CN110515948A - A kind of data query method, system, electronic equipment and storage medium - Google Patents

A kind of data query method, system, electronic equipment and storage medium Download PDF

Info

Publication number
CN110515948A
CN110515948A CN201910784543.XA CN201910784543A CN110515948A CN 110515948 A CN110515948 A CN 110515948A CN 201910784543 A CN201910784543 A CN 201910784543A CN 110515948 A CN110515948 A CN 110515948A
Authority
CN
China
Prior art keywords
matrix
partitioning
data
fpga
inquiry instruction
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN201910784543.XA
Other languages
Chinese (zh)
Inventor
欧明阳
孙颉
赵红博
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Wave Intelligent Technology Co Ltd
Original Assignee
Suzhou Wave Intelligent Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Suzhou Wave Intelligent Technology Co Ltd filed Critical Suzhou Wave Intelligent Technology Co Ltd
Priority to CN201910784543.XA priority Critical patent/CN110515948A/en
Publication of CN110515948A publication Critical patent/CN110515948A/en
Withdrawn legal-status Critical Current

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
    • G06F16/22—Indexing; Data structures therefor; Storage structures
    • G06F16/2228—Indexing structures
    • G06F16/2237—Vectors, bitmaps or matrices
    • 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
    • G06F16/22—Indexing; Data structures therefor; Storage structures
    • G06F16/2282—Tablespace storage structures; Management thereof
    • 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
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2455—Query execution
    • G06F16/24553—Query execution of query operations
    • G06F16/24554—Unary operations; Data partitioning operations

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

This application discloses a kind of data query method, the data query method includes parsing the database list inquiry instruction when receiving database list inquiry instruction, and determine target data matrix according to parsing result and search matrix；The target data matrix is divided into N number of partitioning of matrix according to the size of the target data matrix；Serializing operation is executed to the partitioning of matrix, and the partitioning of matrix after serializing is transmitted to the corresponding memory headroom of the FPGA；The each partitioning of matrix of FPGA parallel computation matrix of consequence corresponding with the inquiry matrix is controlled, and sets the corresponding query result of the database list inquiry instruction for the matrix of consequence.The application can be improved the search efficiency of database list.Disclosed herein as well is a kind of data query system, storage medium and a kind of electronic equipment, have the above beneficial effect.

Description

A kind of data query method, system, electronic equipment and storage medium
Technical field
This application involves database technical field, in particular to a kind of data query method, system, a kind of storage medium and A kind of electronic equipment.
Background technique
Database technology is a core technology of information system, is a kind of method of computer-aided management data, In When carrying out inquiry operation to multiple tables of database, needs the list that will need to inquire to carry out cartesian product operation, obtain one Unified list, then executes inquiry operation again.
Traditional database is designed based on CPU (central processing unit, central processing unit), big in reply When scale, the operation of high concurrent data base read-write, traditional database has that inquiry time delay is high, is executing flute card using CPU When your product operation, data parallel characteristic can not be given full play to and promoted to bring is calculated.In the related technology, exist using GPU Database accelerating database searching operation scheme, but above-mentioned the relevant technologies in face of high-dimensional cartesian product operation when, GPU The deficiency of pipeline depth, seriously limits its performance, and search efficiency is lower.
Therefore, how to improve the search efficiency of database list is that the technology that those skilled in the art need to solve at present is asked Topic.
Summary of the invention
The purpose of the application is to provide a kind of data query method, system, storage medium and a kind of electronic equipment, Neng Gouti The search efficiency of high database list.
In order to solve the above technical problems, the application provides a kind of data query method, which includes:
When receiving database list inquiry instruction, the database list inquiry instruction is parsed, and tie according to parsing Fruit determines target data matrix and searches matrix；
The target data matrix is divided into N number of partitioning of matrix according to the size of the target data matrix；
Serializing operation is executed to the partitioning of matrix, and the partitioning of matrix after serializing is transmitted to the FPGA and is corresponded to Memory headroom；
The each partitioning of matrix of FPGA parallel computation matrix of consequence corresponding with the inquiry matrix is controlled, and will The matrix of consequence is set as the corresponding query result of the database list inquiry instruction.
Optionally, described to determine that target data matrix includes: according to parsing result
Target list is determined according to the parsing result, sets the number of targets for the cartesian product of the target list According to matrix.
Optionally, each partitioning of matrix of FPGA parallel computation result corresponding with the inquiry matrix is being controlled Before matrix, further includes:
The pipeline depth of the FPGA is set according to the dimension of the partitioning of matrix.
Optionally, the partitioning of matrix after serializing is transmitted to the corresponding memory headroom of the FPGA includes:
The partitioning of matrix after the serializing is transmitted to the corresponding memory headroom of the FPGA by PCIE link.
Optionally, by the matrix of consequence be set as the corresponding query result of the database list inquiry instruction it Afterwards, further includes:
The query result is sent to central processing unit by way of RDMA.
Optionally, further includes:
The query result and the lookup matrix are shown to human-computer interaction interface.
Present invention also provides a kind of data query system, which includes:
Parsing module, for parsing the database list inquiry instruction when receiving database list inquiry instruction, And target data matrix is determined according to parsing result and searches matrix；
Matrix division module, for the target data matrix to be divided into N according to the size of the target data matrix A partitioning of matrix；
Transmission module for executing serializing operation to the partitioning of matrix, and the partitioning of matrix after serializing is transmitted To the corresponding memory headroom of the FPGA；
Query result computing module, for controlling each partitioning of matrix of the FPGA parallel computation and the inquiry square The corresponding matrix of consequence of battle array, and the corresponding query result of the database list inquiry instruction is set by the matrix of consequence.
Optionally, the parsing module includes:
Resolution unit is instructed, for when receiving database list inquiry instruction, parsing the database list inquiry Instruction,
Data matrix determining module, for determining target list according to the parsing result, by the flute of the target list Karr product is set as the target data matrix.
Present invention also provides storage mediums, are stored thereon with computer program, and the computer program is realized when executing The step of above-mentioned data query method executes.
Present invention also provides a kind of electronic equipment, including memory and processor, calculating is stored in the memory Machine program, the processor realize the step that above-mentioned data query method executes when calling the computer program in the memory Suddenly.
This application provides a kind of data query methods, including when receiving database list inquiry instruction, parse institute Database list inquiry instruction is stated, and target data matrix is determined according to parsing result and searches matrix；According to the number of targets The target data matrix is divided into N number of partitioning of matrix according to the size of matrix；Serializing operation is executed to the partitioning of matrix, And the partitioning of matrix after serializing is transmitted to the corresponding memory headroom of the FPGA；Control each institute of the FPGA parallel computation Partitioning of matrix matrix of consequence corresponding with the inquiry matrix is stated, and sets the database list for the matrix of consequence and looks into It askes and instructs corresponding query result.
The application determines the target data matrix searched and use after receiving database list inquiry instruction In the inquiry matrix for determining query result.Target data matrix is divided into N number of partitioning of matrix first by the application, after division The partitioning of matrix is transmitted to the memory headroom of FPGA, and then corresponding with the inquiry matrix using the FPGA parallel computation partitioning of matrix Matrix of consequence.Since FPGA has good computing capability on data parallel and pipeline depth, cartesian product is substantially mentioned Search efficiency can be improved the search efficiency of database list.The application additionally provides a kind of data query system, storage simultaneously Medium and a kind of electronic equipment have above-mentioned beneficial effect, and details are not described herein.
Detailed description of the invention
In ord to more clearly illustrate embodiments of the present application, attached drawing needed in the embodiment will be done simply below It introduces, it should be apparent that, the drawings in the following description are only some examples of the present application, for ordinary skill people For member, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of flow chart of data query method provided by the embodiment of the present application；
Fig. 2 is a kind of schematic diagram of database list information integrating method provided by the embodiment of the present application；
Fig. 3 is the flow chart of another kind data query method provided by the embodiment of the present application；
Fig. 4 is a kind of operation method schematic diagram of cartesian product provided by the embodiment of the present application；
Fig. 5 is the data search flow diagram of 3 times of cartesian products of one kind provided by the embodiment of the present application；
Fig. 6 is the flow diagram that a kind of FPGA provided by the embodiments of the present application accelerates cartesian product to search；
Fig. 7 is a kind of structural schematic diagram of data query system provided by the embodiment of the present application.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is Some embodiments of the present application, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art Every other embodiment obtained without making creative work, shall fall in the protection scope of this application.
Below referring to Figure 1, Fig. 1 is a kind of flow chart of data query method provided by the embodiment of the present application.
Specific steps may include:
S101: when receiving database list inquiry instruction, the database list inquiry instruction is parsed, and according to solution Analysis result determines target data matrix and searches matrix；
Wherein, database list inquiry instruction can be a kind of instruction inquired multiple database lists, pass through Parse the available database list for needing to carry out inquiry operation of database list inquiry instruction and inquiry target.It needs to illustrate , when carrying out data base querying, will can first need the list inquired carry out cartesian product operation obtain one it is unified Then list executes inquiry operation again.Target data matrix in this step can be for the number in need for carrying out inquiry operation According to the cartesian product operation result of library list.
As a kind of feasible embodiment, target data matrix is determined according to parsing result in this step and searches matrix Operation can be with specifically: target list is determined according to the parsing result, sets the cartesian product of the target list to The target data matrix.Fig. 2 is referred to, Fig. 2 is a kind of database form information integration side provided by the embodiment of the present application Multiple database lists can be integrated into a database table using mode described in Fig. 2 by the schematic diagram of method, the present embodiment It is single.Inquiry matrix is the constraint condition for carrying out data base querying.
S102: the target data matrix is divided by N number of partitioning of matrix according to the size of the target data matrix；
Wherein, the present embodiment is on the basis of obtaining target data matrix, according to the size of target data matrix by target Data can matrix be divided into N number of partitioning of matrix.As a kind of feasible embodiment, target data matrix can averagely be divided For N number of partitioning of matrix.Specifically, the size of target data matrix and the partitioning of matrix quantity N of division are positively correlated.Since FPGA has There is good and processing capacity, therefore target data matrix is divided into N number of partitioning of matrix to give full play to the performance of FPGA. FPGA (Field-Programmable Gate Array), i.e. field programmable gate array.
S103: executing serializing operation to the partitioning of matrix, and the partitioning of matrix after serializing is transmitted to described The corresponding memory headroom of FPGA；
Wherein, the process that object is converted to byte sequence is called the serializing of object.When two processes are carrying out remotely When communication, various types of data can be sent each other.Either what type of data, all can be in the form of binary sequence It is transmitted on network.Sender needs this object to be converted to byte sequence, could transmit on network；Recipient then needs Byte sequence is reverted to object again.The partitioning of matrix is converted into byte sequence in the present embodiment, and by the syllable sequence after conversion Biographies are transported in the corresponding memory headroom of FPGA.It is understood that in the present embodiment sequence can also be executed to inquiry matrix Columnization operation, and by the inquiry Transfer-matrix after serializing to the corresponding memory headroom of the FPGA.
S104: each partitioning of matrix of FPGA parallel computation result square corresponding with the inquiry matrix is controlled Battle array, and the corresponding query result of the database list inquiry instruction is set by the matrix of consequence.
Wherein, the purpose of this step is the ability using FPGA parallel computation, inquire in each partitioning of matrix with inquiry The corresponding matrix of consequence of matrix, and set the corresponding inquiry of database list inquiry instruction for the matrix of consequence inquired and tie Fruit.Inquiry in this step refers to according to constraint condition, and target value (x is found from the set of (x, y) ∈ X × Y0,y0)。
The present embodiment after receiving database list inquiry instruction, determine the target data matrix searched with For determining the inquiry matrix of query result.Target data matrix is divided into N number of partitioning of matrix first by the present embodiment, will be divided The partitioning of matrix afterwards is transmitted to the memory headroom of FPGA, and then utilizes the FPGA parallel computation partitioning of matrix and the inquiry matrix pair The matrix of consequence answered.Since FPGA has good computing capability on data parallel and pipeline depth, Descartes is substantially mentioned Long-pending search efficiency can be improved the search efficiency of database list.
Fig. 3 is referred to below, Fig. 3 is the flow chart of another kind data query method provided by the embodiment of the present application, this Embodiment may include following operation:
S201: when receiving database list inquiry instruction, the database list inquiry instruction is parsed；
S202: determining target list according to the parsing result, sets described for the cartesian product of the target list Target data matrix；
Wherein, cartesian product (Cartesian product) is an arithmetic operation in mathematical set branch, the operation Return a set by being constituted by operational set product.For example, cartesian product A × set B of set A, B be sequence (a, B) set constituted, wherein a ∈ A, b ∈ B.Use collective formula are as follows:
A × B=(a, b) | a ∈ A and B ∈ B }.
Refer to Fig. 4, Fig. 4 is a kind of operation method schematic diagram of cartesian product provided by the embodiment of the present application, in Fig. 4 Cartesian product operation, available table are done to row set A={ x, y, z } and column set B={ 1,2,3 }.Usually to the set of n dimension It does cartesian product and is also referred to as " n times of cartesian product " (n-fold Cartesian product), array representation can be tieed up with n, Wherein each element is " n tuple " (n-tuple), and the ordered pair in Fig. 4 is 2 tuples.
S203: the target data matrix is divided by N number of partitioning of matrix according to the size of the target data matrix；
S204: serializing operation is executed to the partitioning of matrix, and passes through PCIE link for the matrix after the serializing Block transmission is to the corresponding memory headroom of the FPGA；
Wherein, PCIE (PCI-Express, peripheral component interconnect express) is one Kind high speed serialization computer expansion bus standard.
S205: the pipeline depth of the FPGA is set according to the dimension of the partitioning of matrix；
It wherein, can be according to the partitioning of matrix there may be the operation for the dimension for determining the partitioning of matrix before this step Dimension determines the pipeline depth of FPGA.
S206: each partitioning of matrix of FPGA parallel computation result square corresponding with the inquiry matrix is controlled Battle array, and the corresponding query result of the database list inquiry instruction is set by the matrix of consequence.
Wherein, FPGA is not limited on data search the operation of cartesian product, through FPGA in conjunction with NVM technology, also FPGA can be used, the operation such as generation, sequence, dimensionality reduction, lookup of cartesian product is rapidly completed.
S207: the query result is sent to central processing unit by way of RDMA；
Wherein, RDMA (Remote Direct Memory Access) refers to that remote direct data accesses.
S208: the query result and the lookup matrix are shown to human-computer interaction interface.
Wherein, the present embodiment can show query result to human-computer interaction interface, after obtaining query result into one Step can also show inquiry matrix to human-computer interaction interface, and show inquiry query result and inquiry square in human-computer interaction interface The corresponding relationship of battle array.
Illustrate the process of above-described embodiment description below by embodiment in practical applications.
For the cartesian product of a N-dimensional, if data matrix is WM×N, lookup matrix is FL×N, then matrix of consequence OL×MFor Search the left product of matrix and data matrix:
OL×M=FL×N·WM×N；
Wherein, M is the number of data contained during data are put to the proof, and N represents the dimension of every data, and L refers to the number that needs are inquired According to item number, output matrix returns to the result of inquiry:
According to matrix of consequence OL×MJudgement, obtains cartesian product query result, the knot for including in matrix behavior data matrix Fruit, matrix are classified as the entry for needing to inquire.Respective entries check in data in data matrix, and then return value is greater than 0, such as O in formula (0,0)=5, then it represents that first data comprising required inquiry in data matrix, according to index information, reading database In the value of the position can be obtained required data, if query result is 0, there is no required data in data matrix.Please It is the data search flow diagram of 3 times of cartesian products of one kind provided by the embodiment of the present application referring to Fig. 5, Fig. 5.
When quickly being searched using FPGA, first according to data matrix size, a plurality of line is opened at the end CPU Data matrix piecemeal, then block-by-block are read into memory from database by block, and are serialized by journey.The data of serializing are pressed Sequence is read in FPGA specified memory, similarly, is searched matrix and is also needed to be written in FPGA specified memory after serializing.FPGA is pressed According to the data length to it, data parallel is carried out according to the number of data L of lookup, requires N to deepen pipeline depth according to dimension. Based on the above design, FPGA is given full play in data and the parallel advantage of pipeline depth, the result of inquiry is in order through FPGA Accelerator MAC network consisting packet sends back to CPU, carries out the operation that data are read in returned respectively using by PCIe and MAC, will be effective The bandwidth of each communication interface is played, meanwhile, calculated result is no longer through memory access, but directly network consisting packet is made by MAC With in RDMA (Remote Direct Memory Access, remote direct data access) agreement write-in CPU memory, reduce Transmission path delay, refers to Fig. 6, and Fig. 6 is the process that a kind of FPGA provided by the embodiments of the present application accelerates cartesian product to search Schematic diagram.Method for quickly retrieving of the present embodiment based on FPGA building cartesian product, passes through data and depth using FPGA accelerator card Spend Parallel Implementation cartesian product and search and accelerate, improve the design handled up, using PCIe and MAC execute respectively FPGA device read in Write-in CPU memory carrys out improve data transfer bandwidth, saves data in FPGA memory and moves process, the method for reducing system delay.
Fig. 7 is referred to, Fig. 7 is a kind of structural schematic diagram of data query system provided by the embodiment of the present application；
The system may include:
Parsing module 100 refers to for when receiving database list inquiry instruction, parsing the database list inquiry It enables, and target data matrix is determined according to parsing result and searches matrix；
Matrix division module 200, for being divided the target data matrix according to the size of the target data matrix For N number of partitioning of matrix；
Transmission module 300 for executing serializing operation to the partitioning of matrix, and the partitioning of matrix after serializing is passed Transport to the corresponding memory headroom of the FPGA；
Query result computing module 400 is looked into for controlling each partitioning of matrix of the FPGA parallel computation with described The corresponding matrix of consequence of matrix is ask, and sets the corresponding inquiry of the database list inquiry instruction for the matrix of consequence and ties Fruit.
The present embodiment after receiving database list inquiry instruction, determine the target data matrix searched with For determining the inquiry matrix of query result.Target data matrix is divided into N number of partitioning of matrix first by the present embodiment, will be divided The partitioning of matrix afterwards is transmitted to the memory headroom of FPGA, and then utilizes the FPGA parallel computation partitioning of matrix and the inquiry matrix pair The matrix of consequence answered.Since FPGA has good computing capability on data parallel and pipeline depth, Descartes is substantially mentioned Long-pending search efficiency can be improved the search efficiency of database list.
Further, the parsing module 100 includes:
Resolution unit is instructed, for when receiving database list inquiry instruction, parsing the database list inquiry Instruction,
Data matrix determining module, for determining target list according to the parsing result, by the flute of the target list Karr product is set as the target data matrix.
Further, further includes:
Parameter setting module, for the pipeline depth of the FPGA to be arranged according to the dimension of the partitioning of matrix.
Further, the partitioning of matrix after the serializing is transmitted to by transmission module particularly for by PCIE link The module of the corresponding memory headroom of the FPGA.
Further, further includes:
As a result module occurs, for the query result to be sent to central processing unit by way of RDMA.
Further, further includes:
Display module, for showing the query result and the lookup matrix to human-computer interaction interface.
Since the embodiment of components of system as directed is corresponded to each other with the embodiment of method part, the embodiment of components of system as directed is asked Referring to the description of the embodiment of method part, wouldn't repeat here.
Present invention also provides storage mediums, have computer program thereon, which is performed can be real Step provided by existing above-described embodiment.The storage medium may include: USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic or disk etc. is various to deposit Store up the medium of program code.
Present invention also provides a kind of electronic equipment, may include memory and processor, have meter in the memory Calculation machine program may be implemented provided by above-described embodiment when the processor calls the computer program in the memory Step.Certain electronic equipment can also include various network interfaces, the components such as power supply.
Each embodiment is described in a progressive manner in specification, the highlights of each of the examples are with other realities The difference of example is applied, the same or similar parts in each embodiment may refer to each other.For system disclosed in embodiment Speech, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part illustration .It should be pointed out that for those skilled in the art, under the premise of not departing from the application principle, also Can to the application, some improvement and modification can also be carried out, these improvement and modification also fall into the protection scope of the claim of this application It is interior.
It should also be noted that, in the present specification, relational terms such as first and second and the like be used merely to by One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or The intrinsic element of equipment.Under the situation not limited more, the element limited by sentence "including a ..." is not arranged Except there is also other identical elements in the process, method, article or apparatus that includes the element.

Claims (10)

1. a kind of data query method characterized by comprising
When receiving database list inquiry instruction, the database list inquiry instruction is parsed, and true according to parsing result Set the goal data matrix and lookup matrix；
The target data matrix is divided into N number of partitioning of matrix according to the size of the target data matrix；
Serializing operation is executed to the partitioning of matrix, and the partitioning of matrix after serializing is transmitted to the corresponding memory sky of FPGA Between；
The each partitioning of matrix of FPGA parallel computation matrix of consequence corresponding with the inquiry matrix is controlled, and will be described Matrix of consequence is set as the corresponding query result of the database list inquiry instruction.
2. data query method according to claim 1, which is characterized in that described to determine target data square according to parsing result Battle array include:
Target list is determined according to the parsing result, sets the target data square for the cartesian product of the target list Battle array.
3. data query method according to claim 1, which is characterized in that each described controlling the FPGA parallel computation Before partitioning of matrix matrix of consequence corresponding with the inquiry matrix, further includes:
The pipeline depth of the FPGA is set according to the dimension of the partitioning of matrix.
4. data query method according to claim 1, which is characterized in that the partitioning of matrix after serializing is transmitted to FPGA Corresponding memory headroom includes:
The partitioning of matrix after the serializing is transmitted to the corresponding memory headroom of FPGA by PCIE link.
5. data query method according to claim 1, which is characterized in that setting the data for the matrix of consequence After the corresponding query result of library list inquiry instruction, further includes:
The query result is sent to central processing unit by way of RDMA.
6. according to claim 1 to any one of 5 data query methods, which is characterized in that further include:
The query result and the lookup matrix are shown to human-computer interaction interface.
7. a kind of data query system characterized by comprising
Parsing module, for when receiving database list inquiry instruction, parsing the database list inquiry instruction, and root Target data matrix is determined according to parsing result and searches matrix；
Matrix division module, for the target data matrix to be divided into N number of square according to the size of the target data matrix Battle array piecemeal；
Transmission module for executing serializing operation to the partitioning of matrix, and the partitioning of matrix after serializing is transmitted to The corresponding memory headroom of FPGA；
Query result computing module, for controlling each partitioning of matrix of the FPGA parallel computation and the inquiry matrix pair The matrix of consequence answered, and the corresponding query result of the database list inquiry instruction is set by the matrix of consequence.
8. data query system according to claim 7, which is characterized in that the parsing module includes:
Resolution unit is instructed, for parsing the database list inquiry instruction when receiving database list inquiry instruction,
Data matrix determining module, for determining target list according to the parsing result, by the Descartes of the target list Product is set as the target data matrix.
9. a kind of electronic equipment, which is characterized in that including memory, processor and FPGA, calculating is stored in the memory Machine program, the processor realize the number as described in any one of claim 1 to 6 when calling the computer program in the memory The step of according to querying method.
10. a kind of storage medium, which is characterized in that be stored with computer executable instructions, the calculating in the storage medium When machine executable instruction is loaded and executed by processor, any one of claim 1 to 6 as above data query method is realized Step.
CN201910784543.XA 2019-08-23 2019-08-23 A kind of data query method, system, electronic equipment and storage medium Withdrawn CN110515948A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910784543.XA CN110515948A (en) 2019-08-23 2019-08-23 A kind of data query method, system, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910784543.XA CN110515948A (en) 2019-08-23 2019-08-23 A kind of data query method, system, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN110515948A true CN110515948A (en) 2019-11-29

Family

ID=68627386

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910784543.XA Withdrawn CN110515948A (en) 2019-08-23 2019-08-23 A kind of data query method, system, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN110515948A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111625585A (en) * 2020-05-22 2020-09-04 中科驭数（北京）科技有限公司 Access method, device, host and storage medium of hardware acceleration database
CN117331970A (en) * 2023-10-31 2024-01-02 中科驭数（北京）科技有限公司 Data query method, device, computer storage medium and acceleration card

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111625585A (en) * 2020-05-22 2020-09-04 中科驭数（北京）科技有限公司 Access method, device, host and storage medium of hardware acceleration database
CN117331970A (en) * 2023-10-31 2024-01-02 中科驭数（北京）科技有限公司 Data query method, device, computer storage medium and acceleration card

Similar Documents

Publication Publication Date Title
CN110990638B (en) Large-scale data query acceleration device and method based on FPGA-CPU heterogeneous environment
TWI735545B (en) Model training method and device
US8537160B2 (en) Generating distributed dataflow graphs
US10474652B2 (en) Optimizing wide data-type storage and analysis of data in a column store database
CN104714972B (en) Database divides table foundation and querying method
CN111382327B (en) Character string matching device and method
CN106407303A (en) Data storage method and apparatus, and data query method and apparatus
WO2014031416A2 (en) Hardware implementation of the aggregation/group by operation: hash-table method
CN111352902A (en) Log processing method and device, terminal equipment and storage medium
CN105117355A (en) Memory, memory system and data process method
CN102902763B (en) The method of association, retrieving information process data and process information task and device
CN112015741A (en) Method and device for storing massive data in different databases and tables
CN107273213A (en) A kind of calculation control method, network interface card and electronic equipment
CN105574054A (en) Distributed cache range query method, apparatus and system
CN110515948A (en) A kind of data query method, system, electronic equipment and storage medium
CN107391402A (en) A kind of data operating method, device and a kind of data operation card
US11023825B2 (en) Platform as a service cloud server and machine learning data processing method thereof
CN109947729A (en) A kind of real-time data analysis method and device
CN107205024A (en) Data duplicate removal method and system in a kind of cloud storage system
CN107943846B (en) Data processing method and device and electronic equipment
CN105138649A (en) Data search method and device and terminal
US9229969B2 (en) Management of searches in a database system
CN109726219A (en) The method and terminal device of data query
CN108319604B (en) Optimization method for association of large and small tables in hive
CN113010542A (en) Service data processing method and device, computer equipment and storage medium

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20191129

WW01 Invention patent application withdrawn after publication