CN109086383A - A kind of method and apparatus accelerated for database - Google Patents
A kind of method and apparatus accelerated for database Download PDFInfo
- Publication number
- CN109086383A CN109086383A CN201810831293.6A CN201810831293A CN109086383A CN 109086383 A CN109086383 A CN 109086383A CN 201810831293 A CN201810831293 A CN 201810831293A CN 109086383 A CN109086383 A CN 109086383A
- Authority
- CN
- China
- Prior art keywords
- output
- elements
- output element
- database
- filter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of method and apparatus accelerated for database, which comprises the database accelerator by calculating equipment decompresses one or more compression elements in compressed data library to generate one or more decompression elements;Database accelerator prepares the element after one or more decompressions, generates one or more ready elements, for accelofilter processing;Accelerate filter with executing by database accelerator to one or more ready elements to generate one or more output elements.
Description
Technical field
The present invention relates to technical field of data processing, in particular to a kind of method accelerated for database and set
It is standby.
Background technique
Background technique Modern computing devices may include general-purpose processor core and for executing the various hard of special duty
Part accelerator.Certain calculating equipment may include being presented as one or more accelerators of field programmable gate array (FPGA),
It may include the programmable digital logic resource that can be configured by terminal user or system integrator.In the prior art, with
The appearance of the data formats such as picture video causes requirement of user when using data to database higher and higher, for data
The operating speed in library also requires to be getting faster, and how to carry out the acceleration of database is the problem of current techniques.
Summary of the invention
The invention proposes a kind of methods accelerated for database, which comprises
Database accelerator by calculating equipment decompresses one or more compression elements in compressed data library to generate
One or more decompression element;
Database accelerator prepares the element after one or more decompressions, generates one or more ready elements,
For accelofilter processing;With
Being executed by database accelerator to one or more ready elements accelerates filter to generate one or more
A output element.
The method, wherein the element for preparing one or more of decompressions includes by one or more of decompressions
Each of element of contracting is formatted as the object format compatible with the filter of the acceleration.
The method, further includes:
Output element is compressed by database accelerator to generate compression output element;With
By database accelerator by the output element write-in memory of compression to respond compression output element.
The method further includes polymerizeing output element by database accelerator to generate aggregated data.
The method, in which: the element of each decompression includes signless integer, and one or more decompressions
Element includes the packing array of the element of decompression;It include signless integer with each output element, and one or more defeated
Element includes the packing array of output element out.
The method further includes modifying one or more of output members in response to executing the acceleration filter
Part.
A kind of calculating equipment accelerated for database, the calculating equipment includes database accelerator, wherein the number
Include: according to library accelerator
Database decompressor, for decompressing one or more compression elements in compressed data library, with generate one or
Multiple decompression elements;
Database preparation module, for preparing the element of one or more decompressions, to generate one or more by accelerating
The preparation element of filter process;With
Filter manager executes accelerated filtration to one or more ready elements, one or more defeated to generate
Element out.
The calculating equipment, preparing one or more of decompression elements includes by one or more of decompressions
Each of element is formatted as the object format compatible with the acceleration filter.
The calculating equipment, further includes: database compressing device, for compressing the output element to generate compression output
Element, wherein in response to the compression of the output element memory is written in the compression output element by the calculating equipment.
The calculating equipment, wherein executing accelofilter to one or more of ready elements includes holding
Row extracts, step-by-step logical operation, scans, and generates, and converts, polymerization, sequence or setting membership
The calculating equipment accelerates filter including the use of field programmable gate wherein executing using database accelerator
Array, which executes, accelerates filter.
The calculating equipment, wherein executed using database accelerator and accelerate filter including the use of dedicated integrated electricity
Road, which executes, accelerates filter.
The calculating equipment, the filter manager are also used to polymerize the output element to generate aggregated data.
The calculating equipment, polymerizeing the output element includes calculating total count, minimum value, maximum value or summation.
The calculating equipment, wherein polymerize the output element include: to the output element execute arithmetical operation or
Statistical calculation.
The calculating equipment, in which:
The element of each decompression includes signless integer, and the element of one or more decompressions includes decompression
The packing array of element;With
Each output element includes signless integer, and one or more output elements include the packing battle array of output element
Column.
The calculating equipment, wherein the database preparation module is also used in response to the property for accelerating filter
One or more of output elements can be modified.
The calculating equipment, wherein modifying one or more of output elements includes: each output element of extension
Bit width is truncated the bit width of each output element, value is attached to each output element, or be prepended each defeated
The value of element out.
The calculating equipment, wherein modifying one or more of output elements includes by one or more of outputs
Element is converted to array of indexes, wherein one or more of output elements include bit vector, and wherein each index includes:
Array of indexes corresponds to the setting position in bit vector.
Detailed description of the invention
From following description with reference to the accompanying drawings it will be further appreciated that the present invention.Component in figure is not drawn necessarily to scale,
But it focuses on and shows in the principle of embodiment.In the figure in different views, identical appended drawing reference is specified to be corresponded to
Part.
Fig. 1 is the schematic diagram of the method accelerated for database of the invention.
Specific embodiment
In order to enable the objectives, technical solutions, and advantages of the present invention are more clearly understood, below in conjunction with embodiment, to this
Invention is further elaborated;It should be appreciated that described herein, the specific embodiments are only for explaining the present invention, and does not have to
It is of the invention in limiting.To those skilled in the art, after access is described in detail below, other systems of the present embodiment
System, method and/or feature will become obvious.All such additional systems, method, feature and advantage are intended to be included in
It in this specification, is included within the scope of the invention, and by the protection of the appended claims.In description described in detail below
The other feature of the disclosed embodiments, and these characteristic roots will be apparent according to described in detail below.
Embodiment one:
As shown in Figure 1, for a kind of schematic diagram of the method accelerated for database of the present invention, which comprises
Database accelerator by calculating equipment decompresses one or more compression elements in compressed data library to generate
One or more decompression element;
Database accelerator prepares the element after one or more decompressions, generates one or more ready elements,
For accelofilter processing;With
Being executed by database accelerator to one or more ready elements accelerates filter to generate one or more
A output element.
The method, wherein the element for preparing one or more of decompressions includes by one or more of decompressions
Each of element of contracting is formatted as the object format compatible with the filter of the acceleration.
The method, further includes:
Output element is compressed by database accelerator to generate compression output element;With
By database accelerator by the output element write-in memory of compression to respond compression output element.
The method further includes polymerizeing output element by database accelerator to generate aggregated data.
The method, in which:
The element of each decompression includes signless integer, and the element of one or more decompressions includes decompression
The packing array of element;With
Each output element includes signless integer, and one or more output elements include the packing battle array of output element
Column.The method further includes modifying one or more of output elements in response to executing the acceleration filter.
Embodiment two:
A kind of calculating equipment accelerated for database, the calculating equipment includes database accelerator, wherein the number
Include: according to library accelerator
Database decompressor, for decompressing one or more compression elements in compressed data library, with generate one or
Multiple decompression elements;
Database preparation module, for preparing the element of one or more decompressions, to generate one or more by accelerating
The preparation element of filter process;With
Filter manager executes accelerated filtration to one or more ready elements, one or more defeated to generate
Element out.
The calculating equipment, preparing one or more of decompression elements includes by one or more of decompressions
Each of element is formatted as the object format compatible with the acceleration filter.
The calculating equipment, further includes: database compressing device, for compressing the output element to generate compression output
Element, wherein in response to the compression of the output element memory is written in the compression output element by the calculating equipment.
The calculating equipment, wherein executing accelofilter to one or more of ready elements includes holding
Row extracts, step-by-step logical operation, scans, and generates, and converts, polymerization, sequence or setting membership
The calculating equipment accelerates filter including the use of field programmable gate wherein executing using database accelerator
Array, which executes, accelerates filter.
The calculating equipment, wherein executed using database accelerator and accelerate filter including the use of dedicated integrated electricity
Road, which executes, accelerates filter.
The calculating equipment, the filter manager are also used to polymerize the output element to generate aggregated data.
The calculating equipment, polymerizeing the output element includes calculating total count, minimum value, maximum value or summation.
The calculating equipment, wherein polymerize the output element include: to the output element execute arithmetical operation or
Statistical calculation.
The calculating equipment, in which:
The element of each decompression includes signless integer, and the element of one or more decompressions includes decompression
The packing array of element;With
Each output element includes signless integer, and one or more output elements include the packing battle array of output element
Column.
The calculating equipment, wherein the database preparation module is also used in response to the property for accelerating filter
One or more of output elements can be modified.
The calculating equipment, wherein modifying one or more of output elements includes: each output element of extension
Bit width is truncated the bit width of each output element, value is attached to each output element, or be prepended each defeated
The value of element out.
The calculating equipment, wherein modifying one or more of output elements includes by one or more of outputs
Element is converted to array of indexes, wherein one or more of output elements include bit vector, and wherein each index includes:
Array of indexes corresponds to the setting position in bit vector.
Embodiment three:
One or more computer readable storage mediums, including the multiple instruction being stored thereon, the multiple instruction is rung
Ying Yu is performed and makes to calculate equipment:
By calculating the database accelerator of equipment, one or more compression elements in compression compressed data library, to generate
One or more decompression element;
Database accelerator prepares the element after one or more decompressions, generates one or more ready elements,
For accelofilter processing;With
Accelofilter is executed to generate one or more to one or more ready elements by database accelerator
A output element.
One or more computer readable storage mediums, wherein preparing the element of one or more of decompressions
Including each of the element of one or more of decompressions is formatted as the mesh compatible with the filter of the acceleration
Style formula.
One or more computer readable storage mediums further include the multiple instruction being stored thereon, the finger
It enables in response to being performed, so that the calculating equipment:
Output element is compressed by database accelerator to generate compression output element;With
The output element write-in memory of compression is responded to compression output element by database accelerator.
One or more computer readable storage mediums further include the multiple instruction being stored thereon, described more
A instruction makes the calculating equipment polymerize the output element by the database accelerator to generate in response to being performed
Aggregated data.
One or more computer readable storage mediums, in which:
The element of each decompression includes signless integer, and the element of one or more decompressions includes decompression
The packing array of element;With
Each output element includes signless integer, and one or more output elements include the packing battle array of output element
Column.
One or more computer readable storage mediums further include the multiple instruction being stored thereon, described more
A instruction make in response to being performed the calculatings equipment modified in response to the execution accelofilter it is one or
Multiple output elements.
Although describing the present invention by reference to various embodiments above, but it is to be understood that of the invention not departing from
In the case where range, many changes and modifications can be carried out.That is methods discussed above, system or equipment etc. show
Example.Various configurations can be omitted suitably, replace or add various processes or component.For example, in alternative configuration, can with
Described order in a different order executes method, and/or can add, and omits and/or combine the various stages.Moreover, about
The feature of certain configuration descriptions can be combined with various other configurations.Can combine in a similar way configuration different aspect and
Element.In addition, many elements are only range of the example without limiting the disclosure or claims with the development of technology.
Give detail in the description to provide to the thorough understanding for including the exemplary configuration realized.However,
Configuration can be practiced without these specific details for example, having been illustrated with well-known circuit, process, calculation
Method, structure and technology are without unnecessary details, to avoid fuzzy configuration.The description only provides example arrangement, and unlimited
The scope of the claims processed, applicability or configuration.It is used on the contrary, front will provide the description of configuration for those skilled in the art
Realize the enabled description of described technology.It, can be to the function of element without departing from the spirit or the scope of the present disclosure
It can and arrange and carry out various changes.
In addition, many operations can be in parallel or concurrently although each operation can describe the operations as sequential process
It executes.Furthermore it is possible to rearrange the sequence of operation.One process may have other steps.Furthermore, it is possible to pass through hardware, soft
Part, firmware, middleware, code, hardware description language or any combination thereof carry out the example of implementation method.When software, firmware, in
Between when realizing in part or code, program code or code segment for executing necessary task can store in such as storage medium
In non-transitory computer-readable medium, and described task is executed by processor.
To sum up, be intended to foregoing detailed description be considered as it is illustrative and not restrictive, and it is to be understood that described
Claim (including all equivalents) is intended to limit the spirit and scope of the present invention.The above embodiment is interpreted as only using
In illustrating the present invention rather than limit the scope of the invention.After the content for having read record of the invention, technology
Personnel can make various changes or modifications the present invention, these equivalence changes and modification equally fall into the claims in the present invention and limited
Fixed range.
Claims (10)
1. a kind of method accelerated for database, which is characterized in that the described method includes:
Database accelerator by calculating equipment decompresses one or more compression elements in compressed data library to generate one
Or multiple decompression elements;
Database accelerator prepares the element after one or more decompressions, one or more ready elements is generated, for adding
Fast filter process;With
It is one or more defeated to generate that acceleration filter is executed to one or more ready elements by database accelerator
Element out.
2. the method as described in claim 1, which is characterized in that the element for preparing one or more of decompressions includes by institute
Each of the element for stating one or more decompressions is formatted as the object format compatible with the filter of the acceleration.
3. the method as described in claim 1, which is characterized in that further include:
Output element is compressed by database accelerator to generate compression output element;With
By database accelerator by the output element write-in memory of compression to respond compression output element.
4. the method as described in claim 1, which is characterized in that further include polymerizeing output element by database accelerator to generate
Aggregated data.
5. the method as described in claim 1, which is characterized in that the element of each decompression includes signless integer, and one
The element of a or multiple decompressions includes the packing array of the element of decompression;It include signless integer with each output element,
And one or more output elements include the packing array of output element.
6. method as claimed in claim 5, which is characterized in that further include modifying institute in response to executing the acceleration filter
State one or more output elements.
7. a kind of calculating equipment accelerated for database, which is characterized in that the calculating equipment includes database accelerator,
Described in database accelerator include:
Database decompressor, for decompressing one or more compression elements in compressed data library, to generate one or more
Decompress element;
Database preparation module, it is one or more by accelerating to filter to generate for preparing the element of one or more decompressions
The preparation element of device processing;With
Filter manager executes accelerated filtration to one or more ready elements, to generate one or more output members
Element.
8. calculating equipment as claimed in claim 7, which is characterized in that the one or more of decompression elements of preparation include will
Each of one or more of decompression elements are formatted as the object format compatible with the acceleration filter;Also wrap
It includes: database compressing device, for compressing the output element to generate compression output element, wherein the calculating equipment response
Memory is written into the compression output element in the compression of the output element.
9. calculating equipment as claimed in claim 7, which is characterized in that execute and add to one or more of ready elements
Fast filter includes executing extraction, and step-by-step logical operation scans, and generates, and is converted, polymerization, sequence or setting membership;Benefit
It is executed with database accelerator and filter is accelerated to execute acceleration filter including the use of field programmable gate array;Utilize database
Accelerator, which executes, accelerates filter to execute acceleration filter including the use of specific integrated circuit;The filter manager is also used to
It polymerize the output element to generate aggregated data.
10. calculating equipment as claimed in claim 7, which is characterized in that the polymerization output element includes calculating total count,
Minimum value, maximum value or summation;Polymerizeing the output element includes: to execute arithmetical operation or statistics fortune to the output element
It calculates;
The element of each decompression includes signless integer, and the element of one or more decompressions includes the element of decompression
Packing array;It include signless integer with each output element, and one or more output elements include output element
It is packaged array;The database preparation module is also used to one or more to modify in response to the performance for accelerating filter
A output element;The bit width that one or more of output elements include: each output element of extension is modified, truncation is each
The bit width for exporting element, is attached to each output element for value, or is prepended the value of each output element;Modification institute
State one or more output elements include one or more of output elements are converted into array of indexes, wherein it is one or
Multiple output elements include bit vector, and wherein each index includes: the setting position that array of indexes corresponds in bit vector.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810831293.6A CN109086383A (en) | 2018-07-26 | 2018-07-26 | A kind of method and apparatus accelerated for database |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810831293.6A CN109086383A (en) | 2018-07-26 | 2018-07-26 | A kind of method and apparatus accelerated for database |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109086383A true CN109086383A (en) | 2018-12-25 |
Family
ID=64838566
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810831293.6A Withdrawn CN109086383A (en) | 2018-07-26 | 2018-07-26 | A kind of method and apparatus accelerated for database |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109086383A (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180150471A1 (en) * | 2016-11-29 | 2018-05-31 | Intel Corporation | Technologies for big data analytics accelerator |
-
2018
- 2018-07-26 CN CN201810831293.6A patent/CN109086383A/en not_active Withdrawn
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180150471A1 (en) * | 2016-11-29 | 2018-05-31 | Intel Corporation | Technologies for big data analytics accelerator |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10884744B2 (en) | System and method of loop vectorization by compressing indices and data elements from iterations based on a control mask | |
CN110008045A (en) | Polymerization, device, equipment and the storage medium of micro services | |
US11275568B2 (en) | Generating a synchronous digital circuit from a source code construct defining a function call | |
CN110489428B (en) | Multi-dimensional sparse matrix compression method, decompression method, device, equipment and medium | |
WO2022166258A1 (en) | Behavior recognition method and apparatus, terminal device, and computer-readable storage medium | |
CN111966631A (en) | Mirror image file generation method, system, equipment and medium capable of being rapidly distributed | |
CN113808044B (en) | Encryption mask determining method, device, equipment and storage medium | |
JP2019508774A (en) | Database system optimization method, system, electronic device and storage medium | |
JP7324891B2 (en) | Backbone network generation method, apparatus, electronic equipment, storage medium and computer program | |
CN111860557B (en) | Image processing method and device, electronic equipment and computer storage medium | |
CN109086383A (en) | A kind of method and apparatus accelerated for database | |
CN113033373A (en) | Method and related device for training face recognition model and recognizing face | |
CN112529161A (en) | Training method for generating countermeasure network, and method and device for translating human face image | |
US8930939B2 (en) | Comparing system engram with product engram to determine compatibility of product with system | |
CN115904240A (en) | Data processing method and device, electronic equipment and storage medium | |
CN111475304B (en) | Feature extraction acceleration method and system | |
CN115495212A (en) | Task queue processing method, device, equipment, storage medium and program product | |
CN115146226A (en) | Stream data processing method, device and equipment based on tensor compression method | |
CN110032445B (en) | Big data aggregation calculation method and device | |
CN107835158B (en) | Method and device for transmitting interactive data under SNA network system | |
US8270742B1 (en) | Data compression for computer-aided design systems | |
CN112100453A (en) | Method, system, equipment and computer storage medium for character string distribution statistics | |
CN115034198B (en) | Method for optimizing computation of embedded module in language model | |
CN113760765B (en) | Code testing method and device, electronic equipment and storage medium | |
CN113962191B (en) | Sequence randomization method and device, electronic 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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181225 |