CN105117415A - Optimized SSD data updating method - Google Patents
Optimized SSD data updating method Download PDFInfo
- Publication number
- CN105117415A CN105117415A CN201510458844.5A CN201510458844A CN105117415A CN 105117415 A CN105117415 A CN 105117415A CN 201510458844 A CN201510458844 A CN 201510458844A CN 105117415 A CN105117415 A CN 105117415A
- Authority
- CN
- China
- Prior art keywords
- data
- ssd
- resident
- memory
- disk
- 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
- G06F16/23—Updating
- G06F16/2365—Ensuring data consistency and integrity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/1847—File system types specifically adapted to static storage, e.g. adapted to flash memory or SSD
-
- 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
- G06F16/2246—Trees, e.g. B+trees
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)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an optimized SSD data updating method. For a character type of data stored on an SSD, and with combined use of the two data structures of a line segment tree and a B tree, space occupied by an index structure is reduced with query efficiency being ensured; when character data is updated, different values of a same key in two data blocks are updated by using of a characteristic of a set union-intersection calculation, and other data is written back as original; and random updating of the character data on the SSD is converted to a sequential writing of data by using an LSM method based on a line segment B tree. The method disclosed by the invention effectively avoids a writing amplification problem of the SSD, increases a data writing speed of the SSD, and improves the operation efficiency of a database.
Description
Technical field
The invention belongs to field of computer technology, be specifically related to a kind of SSD data-updating method of optimization.
Background technology
When NoSQL database design, needing can persistent storage by the data in internal memory.Use SSD (solid state hard disc) as can persistent storage time, the read-write processing power of data can be accelerated, the overall performance of elevator system.Therefore write back in SSD by data, the two-level memory framework of composition internal memory-SSD, for application provides capacity larger, compared to the mechanical disk storage of tradition, access speed is Database Systems faster.In order to the quick data to being stored on SSD conduct interviews, need to set up index to data.But traditional B tree, the index structures such as B+ tree can produce the request of a large amount of random I/O operations.If the write SSD directly a large amount of random I/O request msgs not being added process can cause the serious reduction of SSD performance.Because when upper level applications needs to upgrade the data on SSD, the write amplification characteristic that SSD is intrinsic, adds the delay of write operation.
The people such as O'Nei are in the thought of the log-structured merging of proposition inwardly (LogStructureMerge) method of Journaling File System, according to the constantly additional write feature of log information, in conjunction with B data tree structure, sacrificial section reads performance, be used for significantly improving write performance, between read-write, obtain the balance of better performance.This LSM method provides a kind of delay update mechanism, the renewal rewards theory of random small data quantity is fused to the renewal rewards theory of large succession, improves the bandwidth availability ratio stored.The These characteristics of LSM method makes it write in the two-level memory framework of internal memory-SSD and shows good performance more than in the Database Systems read.
But there will be some new problems during LSM method directly acts on based on SSD NoSQL Database Systems.LSM method based on key word for integer type designs, be in the database of the character string type of key-value in storage key, if set up index to the key value of character string type, when again index structure being write back SSD, need to change, because SSD directly can not store the data of pointer type.On the other hand, the data of character string type are due to its form, the diversity of length etc., its data cannot be represented with the space distributed in advance, and separately for each key value sets up index, the data volume in the b-tree indexed structure in LSM method in SSD will be made to increase, in this case data are conducted interviews, accessed path will be caused to increase, and data access delay increases.For this series of problem, the present invention proposes a kind of improvement index structure for character string type data, can accelerate the access speed of character string type data on SSD.
Summary of the invention
In order to overcome the shortcoming of above-mentioned prior art, the object of the present invention is to provide a kind of SSD data-updating method of optimization, improve data access efficiency.
In order to achieve the above object, the technical scheme that the present invention takes is:
A SSD data-updating method for optimization, comprises the following steps:
The first step, set up line segment B tree construction:
In memory database, first utilize the shared prefix information of string data, the character string of shared same prefix is formed a character string interval; Then being utilized by this interval censored data B tree construction to write algorithm is inserted in line segment B tree; Finally, the logical view of whole data structure is a B tree, but its keyword strings block information stored;
Second step, use pLSM method to complete the renewal rewards theory of data, the merging process of the pLSM method between multicompartment is identical with the merging process step of the pLSM method of two assemblies, and the merging process of the pLSM method of two assemblies is as follows:
1) from memory-resident C
0in read in the leaf node data do not merged, insert and merge in block;
2) from disk resident C
1in read in the leaf node data do not merged, insert and merge in block;
3) data in involutory blocking carry out merge sort, if when running into equal key word, with memory-resident C
0in data be latest data, carry out renewal rewards theory;
4) repeat step 1), 2), 3), when merging block and being full, additional writes back disk, then again reads memory-resident C
0with disk resident C
1in the leaf node data that do not merge;
5) as memory-resident C
0with disk resident C
1all leaf nodes all carried out union operation after, represent that a merging process terminates, memory-resident C
0in renewal rewards theory data be integrated in disk;
By above-mentioned combining step, for the renewal of data on SSD dish, structure in first write memory, then adopt the strategy merged step by step to write back SSD, writing back in process, directly the data in SSD are not being modified, but produce the data after renewal by merging process, and add in the new file of write, after merging completes, delete ancient deed;
3rd step, uses line segment B tree construction to provide index to the data on SSD dish:
After using pLSM method, owing to storing multiple data file in SSD dish, for the search operation of data, be first arranged in the memory-resident C of internal memory
0search in structure, if do not found, search in disk structure so step by step, from disk resident C
1, C
2until C
k, until find;
In search procedure, if same data field occurs in multiple structure simultaneously, pLSM method ensures can fetch up-to-date and correct data in each reading, in search operation, if in do not find desired data, just from disk, read data;
When searching in the index structure in SSD dish, key word be relatively converted into test data to be found whether in interval, if find the block information comprising data to be found, then read its side-play amount in the data file, by document misregistration amount visit data file, obtain the value that key to be found is corresponding, whole interval censored data in once accessing is buffered in internal memory, form a buffer zone, when the same interval inner data access of next time arrives, then directly to operate in internal memory, the tissue of buffering adopts doubly linked list tissue, and eliminate by LRU method the buffered data be of little use.
The invention has the beneficial effects as follows:
Present invention achieves a kind of LSM method of optimization, pLSM method, provide the update strategy of the character string type data stored for SSD.On the one hand, use pLSM method, by upgrading the file appending write being converted into order to the random data of SSD dish files, avoid the write scale-up problem of SSD.On the other hand, use line segment B tree construction to provide index for the data in SSD dish, avoid whole traversals during data search, data search time complexity is reduced to O (LogN) from O (N), improves data access efficiency.
Accompanying drawing explanation
Fig. 1 is two assembly LSM structural representations.
Fig. 2 is LSM method merging process schematic diagram.
Embodiment
A SSD data-updating method for optimization, comprises the following steps:
The first step, set up line segment B tree construction:
In memory database, first utilize the shared prefix information of string data, the character string of shared same prefix is formed a character string interval; Then being utilized by this interval censored data B tree construction to write algorithm is inserted in line segment B tree; Finally, the logical view of whole data structure is a B tree, but its keyword strings block information stored;
Second step, use pLSM method to complete the renewal rewards theory of data, the merging process of the pLSM method between multicompartment is identical with the merging process step of the pLSM method of two assemblies, and with reference to Fig. 1, the merging process of the pLSM method of two assemblies is as follows:
1) from memory-resident C
0in read in the leaf node data do not merged, insert and merge in block;
2) from disk resident C
1in read in the leaf node data do not merged, insert and merge in block;
3) data in involutory blocking carry out merge sort, if when running into equal key word, with memory-resident C
0in data be latest data, carry out renewal rewards theory, such as, memory-resident C
0middle key is the Data Identification of 001 is deletion, and C
1middle key is the data of 001 correspondence is 12345, so merge in block carry out merge sort time, due to memory-resident C
0in data be that latest data upgrades, so delete the data of key corresponding to 001;
4) repeat step 1), 2), 3), when merging block and being full, additional writes back disk, then again reads memory-resident C
0with disk resident C
1in the leaf node data that do not merge;
5) as memory-resident C
0with disk resident C
1all leaf nodes all carried out union operation after, represent that a merging process terminates, memory-resident C
0in renewal rewards theory data be integrated in disk, as shown in Figure 2;
Above-mentioned steps is the merging process of the pLSM algorithm of two assemblies, and the merging process between multicompartment is identical with above-mentioned steps;
By above-mentioned combining step, for the renewal of data on SSD dish, structure in first write memory, then the strategy merged step by step is adopted to write back SSD, writing back in process, directly the data in SSD are not being modified, but producing the data after renewal by merging process, and add in the new file of write, after merging completes, delete ancient deed, adopt in this way, effectively prevent the write scale-up problem of SSD, improve data and write back efficiency;
3rd step, uses line segment B tree construction to provide index to the data on SSD dish:
After using pLSM method, owing to storing multiple data file in SSD dish, for the search operation of data, be first arranged in the memory-resident C of internal memory
0search in structure, if do not found, search in disk structure so step by step, from disk resident C
1, C
2until C
k, until find.
In search procedure, may occur in multiple structure by same data field, and pLSM algorithm can ensure can fetch up-to-date and correct data in each reading simultaneously.This is because, pLSM tree read time based on following hypothesis: up-to-date data are always present in the lower storage organization of rank, if namely there is identical key word in Ck and Ck-1 simultaneously, so lookup result returns the data in Ck-1, and this feature also embodies to some extent in the insertion process of data.In search operation, if in do not find desired data, just need to read data from disk, add the time overhead of search operation.
When searching in the index structure in SSD dish, and the search procedure that common B sets is similar, but, key word be relatively converted into test data to be found whether in interval, its schematic diagram is as shown in drawings.If find the block information comprising data to be found, then read its side-play amount in the data file, by document misregistration amount visit data file, obtain the value that key to be found is corresponding.Because the speed of access SSD dish is slower than the speed of access memory, in conjunction with locality access principle, whole interval censored data in once accessing is buffered in internal memory, form a buffer zone, when the same interval inner data access of next time arrives, then directly to operate in internal memory, reduce access SSD number, the tissue cushioned in the present invention adopts doubly linked list tissue, and eliminates by lru algorithm the buffered data be of little use.
Claims (1)
1. the SSD data-updating method optimized, is characterized in that, comprise the following steps:
The first step, set up line segment B tree construction:
In memory database, first utilize the shared prefix information of string data, the character string of shared same prefix is formed a character string interval; Then being utilized by this interval censored data B tree construction to write algorithm is inserted in line segment B tree; Finally, the logical view of whole data structure is a B tree, but its keyword strings block information stored;
Second step, use pLSM method to complete the renewal rewards theory of data, the merging process of the pLSM method between multicompartment is identical with the merging process step of the pLSM method of two assemblies, and the merging process of the pLSM method of two assemblies is as follows:
1) from memory-resident C
0in read in the leaf node data do not merged, insert and merge in block;
2) from disk resident C
1in read in the leaf node data do not merged, insert and merge in block;
3) data in involutory blocking carry out merge sort, if when running into equal key word, with memory-resident C
0in data be latest data, carry out renewal rewards theory;
4) repeat step 1), 2), 3), when merging block and being full, additional writes back disk, then again reads memory-resident C
0with disk resident C
1in the leaf node data that do not merge;
5) as memory-resident C
0with disk resident C
1all leaf nodes all carried out union operation after, represent that a merging process terminates, memory-resident C
0in renewal rewards theory data be integrated in disk;
By above-mentioned combining step, for the renewal of data on SSD dish, structure in first write memory, then adopt the strategy merged step by step to write back SSD, writing back in process, directly the data in SSD are not being modified, but produce the data after renewal by merging process, and add in the new file of write, after merging completes, delete ancient deed;
3rd step, uses line segment B tree construction to provide index to the data on SSD dish:
After using pLSM method, owing to storing multiple data file in SSD dish, for the search operation of data, be first arranged in the memory-resident C of internal memory
0search in structure, if do not found, search in disk structure so step by step, from disk resident C
1, C
2until C
k, until find;
In search procedure, if same data field occurs in multiple structure simultaneously, pLSM method ensures can fetch up-to-date and correct data in each reading, in search operation, if in do not find desired data, just from disk, read data;
When searching in the index structure in SSD dish, key word be relatively converted into test data to be found whether in interval, if find the block information comprising data to be found, then read its side-play amount in the data file, by document misregistration amount visit data file, obtain the value that key to be found is corresponding, whole interval censored data in once accessing is buffered in internal memory, form a buffer zone, when the same interval inner data access of next time arrives, then directly to operate in internal memory, the tissue of buffering adopts doubly linked list tissue, and eliminate by LRU method the buffered data be of little use.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510458844.5A CN105117415B (en) | 2015-07-30 | 2015-07-30 | A kind of SSD data-updating methods of optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510458844.5A CN105117415B (en) | 2015-07-30 | 2015-07-30 | A kind of SSD data-updating methods of optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105117415A true CN105117415A (en) | 2015-12-02 |
| CN105117415B CN105117415B (en) | 2018-07-03 |
Family
ID=54665405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510458844.5A Active CN105117415B (en) | 2015-07-30 | 2015-07-30 | A kind of SSD data-updating methods of optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105117415B (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106227677A (en) * | 2016-07-20 | 2016-12-14 | 浪潮电子信息产业股份有限公司 | Method for managing variable-length cache metadata |
| CN106708442A (en) * | 2016-12-30 | 2017-05-24 | 武汉安嘉颐科技有限公司 | Massive data storage method simultaneously applicable to disk and solid state disk reading and writing features |
| CN108319625A (en) * | 2017-01-17 | 2018-07-24 | 广州市动景计算机科技有限公司 | Piece file mergence method and apparatus |
| CN108319602A (en) * | 2017-01-17 | 2018-07-24 | 广州市动景计算机科技有限公司 | Data base management method and Database Systems |
| WO2018133762A1 (en) * | 2017-01-17 | 2018-07-26 | 广州市动景计算机科技有限公司 | File merging method and apparatus |
| CN109213445A (en) * | 2018-08-23 | 2019-01-15 | 郑州云海信息技术有限公司 | A kind of management method, management system and the relevant apparatus of storage system metadata |
| CN109271570A (en) * | 2018-10-30 | 2019-01-25 | 郑州云海信息技术有限公司 | A kind of method of metadata management inquiry |
| CN109407985A (en) * | 2018-10-15 | 2019-03-01 | 郑州云海信息技术有限公司 | A kind of method and relevant apparatus of data management |
| WO2019128820A1 (en) * | 2017-12-29 | 2019-07-04 | Huawei Technologies Co., Ltd. | Systems and methods for database management using append-only storage devices |
| CN110502457A (en) * | 2019-08-23 | 2019-11-26 | 北京浪潮数据技术有限公司 | A kind of metadata storing method and device |
| CN110851434A (en) * | 2018-07-27 | 2020-02-28 | 阿里巴巴集团控股有限公司 | Data storage method, device and equipment |
| CN111104403A (en) * | 2019-11-30 | 2020-05-05 | 北京浪潮数据技术有限公司 | LSM tree data processing method, system, equipment and computer medium |
| CN111831622A (en) * | 2020-03-31 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Data index generation method and device, electronic equipment and readable storage medium |
| CN112487095A (en) * | 2020-12-09 | 2021-03-12 | 浪潮云信息技术股份公司 | Method for optimizing transaction data storage of distributed database |
| US11537582B2 (en) | 2021-04-16 | 2022-12-27 | Samsung Electronics Co., Ltd. | Data access method, a data access control device, and a data access system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722449A (en) * | 2012-05-24 | 2012-10-10 | 中国科学院计算技术研究所 | Key-Value local storage method and system based on solid state disk (SSD) |
| US20150058568A1 (en) * | 2013-08-26 | 2015-02-26 | International Business Machines Corporation | HIERARCHICAL STORAGE FOR LSM-BASED NoSQL STORES |
| CN104461384A (en) * | 2014-11-28 | 2015-03-25 | 华为技术有限公司 | Data write-in method and storage device |
-
2015
- 2015-07-30 CN CN201510458844.5A patent/CN105117415B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722449A (en) * | 2012-05-24 | 2012-10-10 | 中国科学院计算技术研究所 | Key-Value local storage method and system based on solid state disk (SSD) |
| US20150058568A1 (en) * | 2013-08-26 | 2015-02-26 | International Business Machines Corporation | HIERARCHICAL STORAGE FOR LSM-BASED NoSQL STORES |
| CN104461384A (en) * | 2014-11-28 | 2015-03-25 | 华为技术有限公司 | Data write-in method and storage device |
Non-Patent Citations (3)
| Title |
|---|
| JIN WANG ET AL: "pLSM: A Highly Efficient LSM-Tree Index Supporting Real-Time Big Data Analysis", 《2013 IEEE 37TH ANNUAL COMPUTER SOFTWARE AND APPLICATIONS CONFERENCE》 * |
| PENG WANG ET AL: "An efficient design and implementation of LSM-tree based key-value store on open-channel SSD", 《PROCEEDINGS OF THE NINTH EUROPEAN CONFERENCE ON COMPUTER SYSTEMS》 * |
| 罗军等: "高效Key-Value持久化缓存***的实现", 《计算机工程》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106227677B (en) * | 2016-07-20 | 2018-11-20 | 浪潮电子信息产业股份有限公司 | Method for managing variable-length cache metadata |
| CN106227677A (en) * | 2016-07-20 | 2016-12-14 | 浪潮电子信息产业股份有限公司 | Method for managing variable-length cache metadata |
| CN106708442A (en) * | 2016-12-30 | 2017-05-24 | 武汉安嘉颐科技有限公司 | Massive data storage method simultaneously applicable to disk and solid state disk reading and writing features |
| CN106708442B (en) * | 2016-12-30 | 2020-02-14 | 硬石科技(武汉)有限公司 | Mass data storage method simultaneously adapting to read-write characteristics of magnetic disk and solid state disk |
| CN108319625B (en) * | 2017-01-17 | 2019-10-25 | 广州市动景计算机科技有限公司 | File mergences method and apparatus |
| CN108319625A (en) * | 2017-01-17 | 2018-07-24 | 广州市动景计算机科技有限公司 | Piece file mergence method and apparatus |
| CN108319602A (en) * | 2017-01-17 | 2018-07-24 | 广州市动景计算机科技有限公司 | Data base management method and Database Systems |
| WO2018133762A1 (en) * | 2017-01-17 | 2018-07-26 | 广州市动景计算机科技有限公司 | File merging method and apparatus |
| CN108319602B (en) * | 2017-01-17 | 2020-10-16 | 阿里巴巴（中国）有限公司 | Database management method and database system |
| CN111936977A (en) * | 2017-12-29 | 2020-11-13 | 华为技术有限公司 | Database management system and method using append only storage device |
| WO2019128820A1 (en) * | 2017-12-29 | 2019-07-04 | Huawei Technologies Co., Ltd. | Systems and methods for database management using append-only storage devices |
| US11921684B2 (en) | 2017-12-29 | 2024-03-05 | Huawei Cloud Computing Technologies Co., Ltd. | Systems and methods for database management using append-only storage devices |
| US11461296B2 (en) | 2017-12-29 | 2022-10-04 | Huawei Cloud Computing Technologies Co., Ltd. | Systems and methods for database management using append-only storage devices |
| US10725983B2 (en) | 2017-12-29 | 2020-07-28 | Huawei Technologies Co., Ltd. | Systems and methods for database management using append-only storage devices |
| CN110851434B (en) * | 2018-07-27 | 2023-07-18 | 阿里巴巴集团控股有限公司 | Data storage method, device and equipment |
| CN110851434A (en) * | 2018-07-27 | 2020-02-28 | 阿里巴巴集团控股有限公司 | Data storage method, device and equipment |
| CN109213445A (en) * | 2018-08-23 | 2019-01-15 | 郑州云海信息技术有限公司 | A kind of management method, management system and the relevant apparatus of storage system metadata |
| CN109407985A (en) * | 2018-10-15 | 2019-03-01 | 郑州云海信息技术有限公司 | A kind of method and relevant apparatus of data management |
| CN109271570A (en) * | 2018-10-30 | 2019-01-25 | 郑州云海信息技术有限公司 | A kind of method of metadata management inquiry |
| CN110502457B (en) * | 2019-08-23 | 2022-02-18 | 北京浪潮数据技术有限公司 | Metadata storage method and device |
| CN110502457A (en) * | 2019-08-23 | 2019-11-26 | 北京浪潮数据技术有限公司 | A kind of metadata storing method and device |
| CN111104403B (en) * | 2019-11-30 | 2022-06-07 | 北京浪潮数据技术有限公司 | LSM tree data processing method, system, equipment and computer medium |
| CN111104403A (en) * | 2019-11-30 | 2020-05-05 | 北京浪潮数据技术有限公司 | LSM tree data processing method, system, equipment and computer medium |
| CN111831622A (en) * | 2020-03-31 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Data index generation method and device, electronic equipment and readable storage medium |
| CN112487095A (en) * | 2020-12-09 | 2021-03-12 | 浪潮云信息技术股份公司 | Method for optimizing transaction data storage of distributed database |
| US11537582B2 (en) | 2021-04-16 | 2022-12-27 | Samsung Electronics Co., Ltd. | Data access method, a data access control device, and a data access system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105117415B (en) | 2018-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105117415A (en) | Optimized SSD data updating method | |
| CN110825748B (en) | High-performance and easily-expandable key value storage method by utilizing differentiated indexing mechanism | |
| CN110347852B (en) | File system embedded with transverse expansion key value storage system and file management method | |
| CN105574104B (en) | A kind of LogStructure storage system and its method for writing data based on ObjectStore | |
| US8732139B2 (en) | Method and system for dynamically partitioning very large database indices on write-once tables | |
| CN110188108B (en) | Data storage method, device, system, computer equipment and storage medium | |
| Ahn et al. | ForestDB: A fast key-value storage system for variable-length string keys | |
| CN109683811B (en) | Request processing method for hybrid memory key value pair storage system | |
| CN109299113B (en) | Range query method with storage-aware mixed index | |
| KR20190019805A (en) | Method and device for storing data object, and computer readable storage medium having a computer program using the same | |
| CN104484471B (en) | A kind of implementation method of high-performance data storage engines | |
| US8086641B1 (en) | Integrated search engine devices that utilize SPM-linked bit maps to reduce handle memory duplication and methods of operating same | |
| US10289709B2 (en) | Interleaved storage of dictionary blocks in a page chain | |
| WO2018205151A1 (en) | Data updating method and storage device | |
| WO2015152830A1 (en) | Method of maintaining data consistency | |
| Lv et al. | Log-compact R-tree: an efficient spatial index for SSD | |
| CN111159176A (en) | Method and system for storing and reading mass stream data | |
| CN101236564A (en) | Mass data high performance reading display process | |
| US7987205B1 (en) | Integrated search engine devices having pipelined node maintenance sub-engines therein that support database flush operations | |
| CN104156432A (en) | File access method | |
| CN112732725B (en) | NVM (non volatile memory) hybrid memory-based adaptive prefix tree construction method, system and medium | |
| JP7345482B2 (en) | Maintaining shards in KV store with dynamic key range | |
| Petrov | Algorithms behind modern storage systems: Different uses for read-optimized b-trees and write-optimized lsm-trees | |
| US7953721B1 (en) | Integrated search engine devices that support database key dumping and methods of operating same | |
| CN110515897B (en) | Method and system for optimizing reading performance of LSM storage system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |