CN108228801A - A kind of skip list multithreading optimization method and equipment based on multi-core processor - Google Patents
A kind of skip list multithreading optimization method and equipment based on multi-core processor Download PDFInfo
- Publication number
- CN108228801A CN108228801A CN201711480205.4A CN201711480205A CN108228801A CN 108228801 A CN108228801 A CN 108228801A CN 201711480205 A CN201711480205 A CN 201711480205A CN 108228801 A CN108228801 A CN 108228801A
- Authority
- CN
- China
- Prior art keywords
- data
- lock
- list
- read
- inquiry
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
- G06F16/24532—Query optimisation of parallel queries
-
- 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
- G06F16/23—Updating
- G06F16/2365—Ensuring data consistency and integrity
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)
- Computer Security & Cryptography (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of skip list multithreading optimization methods and equipment based on multi-core processor, belong to technical field of data processing, are 0 including initialization Read-Write Locks Lock；Skip list is successively searched since the highest level index data of skip list, obtain inquiry data or is inserted into data index data nearest in data list；When operation is data query, read lock is applied for Read-Write Locks Lock, then entering the data list in a manner of parallel processing performs inquiry operation；When operation is inserted into for data, lock is write to Read-Write Locks Lock applications, insertion operation is performed subsequently into the data list.When being performed simultaneously multiple queries or insertion operation in multiple processing cores of multi-core processor, by the locking operation to entire data list, ensure to be inserted into the data consistency with inquiry operation.The resource of multi-core processor can be made full use of in this way, improve the efficiency that data are inserted into and are inquired.
Description
Technical field
The present invention relates to technical field of data processing, more particularly to a kind of skip list multithreading based on multi-core processor is excellent
Change method and apparatus.
Background technology
Memory database refers to all putting data into the database operated in memory, with disk as main storage
The database of medium is compared, and memory database is mainly characterized by that speed is fast, handling capacity is high.At present in big data processing system,
The buffer memory of data is carried out usually using memory database, to promote the process performance of frequently-used data.It is indexed in database
The dictionary structure of (mainly inverted index) directly influences the effect of index, needs to find one in index structure and is suitble to work as
The data structure of preceding business is stored.Wherein, data structure generally comprise skip list, B+ trees, prefix trees, suffix tree, from
Dynamic state machine and Hash table these types data structure.
Wherein, skip list refers on the basis of ordered list, increases a kind of number of index data in a manner of randomization
According to structure.By these additional index datas when searching in lists in this way, it rapidly can skip partial list and be promoted
Search speed.Skip list have the characteristics that it is simple in structure, efficient, it is main in Redis, Memsql etc. as index data structure
It is used widely in stream memory database.
Multi-core processor provides multiple physics and performs core so that may be performed simultaneously multiple threads on processor, this will
The parallel processing capability of processor is substantially improved.For skip list, when having, multiple queries or insertion operation on multi-core processor are same
During Shi Zhihang, necessary synchronization mechanism is needed to ensure the consistency of data.
Existing skip list multi-thread concurrent algorithm is since synchronization mechanism is inefficient, hardware resource utilization is low, parallel efficiency
Low drawback and influence parallel search and be inserted into performance.
Invention content
The purpose of the present invention is to provide a kind of skip list multithreading optimization methods based on multi-core processor, are jumped with improving
Parallel search and insertion performance in the list data structure that jumps.
In order to achieve the above object, the technical solution adopted by the present invention is：
On the one hand, a kind of skip list multithreading optimization method based on multi-core processor is provided, including：
S1, initialization Read-Write Locks Lock are 0, ongoing data manipulation in Read-Write Locks Lock real-time recorded data lists
State；
S2, skip list is successively searched since the highest level index data of skip list, obtains inquiry data or insert number
According to nearest index data in data list；
S3, when operation for data query when, to Read-Write Locks Lock apply read lock, institute is then entered in a manner of parallel processing
It states data list and performs inquiry operation；
S4, when operation is inserted into for data, lock is write to Read-Write Locks Lock applications, is performed subsequently into the data list slotting
Enter operation.
Preferably, the data list that entered in a manner of parallel processing in step S3 performs inquiry operation, specific to wrap
It includes：
S31, check whether the value of variables L ock is greater than or equal to zero, if performing step S32, after hanging up 0.1 second if not
Continue to execute step S31；
S32,1 is added to Lock execution atomic operations, is continued to execute if running succeeded and jumping to S33, if unsuccessfully jumping to S31
It continues to execute；
S33, inquiry operation is performed into data list；
S34, subtract 1 to Lock execution atomic operations, terminate inquiry operation；
Preferably, the entrance data list in step S4 performs insertion operation, specifically includes：
S41, it checks whether the value of variables L ock is equal to zero, if performing step S42, continues to hold after hanging up 0.1 second if not
Row step S41；
S42, subtract 1 to Lock execution atomic operations, continued to execute if running succeeded and jumping to S43, if unsuccessfully jumping to S41
It continues to execute；
S43, insertion operation is performed into data list；
S44,1 is added to Lock execution atomic operations, terminates insertion operation.
Second aspect provides a kind of skip list multithreading optimization equipment based on multi-core processor, including：
Processor and storage device；
Processor is used to implement each instruction；
Storage device, for storing each instruction, each instruction is loaded and is performed by processor：
It is 0 to initialize Read-Write Locks Lock, ongoing data manipulation shape in Read-Write Locks Lock real-time recorded data lists
State；
Skip list is successively searched since the highest level index data of skip list, inquiry data is obtained or insertion data exists
Nearest index data in data list；
When operation for data query when, to Read-Write Locks Lock apply read lock, then entered in a manner of parallel processing described in
Data list performs inquiry operation；
When operation is inserted into for data, lock is write to Read-Write Locks Lock applications, insertion is performed subsequently into the data list
Operation.
Preferably, the processor is performed enters data list execution inquiry operation, tool in a manner of parallel processing
Body includes：
A, it checks whether the value of variables L ock is greater than or equal to zero, if so, performing step b, continues after hanging up 0.1 second if not
Perform step a；
B, atomic operation is performed to Lock and adds 1, continued to execute if running succeeded and jumping to step c, if unsuccessfully jumping to step
Rapid a is continued to execute；
C, inquiry operation is performed into data list；
D, atomic operation is performed to Lock and subtracts 1, terminate inquiry operation；
Preferably, the processor, which performs, enters data list execution insertion operation, specifically includes：
E, it checks whether the value of variables L ock is equal to zero, if so, performing step f, step is continued to execute after hanging up 0.1 second if not
Rapid e；
F, atomic operation is performed to Lock and subtracts 1, continued to execute if running succeeded and jumping to g, continue to hold if unsuccessfully jumping to e
Row；
G, insertion operation is performed into data list；
H, atomic operation is performed to Lock and adds 1, terminate insertion operation.
The third aspect provides a kind of skip list multithreading optimization equipment based on multi-core processor, including：
Initialization module, for initializing Read-Write Locks Lock as 0, in Read-Write Locks Lock real-time recorded data lists into
Capable data manipulation state；
Searching module for successively searching skip list since the highest level index data of skip list, obtains inquiry data
Or it is inserted into data index data nearest in data list；
Inquiry operation execution module is applied for read lock to Read-Write Locks Lock, the number is then entered in a manner of parallel processing
Inquiry operation is performed according to list；
Insertion operation execution module writes lock to Read-Write Locks Lock applications, is performed subsequently into the data list and be inserted into behaviour
Make.
Preferably, inquiry operation execution module specifically includes：
First checks unit, for checking whether the value of variables L ock is greater than or equal to zero；
Inquiry operation execution unit, for checking unit judges when being, to Lock plus 1, and to enter data and arranging first
Table performs inquiry operation, subtracts 1 to Lock after operation；
And first check unit judges for it is no when, wait until Lock be greater than or equal to zero when perform inquiry operation again.
Preferably, insertion operation execution module specifically includes：
Second checks unit, for checking whether the value of variables L ock is equal to zero；
Insertion operation execution unit for checking unit judges when being, to subtract 1 to Lock second, and enters data and arranges
Table performs insertion operation, adds 1 to Lock after operation；
And second check unit judges for it is no when, wait until Lock be equal to zero when perform insertion operation again.
Compared with prior art, there are following technique effects by the present invention：The present invention passes through at multiple places of multi-core processor
Multiple or inquiry operation is performed simultaneously in reason core, by entire data list locking operation, ensureing insertion and inquiry operation
Data consistency.In this way can abundant processor resource utilization, improve data and be inserted into and the efficiency of inquiry.
Description of the drawings
Below in conjunction with the accompanying drawings, the specific embodiment of the present invention is described in detail：
Fig. 1 is a kind of flow diagram of the skip list multithreading optimization method based on multi-core processor；
Fig. 2 is the inquiry of skip list and the schematic diagram of insertion operation；
Fig. 3 is parallel query operating process schematic diagram in skip list；
Fig. 4 is insertion operation flow diagram in skip list；
Fig. 5 is a kind of structure diagram of the skip list multithreading optimization equipment based on multi-core processor.
Specific embodiment
In order to illustrate further the feature of the present invention, reference should be made to the following detailed description and accompanying drawings of the present invention.Institute
Attached drawing is only for reference and purposes of discussion, is not used for limiting protection scope of the present invention.
As shown in Figure 1, present embodiment discloses a kind of skip list multithreading optimization method based on multi-core processor, including
Following steps S1 to S4
S1, initialization Read-Write Locks Lock are 0, ongoing data manipulation in Read-Write Locks Lock real-time recorded data lists
State；
S2, skip list is successively searched since the highest level index data of skip list, obtains inquiry data or insert number
According to nearest index data in data list；
S3, when operation for data query when, to Read-Write Locks Lock apply read lock, institute is then entered in a manner of parallel processing
It states data list and performs inquiry operation；
S4, when operation is inserted into for data, lock is write to Read-Write Locks Lock applications, is performed subsequently into the data list slotting
Enter operation.
Specifically, carry out the ongoing data manipulation in real-time recorded data list in the present embodiment using variables L ock
State.When Lock is more than 0, indicate that multiple queries operation actually performs in data list；When Lock is less than 0,1 is indicated
Insertion operation performs in data list；When Lock is equal to 0, the specific inquiry not being carrying out in data list at this time is represented
Or insertion operation.Therefore in initialization, the value for locking variables L ock is set to 0, represents do not have any operation in data list.
As shown in Fig. 2, when successively consulting skip list, Search (K) is denoted as the operation to inquiring data K, Insert
(J) it is denoted as the operation for being inserted into data J.Search (K) and Insert (J) operations are carrying out being inserted into data or are inquiring data manipulation
When, it is required to inquire data K and data J indexes nearest in data list in the multilayer index data of skip list in advance
Data.The inquiry operation of skip list successively carries out as shown in Figure 2.
It should be noted that since any content is not changed in the inquiry to skip list, so Search (K) and Insert
(J) inquiry operation can perform parallel, the consistency without destroying skip list index data structure.
Preferably, for the inquiry operation of data list, multiple queries can be allowed to be carried out at the same time.Whether to first look at
There is inquiry operation, if there is inquiry operation or without any operation, be then directly entered inquiry.If having insertion operation,
It waits for until insertion operation terminates to enter back into.The value of variables L ock has recorded the mode of operation of data list.As shown in figure 3, step
Rapid S3 specifically comprises the following steps：
S31, check whether the value of variables L ock is greater than or equal to 0, if so then execute step S33, if otherwise performing step
S32；
S32, after current queries are operated thread suspension setting time, step S31 is re-executed；
S33, atomic operation is performed, enables Lock=Lock+1；
S34, check step S33 atomic operation whether successful execution, continued to execute if so then execute step S35, if not
Then perform step S31；
S35, in the data list, perform inquiry operation；
S36, atomic operation is performed, enables Lock=Lock-1；
S37, check step S36 atomic operation whether successful execution, if then flow terminates, if otherwise performing step
S39；
S39, after current queries are operated thread suspension setting time, step S36 is performed.
Preferably, to the insertion operation of data list, only 1 operation is allowed to be performed in data list.First looking at is
If no existing inquiry or insertion operation without any operation, are directly entered and perform data list insertion operation.If have
Inquiry or insertion operation are then waited for until the operation in data list terminates to enter back into.The value of variables L ock has recorded data row
The mode of operation of table.As shown in figure 4, step S4 is specifically included：
S41, check whether the value of variables L ock is equal to 0, if so then execute step S43, if otherwise performing step S42；
S42, after operation thread suspension setting time will be currently inserted into, step S41 is then performed；
S43, atomic operation is performed, enables Lock=Lock-1；
S44, check step S43 atomic operation whether successful execution, if so then execute step S45, if otherwise performing step
Rapid S41；
S45, in the data list, perform insertion operation；
S46, atomic operation is performed, enables Lock=Lock+1；
S47, check step S46 atomic operation whether successful execution, if then flow terminates, if otherwise performing step
S48；
S48, step S46 will be performed after operation thread suspension setting time will be currently inserted into.
Specifically, in practical applications, current queries operation thread can be set according to actual conditions or be currently inserted into operation
The suspension time of thread, such as 0.1 second etc..
It should be noted that the present embodiment is operated by the lock to data list, multiple queries operation can be realized simultaneously
It is searched into data list.Because inquiry operation does not change the numerical value of data list, multiple queries is allowed to operate so same
Consistency of the Shi Jinhang without influencing data.Since inquiry operation will be far more than insertion operation, so originally in practical application scene
Scheme can effectively promote the concurrent processing efficiency of skip list.
The present embodiment also discloses a kind of skip list multithreading optimization equipment based on multi-core processor, including processor and
Storage device；
Processor is used to implement each instruction；
Storage device, for storing each instruction, each instruction is loaded and is performed by processor：
It is 0 to initialize Read-Write Locks Lock, ongoing data manipulation shape in Read-Write Locks Lock real-time recorded data lists
State；
Skip list is successively searched since the highest level index data of skip list, inquiry data is obtained or insertion data exists
Nearest index data in data list；
When operation for data query when, to Read-Write Locks Lock apply read lock, then entered in a manner of parallel processing described in
Data list performs inquiry operation；
When operation is inserted into for data, lock is write to Read-Write Locks Lock applications, insertion is performed subsequently into the data list
Operation.
Preferably, processor is performed enters data list execution inquiry operation in a manner of parallel processing, specific to wrap
It includes：
A, check whether the value of variables L ock is greater than or equal to zero；
B, if so, adding 1 to Lock and entering data list execution inquiry operation, subtract 1 to Lock after operation；
C, step b is performed again when Lock is greater than or equal to zero if it is not, waiting until.
Preferably, processor, which performs, enters data list execution insertion operation, specifically includes：
D, check whether the value of variables L ock is equal to zero；
E, if so, subtracting 1 to Lock, and enter data list and perform insertion operation, 1 is added to Lock after operation；
F, the step e is performed again when Lock is equal to zero if it is not, waiting until.
As shown in figure 5, a kind of skip list multithreading optimization equipment based on multi-core processor, packet is also disclosed in the present embodiment
It includes：
Initialization module 10 is 0 for initializing Read-Write Locks Lock, in Read-Write Locks Lock real-time recorded data lists
The data manipulation state of progress；
Searching module 20 for successively searching skip list since the highest level index data of skip list, obtains inquiry number
According to or be inserted into data index data nearest in data list；
Inquiry operation execution module 30, to Read-Write Locks Lock apply read lock, then entered in a manner of parallel processing described in
Data list performs inquiry operation；
Insertion operation execution module 40 writes lock to Read-Write Locks Lock applications, insertion is performed subsequently into the data list
Operation.
Preferably, inquiry operation execution module specifically includes：
First checks unit, for checking whether the value of variables L ock is greater than or equal to zero；
Inquiry operation execution unit, for checking unit judges when being, to Lock plus 1, and to enter data and arranging first
Table performs inquiry operation, subtracts 1 to Lock after operation；
And first check unit judges for it is no when, wait until Lock be greater than or equal to zero when perform inquiry operation again.
Preferably, insertion operation execution module specifically includes：
Second checks unit, for checking whether the value of variables L ock is equal to zero；
Insertion operation execution unit for checking unit judges when being, to subtract 1 to Lock second, and enters data and arranges
Table performs insertion operation, adds 1 to Lock after operation；
And second check unit judges for it is no when, wait until Lock be equal to zero when perform insertion operation again.
It should be noted that a kind of skip list multithreading optimization equipment based on multi-core processor disclosed in the present embodiment,
With above-described embodiment disclosed in the skip list multithreading optimization method based on multi-core processor have the corresponding technology of identical Hu thin
Section and identical technique effect are not being repeated at this.
Existing skip list parallel query and interleaved plan are realized based on traditional lock synchronization mechanism.When data arrange
When having thread execution data query operation in table, other data query threads will be blocked, and can not realize data query thread
Concurrently perform.This programme is set by the Read-Write Locks to Lock, can be under the premise of data consistency is ensured, multi-core processor
Multiple threads simultaneously data query operation is performed in data list.Due in memory database practical application scene, data
Inquiry operation be much larger than data insertion operation, so this programme can effectively be promoted skip list in actual scene concurrently inquire with
The execution efficiency of insertion operation.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.
Claims (9)
1. a kind of skip list multithreading optimization method based on multi-core processor, which is characterized in that including：
S1, initialization Read-Write Locks Lock are 0, ongoing data manipulation shape in Read-Write Locks Lock real-time recorded data lists
State；
S2, skip list is successively searched since the highest level index data of skip list, obtains inquiry data or insertion data exist
Nearest index data in data list；
S3, when operation for data query when, to Read-Write Locks Lock apply read lock, then in a manner of parallel processing enter the number
Inquiry operation is performed according to list；
S4, when operation for data be inserted into when, to Read-Write Locks Lock application write lock, subsequently into the data list perform be inserted into grasp
Make.
2. the skip list multithreading optimization method based on multi-core processor as described in claim 1, which is characterized in that the step
The data list that entered in a manner of parallel processing in rapid S3 performs inquiry operation, specifically includes：
S31, it checks whether the value of variables L ock is greater than or equal to zero, if performing step S32, continues after hanging up 0.1 second if not
Perform step S31；
S32,1 is added to Lock execution atomic operations, is continued to execute if running succeeded and jumping to S33, if unsuccessfully jumping to S31 continuation
It performs；
S33, inquiry operation is performed into data list；
S34, subtract 1 to Lock execution atomic operations, terminate inquiry operation.
3. the skip list multithreading optimization method based on multi-core processor as described in claim 1, which is characterized in that the step
The entrance data list in rapid S4 performs insertion operation, specifically includes：
S41, it checks whether the value of variables L ock is equal to zero, if performing step S42, step is continued to execute after hanging up 0.1 second if not
Rapid S41；
S42, subtract 1 to Lock execution atomic operations, continued to execute if running succeeded and jumping to S43, if unsuccessfully jumping to S41 continuation
It performs；
S43, insertion operation is performed into data list；
S44,1 is added to Lock execution atomic operations, terminates insertion operation.
4. a kind of skip list multithreading optimization equipment based on multi-core processor, which is characterized in that including：Processor and storage are set
It is standby；
Processor is used to implement each instruction；
Storage device, for storing each instruction, each instruction is loaded and is performed by processor：
It is 0 to initialize Read-Write Locks Lock, ongoing data manipulation state in Read-Write Locks Lock real-time recorded data lists；
Skip list is successively searched since the highest level index data of skip list, obtain inquiry data or is inserted into data in data
Nearest index data in list；
When operation is data query, read lock is applied to Read-Write Locks Lock, the data are then entered in a manner of parallel processing
List performs inquiry operation；
When operation is inserted into for data, lock is write to Read-Write Locks Lock applications, insertion operation is performed subsequently into the data list.
5. the skip list multithreading optimization equipment based on multi-core processor as claimed in claim 4, which is characterized in that the place
It manages device and performs and enter data list execution inquiry operation in a manner of parallel processing, specifically include：
A, it checks whether the value of variables L ock is greater than or equal to zero, if so, performing step b, is continued to execute after hanging up 0.1 second if not
Step a；
B, to Lock perform atomic operation add 1, continued to execute if running succeeded and jumping to step c, if unsuccessfully jump to step a after
It is continuous to perform；
C, inquiry operation is performed into data list；
D, atomic operation is performed to Lock and subtracts 1, terminate inquiry operation.
6. the skip list multithreading optimization equipment based on multi-core processor as claimed in claim 4, which is characterized in that the place
It manages device and performs and enter data list execution insertion operation, specifically include：
E, it checks whether the value of variables L ock is equal to zero, if so, performing step f, step e is continued to execute after hanging up 0.1 second if not；
F, atomic operation is performed to Lock and subtracts 1, continued to execute if running succeeded and jumping to g, continued to execute if unsuccessfully jumping to e；
G, insertion operation is performed into data list；
H, atomic operation is performed to Lock and adds 1, terminate insertion operation.
7. a kind of skip list multithreading optimization equipment based on multi-core processor, which is characterized in that including：
Initialization module is 0 for initializing Read-Write Locks Lock, ongoing in Read-Write Locks Lock real-time recorded data lists
Data manipulation state；
Searching module, for successively searching skip list since the highest level index data of skip list, obtain inquiry data or
It is inserted into data index data nearest in data list；
Inquiry operation execution module is applied for read lock to Read-Write Locks Lock, is held in a manner of parallel processing into the data list
Row inquiry operation；
Insertion operation execution module writes lock to Read-Write Locks Lock applications, insertion operation is performed into the data list.
8. the skip list multithreading optimization equipment based on multi-core processor as claimed in claim 7, which is characterized in that described to look into
Operation executing module is ask, is specifically included：
First checks unit, for checking whether the value of variables L ock is greater than or equal to zero；
Inquiry operation execution unit, for checking unit judges when being, to be held to Lock plus 1, and into data list first
Row inquiry operation, subtracts 1 after operation to Lock；
And first check unit judges for it is no when, wait until Lock be greater than or equal to zero when perform inquiry operation again.
9. the skip list multithreading optimization equipment based on multi-core processor as claimed in claim 7, which is characterized in that described to insert
Enter operation executing module, specifically include：
Second checks unit, for checking whether the value of variables L ock is equal to zero；
Insertion operation execution unit for checking unit judges when being, to subtract 1 to Lock second, and is held into data list
Row insertion operation, adds 1 after operation to Lock；
And second check unit judges for it is no when, wait until Lock be equal to zero when perform insertion operation again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711480205.4A CN108228801B (en) | 2017-12-29 | 2017-12-29 | Jump table multithreading optimization method and device based on multi-core processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711480205.4A CN108228801B (en) | 2017-12-29 | 2017-12-29 | Jump table multithreading optimization method and device based on multi-core processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108228801A true CN108228801A (en) | 2018-06-29 |
| CN108228801B CN108228801B (en) | 2021-08-31 |
Family
ID=62646131
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711480205.4A Active CN108228801B (en) | 2017-12-29 | 2017-12-29 | Jump table multithreading optimization method and device based on multi-core processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108228801B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116069267A (en) * | 2023-04-06 | 2023-05-05 | 苏州浪潮智能科技有限公司 | Writing cache method, system, equipment and storage medium of RAID card |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090132563A1 (en) * | 2007-11-19 | 2009-05-21 | Sun Microsystems, Inc. | Simple optimistic skiplist |
| CN101853279A (en) * | 2010-05-13 | 2010-10-06 | 复旦大学 | Improved lockout method of internal memory database |
| CN102880585A (en) * | 2012-09-28 | 2013-01-16 | 无锡江南计算技术研究所 | Synchronizer for processor system with multiple processor cores |
| CN103279428A (en) * | 2013-05-08 | 2013-09-04 | 中国人民解放军国防科学技术大学 | Explicit multi-core Cache consistency active management method facing flow application |
| US20160098361A1 (en) * | 2014-10-07 | 2016-04-07 | Check Point Software Technologies Ltd. | Optimization of Data Locks for Improved Write Lock Performance and CPU Cache usage in Mulit Core Architectures |
-
2017
- 2017-12-29 CN CN201711480205.4A patent/CN108228801B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090132563A1 (en) * | 2007-11-19 | 2009-05-21 | Sun Microsystems, Inc. | Simple optimistic skiplist |
| CN101853279A (en) * | 2010-05-13 | 2010-10-06 | 复旦大学 | Improved lockout method of internal memory database |
| CN102880585A (en) * | 2012-09-28 | 2013-01-16 | 无锡江南计算技术研究所 | Synchronizer for processor system with multiple processor cores |
| CN103279428A (en) * | 2013-05-08 | 2013-09-04 | 中国人民解放军国防科学技术大学 | Explicit multi-core Cache consistency active management method facing flow application |
| US20160098361A1 (en) * | 2014-10-07 | 2016-04-07 | Check Point Software Technologies Ltd. | Optimization of Data Locks for Improved Write Lock Performance and CPU Cache usage in Mulit Core Architectures |
Non-Patent Citations (1)
| Title |
|---|
| NURIT MOSCOVICI: "A GPU-Friendly Skiplist Algorithm", 《2017 26TH INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES (PACT)》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116069267A (en) * | 2023-04-06 | 2023-05-05 | 苏州浪潮智能科技有限公司 | Writing cache method, system, equipment and storage medium of RAID card |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108228801B (en) | 2021-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11693830B2 (en) | Metadata management method, system and medium | |
| CN105468298B (en) | A kind of key assignments storage method based on log-structured merging tree | |
| CN105117417B (en) | A kind of memory database Trie tree indexing means for reading optimization | |
| CN105786448B (en) | A kind of instruction dispatching method and device | |
| CN110083601A (en) | Index tree constructing method and system towards key assignments storage system | |
| CN105095425B (en) | A kind of inter-library method and device of carrying down of database | |
| US20240104059A1 (en) | Method for Service Processing and System, Device, and Medium | |
| CN107153643A (en) | Tables of data connection method and device | |
| CN104601562A (en) | Interactive method and system of game server and database | |
| US9785695B2 (en) | Providing multiple concurrent transactions on a single database schema using a single concurrent transaction database infrastructure | |
| CN104424222B (en) | Database index method and device | |
| CN107908714A (en) | A kind of aggregation of data sort method and device | |
| CN108228801A (en) | A kind of skip list multithreading optimization method and equipment based on multi-core processor | |
| CN106874343A (en) | The data-erasure method and system of a kind of time series database | |
| US10083065B2 (en) | Creating multiple rules for a device to allow concurrent access to the device by different virtual machines | |
| CN110442594A (en) | A kind of Dynamic Execution method towards Spark SQL Aggregation Operators | |
| US8490098B2 (en) | Concomitance scheduling commensal threads in a multi-threading computer system | |
| CN111198660A (en) | B + tree traversal method and device | |
| CN109582467A (en) | Processing method, system and the relevant apparatus of I/O request in a kind of storage system | |
| CN116069480B (en) | Processor and computing device | |
| CN108021678B (en) | Key value pair storage structure with compact structure and quick key value pair searching method | |
| CN103353891B (en) | Data base management system and its data processing method | |
| CN113722320A (en) | Graph data index persistence method based on parallelism | |
| US20140214827A1 (en) | Data Caveats for Database Tables | |
| JPS63307553A (en) | File control system |
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 | ||
| CB02 | Change of applicant information |
Address after: 230000 Yafu Park, Juchao Economic Development Zone, Chaohu City, Hefei City, Anhui Province Applicant after: ANHUI HUISHI JINTONG TECHNOLOGY Co.,Ltd. Address before: 102, room 602, C District, Hefei National University, Mount Huangshan Road, 230000 Hefei Road, Anhui, China Applicant before: ANHUI HUISHI JINTONG TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |