes, inquiring local indexes of the k-d tree in a cache of the user, returning k-d tree nodes meeting conditions, and finding out corresponding rectangular areas based on an index method of the k-d tree to serve as the inquiry results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910111126.9A CN109857743A (en) | 2019-02-12 | 2019-02-12 | The construction method and device querying method and system of symmetrical canonical multi-dimensional indexing platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910111126.9A CN109857743A (en) | 2019-02-12 | 2019-02-12 | The construction method and device querying method and system of symmetrical canonical multi-dimensional indexing platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109857743A true CN109857743A (en) | 2019-06-07 |
Family
ID=66897739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910111126.9A Pending CN109857743A (en) | 2019-02-12 | 2019-02-12 | The construction method and device querying method and system of symmetrical canonical multi-dimensional indexing platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109857743A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324642A (en) * | 2012-03-23 | 2013-09-25 | 日电（中国）有限公司 | Data index establishing system and method as well as data query method |
| CN103617162A (en) * | 2013-10-14 | 2014-03-05 | 南京邮电大学 | Method of constructing Hilbert R-tree index on equivalent cloud platform |
| CN105550332A (en) * | 2015-12-21 | 2016-05-04 | 河海大学 | Dual-layer index structure based origin graph query method |
| CN108182242A (en) * | 2017-12-28 | 2018-06-19 | 湖南大学 | A kind of indexing means for the inquiry of magnanimity multi dimensional numerical data area |
| WO2018148267A1 (en) * | 2017-02-08 | 2018-08-16 | The Research Foundation For The State University Of New York | System and method associated with progressive spatial analysis of prodigious 3d data including complex structures |
| US20180349374A1 (en) * | 2017-03-01 | 2018-12-06 | Sap Se | Lock-free hash indexing |
-
2019
- 2019-02-12 CN CN201910111126.9A patent/CN109857743A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324642A (en) * | 2012-03-23 | 2013-09-25 | 日电（中国）有限公司 | Data index establishing system and method as well as data query method |
| CN103617162A (en) * | 2013-10-14 | 2014-03-05 | 南京邮电大学 | Method of constructing Hilbert R-tree index on equivalent cloud platform |
| CN105550332A (en) * | 2015-12-21 | 2016-05-04 | 河海大学 | Dual-layer index structure based origin graph query method |
| WO2018148267A1 (en) * | 2017-02-08 | 2018-08-16 | The Research Foundation For The State University Of New York | System and method associated with progressive spatial analysis of prodigious 3d data including complex structures |
| US20180349374A1 (en) * | 2017-03-01 | 2018-12-06 | Sap Se | Lock-free hash indexing |
| CN108182242A (en) * | 2017-12-28 | 2018-06-19 | 湖南大学 | A kind of indexing means for the inquiry of magnanimity multi dimensional numerical data area |
Non-Patent Citations (3)
| Title |
|---|
| GANG CHEN 等: "A Framework for Supporting DBMS-like Indexes in the Cloud", 《PROCEEDINGS OF THE VLDB ENDOWMENT》 * |
| LAN LI 等: "A Novel Cayley-based Index in Cloud Storage", 《2015 IEEE INTERNATIONAL CONFERENCES ON SMART CITY》 * |
| XIANGYU ZHANG 等: "An efficient multi-demensional index for cloud data management", 《IN PROCEEDINGS OF THE CIKM WORKSHOP ON CLOUD DATA MANAGEMENT》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ding et al. | An efficient quad-tree based index structure for cloud data management | |
| Schmidt et al. | Flexible information discovery in decentralized distributed systems | |
| Wang et al. | Indexing multi-dimensional data in a cloud system | |
| CN110347680B (en) | Space-time data indexing method for interpyury environment | |
| CN101853283B (en) | Construction method for multidimensional data-oriented semantic indexing peer-to-peer network | |
| Du et al. | Spatio-temporal data index model of moving objects on fixed networks using hbase | |
| Li et al. | DPTree: A balanced tree based indexing framework for peer-to-peer systems | |
| Yang et al. | Distributed similarity queries in metric spaces | |
| Kumar et al. | M-Grid: a distributed framework for multidimensional indexing and querying of location based data | |
| Shen et al. | Leveraging a compound graph-based DHT for multi-attribute range queries with performance analysis | |
| Batko et al. | Scalability comparison of peer-to-peer similarity search structures | |
| CN109857743A (en) | The construction method and device querying method and system of symmetrical canonical multi-dimensional indexing platform | |
| Ryu et al. | MapReduce-based skyline query processing scheme using adaptive two-level grids | |
| Tsatsanifos et al. | Index-based query processing on distributed multidimensional data | |
| Tsatsanifos et al. | Midas: multi-attribute indexing for distributed architecture systems | |
| Huang et al. | PChord: a distributed hash table for P2P network | |
| Nam et al. | Analyzing design choices for distributed multidimensional indexing | |
| Surati et al. | BMMI-tree: A Peer-to-Peer m-ary tree using 1-m node splitting for an efficient multidimensional complex query search | |
| CN114153987A (en) | Distributed knowledge graph query method, device and storage medium | |
| Chen et al. | Multidimensional declustering schemes using golden ratio and kronecker sequences | |
| Chazapis et al. | Replica-aware, multi-dimensional range queries in distributed hash tables | |
| Li et al. | A PR-quadtree based multi-dimensional indexing for complex query in a cloud system | |
| Brodal et al. | D 2-tree: A new overlay with deterministic bounds | |
| Surati et al. | A Peer-to-Peer multiway tree network with efficient range query search using multidimensional indexing | |
| Zhelev et al. | A DHT-based scalable and fault-tolerant cloud information service |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190607 |
|
| RJ01 | Rejection of invention patent application after publication |