CN104156380B - A kind of distributed memory hash indexing method and system - Google Patents
A kind of distributed memory hash indexing method and system Download PDFInfo
- Publication number
- CN104156380B CN104156380B CN201410077069.4A CN201410077069A CN104156380B CN 104156380 B CN104156380 B CN 104156380B CN 201410077069 A CN201410077069 A CN 201410077069A CN 104156380 B CN104156380 B CN 104156380B
- Authority
- CN
- China
- Prior art keywords
- key
- node
- memory node
- memory
- steps
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 14
- 238000013507 mapping Methods 0.000 claims abstract description 21
- 238000004321 preservation Methods 0.000 claims description 2
- 230000010365 information processing Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 230000002688 persistence Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9014—Indexing; Data structures therefor; Storage structures hash tables
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
Abstract
The invention belongs to database, distributed information processing field more particularly to a kind of distributed memory hash indexing method and systems.In the present invention, the address range of Hash table is mapped in multiple memory nodes, the cryptographic Hash of host node calculation key key, memory node number is obtained according to the cryptographic Hash for the key key being calculated;Then it is numbered and the mapping relationship searching of memory node IP to corresponding memory node address according to memory node, memory node is corresponded to memory node address to be written, read or delete operation, due to having expanded memory capacity by multiple memory nodes, simultaneously by searching based on key assignments (key-value) distributed hash, the readwrite performance of Database Systems is improved.
Description
Technical field
The invention belongs to database, distributed information processing field more particularly to a kind of distributed memory Hash
Indexing means and system.
Background technique
The data that needs in today of rapid technological growth, social every field count all are skyrocketed through, magnanimity number
According to record, look for computer technology and bring new developing direction, new technical task is provided.Hash of the tradition based on disk
Table, magnetic disc i/o will take a substantial amount of time each time, when being difficult to meet modern transaction system (such as stock) to real-time respond
Between requirement.Further for mass data, single machine node is interior since all indexes can not be stored in by the limitation of memory capacity
In depositing.
Summary of the invention
The purpose of the present invention is to provide a kind of distributed memory Hash based on more memory nodes and key-value
Indexing means, it is intended to solve existing disk Hash table I/O time-consuming and single machine node due to memory capacity limitation nothing
All indexes are stored in the problems in memory by method.
The invention is realized in this way a kind of distributed memory hash indexing method, comprising:
The address range of Hash table is mapped in multiple memory nodes;
The cryptographic Hash of host node calculation key key obtains memory node according to the cryptographic Hash for the key key being calculated and compiles
Number;
According to memory node number and the mapping relationship searching of memory node IP to corresponding memory node address,
The memory node number and the mapping relations of memory node IP store in the routing table, and the routing table is stored in main section
Point in;
Memory node is corresponded to the memory node address to be written, read or delete operation.
Further, the cryptographic Hash for the key assignments that the basis is calculated obtains node serial number, specifically:
It is used as node serial number by preceding m of the cryptographic Hash for the key being calculated, the m is the number of memory node.
Further, the method also includes:
Disk node saves the mirror image of the Hash table in memory node by the Hash table based on disk, to disk node
Carry out write-in or delete operation.
The present invention also provides a kind of distributed memory hash index systems, comprising: mapping block, computing module,
One obtains module, the first searching module, writing module, read module, removing module;
The mapping block, for the address range of Hash table to be mapped in multiple memory nodes;
The computing module, the cryptographic Hash for host node calculation key key;
Described first obtains module, obtains memory node number for the cryptographic Hash according to the key key being calculated;
First searching module, for being arrived according to memory node number and the mapping relationship searching of memory node IP
Corresponding memory node address, the memory node number and the mapping relations of memory node IP are stored in routing table
In, the routing table storage is in the master node;
The write module carries out write operation for corresponding to memory node to the memory node address;
The read module is read for corresponding to memory node to the memory node address;
The removing module carries out delete operation for corresponding to memory node to the memory node address.
Further, described first module is obtained further include:
First acquisition submodule, for being used as node serial number for preceding m of the cryptographic Hash for the key being calculated, the m is
The number of memory node.
Further, the system also includes memory modules, pass through the Hash table based on disk for disk node and save
The mirror image of Hash table in memory node;
Further, the write module is also used to carry out write operation to disk node;The removing module is also used to
Delete operation is carried out to disk node.
In the present invention, the address range of Hash table is mapped in multiple memory nodes, host node calculation key key's
Cryptographic Hash obtains memory node number according to the cryptographic Hash for the key key being calculated;Then according to memory node number with
The mapping relationship searching of memory node IP corresponds to memory section to corresponding memory node address, to memory node address
Click through row write enter, reading or delete operation, due to having expanded memory capacity by multiple memory nodes, while passing through base
It is searched in key assignments (key-value) distributed hash, improves the readwrite performance of Database Systems.
Detailed description of the invention
Fig. 1 is the implementation flow chart of distributed memory hash indexing method provided in an embodiment of the present invention;
Fig. 2 is provided in an embodiment of the present invention to node address corresponding node progress write operation specific implementation flow chart;
Fig. 3 is provided in an embodiment of the present invention to be read specific implementation flow chart to node address corresponding node;
Fig. 4 is provided in an embodiment of the present invention to node address corresponding node progress delete operation specific implementation flow chart;
Fig. 5 is the structural block diagram of distributed memory hash index system provided in an embodiment of the present invention;
Fig. 6 is the structural block diagram of the writing module 55 in Fig. 5 provided in an embodiment of the present invention;
Fig. 7 is the structural block diagram of the read module 56 in Fig. 5 provided in an embodiment of the present invention;
Fig. 8 is the structural block diagram of the removing module 57 in Fig. 5 provided in an embodiment of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Fig. 1 shows a kind of implementation process of distributed memory hash indexing method provided in an embodiment of the present invention, in detail
It states as follows:
The address range of Hash table is mapped in multiple memory nodes by step 101.
The address range of Hash table is mapped in multiple memory nodes first, each memory node storage is corresponding
Hash table namely a part for saving concordance list.The cluster of all memory node compositions is also reservoir index point.
The cryptographic Hash of step 102, host node calculation key key obtains memory according to the cryptographic Hash for the key key being calculated
Node serial number.
Which memory node one key (key), which should be mapped to, host node decision, the Hash of host node calculation key key
Value obtains memory node number according to the cryptographic Hash for the key key being calculated.In embodiments of the present invention, according to calculating
The cryptographic Hash of the key assignments arrived obtains node serial number, specifically: preceding m of the cryptographic Hash for the key being calculated are compiled as node
Number, wherein m is the number of memory node.
Step 103, according to memory node number and the mapping relationship searching of memory node IP to corresponding memory
Node address.
After host node obtains memory node number, closed according to memory node number and the mapping of memory node IP
System finds corresponding memory node address.Memory node is numbered to be previously stored in the mapping relations of memory node IP
In routing table, and routing table storage is in the master node.
Step 104 corresponds to memory node to above-mentioned memory node address and is written, is read or delete operation.
After host node finds the address of corresponding memory node, to memory node address correspond to memory node into
Row write enters (put), reads (get) or deletes (delete) operation.
The embodiment of the present invention, due to having expanded memory capacity by multiple memory nodes, while by being based on key assignments
(key-value) distributed hash is searched, and improves the readwrite performance of Database Systems.
Further, in order to index holding persistence, the above method further include: disk node passes through the Kazakhstan based on disk
Uncommon table saves the mirror image of the Hash table in memory node, carries out write-in or delete operation to disk node.Specifically, reading behaviour
Make to carry out just for memory node, disk node can't be searched, and memory section can be directed to delete operation simultaneously by being written
Point operated with disk node, still, for memory node operation be directed to disk node operation be it is asynchronous into
Capable, the operation of memory node does not need to wait the completion of disk nodal operation, therefore, the operation of disk node not shadow
Acoustic system overall performance.The purpose that disk node is added is to have no effect on the place of system to carry out persistence preservation to index
Manage speed.
Further, in order to carry out write operation to node address corresponding node in more detailed description step 104,
Combination of embodiment of the present invention attached drawing 2 do the following specifically describes:
Step 201 is searched the key in the bloom filter of host node and be whether there is, if it is, executing step
Rapid 202, if it is not, then being performed simultaneously step 203 and 204.
Bloom filter is stored in the master node, and the effect of bloom filter is can to lead to too small amount of memory
Quickly to judge whether given key is not present.However, can not agree at this time if bloom filter determines that given key exists
Determine key to be implicitly present in, needs to further search for memory node.
Step 202 searches key whether in current memory node, has existed if so, returning to key, if not
Then follow the steps 203.
Step 203, according to memory node address, key and corresponding value value are written in memory node, it
End operation afterwards.
If memory node has above-mentioned key, key and corresponding value value are written in memory node.
Key and corresponding value are sent disk node by step 204, then executes step 205.
Step 205 judges whether disk node log buffer area log cache has expired, if so, after thening follow the steps 206
Step 207 is executed, if it is not, then directly executing step 207.
The content of current log cache is written in disk by step 206, empties log cache.
Key and corresponding value is written in log cache step 207, later end operation.
This completes the write operations to memory node and disk node.Based on looking into for key-value mode
It looks for and write operation, effectively increases write efficiency.
Further, for being read to node address corresponding node in more detailed description step 104,
Combination of embodiment of the present invention attached drawing 3 do the following specifically describes:
Step 301 is searched the key in the bloom filter of host node and be whether there is, if it is, executing step
Rapid 302, if not, thening follow the steps 304.
Equally, store bloom filter in the master node, the effect of bloom filter be can lead to it is too small amount of
Memory quickly to judge whether given key is not present.However, if bloom filter determines that given key exists, at this time simultaneously
It cannot affirm that key is implicitly present in, need to further search for memory node.
Step 302 searches key whether in current memory node according to memory node IP, if it is, executing
Step 303, if not, thening follow the steps 304.
Step 303 returns to key and corresponding value value.
When memory node search key in current memory node, then return to key and corresponding value value.
Step 304, return key are not present.
This completes the read operations to memory node.Equally, lookup and reading based on key-value mode
Operation, effectively increases reading efficiency.
Further, for being read to node address corresponding node in more detailed description step 104,
Combination of embodiment of the present invention attached drawing 4 do the following specifically describes:
Step 401 is searched the key in the bloom filter of host node and be whether there is, if it is, executing step
Rapid 402, if not, thening follow the steps 405.
Equally, store bloom filter in the master node, the effect of bloom filter be can lead to it is too small amount of
Memory quickly to judge whether given key is not present.However, if bloom filter determines that given key exists, at this time simultaneously
It cannot affirm that key is implicitly present in, need to further search for memory node.
Step 402 searches key whether in current memory node according to memory node IP, if it is, simultaneously
Step 403 and step 404 are executed, if not, thening follow the steps 405.
Key and corresponding value value in step 403, deletion memory node.
Key and corresponding value value in step 404, deletion disk node.
Step 405, return key are not present.
The embodiment of the present invention passes through the bloom filter of host node and lookup and reading based on key-value mode
Extract operation effectively increases delete operation efficiency.
It should be noted that for the various method embodiments described above, for simple description, therefore, it is stated as a series of
Combination of actions, but those skilled in the art should understand that, the present invention is not limited by the sequence of acts described because
According to the present invention, some steps may be performed in other sequences or simultaneously.
Fig. 5 is the embodiment of the invention provides a kind of distributed memory hash index system, which is able to achieve aforementioned
Method described in embodiment.For ease of description, only the parts related to this embodiment are shown.As shown in figure 5, the system
It include: mapping block 51, computing module 52, first acquisition module 53, the first searching module 54, writing module 55, read module
56, removing module 57.
Mapping block 51, for the address range of Hash table to be mapped in multiple memory nodes;
Computing module 52, the cryptographic Hash for host node calculation key key;
First obtains module 53, obtains memory node number for the cryptographic Hash according to the key key being calculated.Into one
Step, first obtains module 53 further include: the first acquisition submodule 531, for by the preceding m of the cryptographic Hash for the key being calculated
Position is used as node serial number, and the m is the number of memory node.
First searching module 54, for according to memory node number and the mapping relationship searching of memory node IP to pair
The memory node address answered, the memory node number and the mapping relations of memory node IP store in the routing table,
The routing table storage is in the master node.
Writing module 55 carries out write operation for corresponding to memory node to memory node address.
Read module 56 is read for corresponding to memory node to memory node address.
Removing module 57 carries out delete operation for corresponding to memory node to memory node address.
Further, above system further include: memory module 58 is protected for disk node by the Hash table based on disk
Deposit the mirror image of the Hash table in memory node.
Further, writing module 55 is also used to carry out write operation to disk node;Removing module 57 is also used to magnetic
Disk node carries out delete operation.
In embodiments of the present invention, due to having expanded memory capacity by multiple memory nodes, while by being based on
Key assignments (key-value) distributed hash is searched, and the readwrite performance of Database Systems is improved.
Fig. 6 is the structural block diagram of writing module 55 in the Fig. 5 provided in the embodiment of the present invention, and writing module 55 includes second
Submodule 553, the first sending module 554, first judgment module is written in searching module 551, third searching module 552, first
555, submodule 557, the first return module 558 is written in the second write-in submodule 556, third.
Second searching module 551, the key for searching described in the bloom filter of host node whether there is.
Third searching module 552, for searching whether key is working as in the presence of the second searching module 551 finds key
In preceding memory node.
First write-in submodule 553, for not found in current memory node in third searching module 552
When key, according to memory node IP address, key and corresponding value value are written in memory node.
First sending module 554, for not finding key in current memory node in third searching module 552
When, disk node is sent by key and corresponding value.
First judgment module 555, for judging whether disk node log buffer area log cache has expired.
Second write-in submodule 556, for judging disk node log buffer area log in first judgment module 555
When cache has expired, the content of current log cache is written in disk, and empty log cache.
Submodule 557 is written in third, for key and corresponding value to be written in log cache.
First return module 558, for when third searching module 552 finds key in current memory node,
Return to key and presence.
Fig. 7 is the structural block diagram of read module 56 in the Fig. 5 provided in the embodiment of the present invention.Read module 56, including the
Four searching modules 561, the 5th searching module 562, the second return module 563.
4th searching module 561, the key for searching described in the bloom filter of host node whether there is.
5th searching module 562, for searching whether key is working as in the presence of the 4th searching module 561 finds key
In preceding memory node.
Second return module 563, for returning when the 5th searching module 562 searches key in current memory node
Return key and corresponding value value.Second return module 563 be also used to the 4th searching module 561 find key be not present or
When the 5th searching module 562 of person searches key not in current memory node, returns to key and be not present.
Fig. 8 is the structural block diagram of removing module 57 in the Fig. 5 provided in the embodiment of the present invention.Removing module 57 includes:
Six searching modules 571, the 7th searching module 572, first delete submodule 573, second and delete submodule 574 and third return
Module 575.
6th searching module 571, the key for searching described in the bloom filter of host node whether there is.
7th searching module 572, the key of the bloom filter for finding host node in the 6th searching module 571
In the presence of, key is searched whether in current memory node according to memory node IP.
First deletes submodule 573, when searching key in current memory node for the 7th searching module 572, deletes
Except the key and corresponding value value in memory node.
Second deletes submodule 574, when searching key in current memory node for the 7th searching module 572, deletes
Except the key and corresponding value value in disk node.
Third return module 575 finds key for the 6th searching module 571 and is not present or the 7th searching module 572
When searching key not in current memory node, returns to key and be not present.
The embodiment of the present invention, by the bloom filter of host node and the lookup based on key-value mode with write
Enter, read, delete operation, effectively increasing operating efficiency.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (2)
1. a kind of distributed memory hash indexing method characterized by comprising
The address range of Hash table is mapped in multiple memory nodes;
The cryptographic Hash of host node calculation key key is used as node serial number, the m for preceding m of the cryptographic Hash for the key being calculated
For the number of memory node;
It is numbered with the mapping relationship searching of memory node IP according to memory node to corresponding memory node address, it is described
Memory node number and the mapping relations of memory node IP store in the routing table, and the routing table is stored in host node
In;
Memory node is corresponded to the memory node address to be written, read or delete operation;
Disk node saves the mirror image of the Hash table in memory node by the Hash table based on disk, carries out to disk node
Write-in or delete operation;
It is described that memory node progress write operation is corresponded to the memory node address, specific following steps:
Step A1, the key is searched in the bloom filter of host node whether there is, if so, A2 is thened follow the steps,
If it is not, then being performed simultaneously step A3 and A4;
Step A2, key is searched whether in current memory node, if so, A5 is thened follow the steps, if otherwise executing step
Rapid A3;
Step A3, according to memory node IP, key and corresponding value value is written in memory node, terminate to grasp later
Make;
Step A5, disk node is sent by key and corresponding value, then executes step A6;
Step A6, judge whether disk node log buffer area log cache has expired, if so, executing step after thening follow the steps A7
Rapid A8, if it is not, then directly executing step A8;
Step A7, the content of current log cache is written in disk, empties log cache;
Step A8, key and corresponding value is written in log cache, later end operation;
It is described memory node is corresponded to the memory node address to be read, specific following steps:
Step B1, the key is searched in the bloom filter of host node whether there is, if so, B2 is thened follow the steps,
If not, thening follow the steps B4;
Step B2, key is searched whether in current memory node according to memory node IP, if so, thening follow the steps
B3, if not, thening follow the steps B4;
Step B3, key and corresponding value value is returned;
Step B4, key is returned to be not present;
It is described that memory node progress delete operation is corresponded to the memory node address, specific following steps:
Step C1, the key is searched in the bloom filter of host node whether there is, if so, C2 is thened follow the steps,
If not, thening follow the steps C5;
Step C2, key is searched whether in current memory node according to memory node IP, if it is, being performed simultaneously
Step C3 and step C4, if not, thening follow the steps C5;
Step C3, the key and corresponding value value in memory node are deleted;
Step C4, the key and corresponding value value in disk node are deleted;
Step C5, key is returned to be not present.
2. a kind of distributed memory hash index system characterized by comprising mapping block, computing module, first are searched
Module, writing module, read module, removing module, memory module;
The mapping block, for the address range of Hash table to be mapped in multiple memory nodes;
The computing module, for the cryptographic Hash of host node calculation key key, by preceding m works of the cryptographic Hash for the key being calculated
For node serial number, the m is the number of memory node;
First searching module, for according to memory node number with the mapping relationship searching of memory node IP to corresponding
Memory node address, the mapping relations storage of memory node number and memory node IP in the routing table, institute
State routing table storage in the master node;
The write module carries out write operation for corresponding to memory node to the memory node address;
The read module is read for corresponding to memory node to the memory node address;
The removing module carries out delete operation for corresponding to memory node to the memory node address;
The memory module passes through the mirror of the Hash table in the Hash table preservation memory node based on disk for disk node
Picture;
The write module is also used to carry out write operation to disk node;The removing module is also used to carry out disk node
Delete operation;
The write module is specifically used for executing following steps:
Step A1, the key is searched in the bloom filter of host node whether there is, if so, A2 is thened follow the steps,
If it is not, then being performed simultaneously step A3 and A4;
Step A2, key is searched whether in current memory node, if so, A5 is thened follow the steps, if otherwise executing step
Rapid A3;
Step A3, according to memory node IP, key and corresponding value value is written in memory node, terminate to grasp later
Make;
Step A5, disk node is sent by key and corresponding value, then executes step A6;
Step A6, judge whether disk node log buffer area log cache has expired, if so, executing step after thening follow the steps A7
Rapid A8, if it is not, then directly executing step A8;
Step A7, the content of current log cache is written in disk, empties log cache;
Step A8, key and corresponding value is written in log cache, later end operation;
The read module is specifically used for executing following steps:
Step B1, the key is searched in the bloom filter of host node whether there is, if so, B2 is thened follow the steps,
If not, thening follow the steps B4;
Step B2, key is searched whether in current memory node according to memory node IP, if so, thening follow the steps
B3, if not, thening follow the steps B4;
Step B3, key and corresponding value value is returned;
Step B4, key is returned to be not present;
The removing module is specifically used for executing following steps:
Step C1, the key is searched in the bloom filter of host node whether there is, if so, C2 is thened follow the steps,
If not, thening follow the steps C5;
Step C2, key is searched whether in current memory node according to memory node IP, if it is, being performed simultaneously
Step C3 and step C4, if not, thening follow the steps C5;
Step C3, the key and corresponding value value in memory node are deleted;
Step C4, the key and corresponding value value in disk node are deleted;
Step C5, key is returned to be not present.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410077069.4A CN104156380B (en) | 2014-03-04 | 2014-03-04 | A kind of distributed memory hash indexing method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410077069.4A CN104156380B (en) | 2014-03-04 | 2014-03-04 | A kind of distributed memory hash indexing method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104156380A CN104156380A (en) | 2014-11-19 |
CN104156380B true CN104156380B (en) | 2019-03-26 |
Family
ID=51881885
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410077069.4A Expired - Fee Related CN104156380B (en) | 2014-03-04 | 2014-03-04 | A kind of distributed memory hash indexing method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104156380B (en) |
Families Citing this family (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408165B (en) * | 2014-12-08 | 2018-04-13 | 畅捷通信息技术股份有限公司 | Date storage method and device under high concurrent |
US10740290B2 (en) | 2015-04-14 | 2020-08-11 | Jetflow Technologies | Systems and methods for key-value stores |
CN105159985A (en) * | 2015-08-31 | 2015-12-16 | 努比亚技术有限公司 | Data query device and method based on redis cluster |
CN105426451A (en) * | 2015-11-11 | 2016-03-23 | 深圳市华讯方舟科技有限公司 | Key value pair-based data processing method and system |
CN105930510A (en) * | 2016-05-12 | 2016-09-07 | 广西尊达电子商务有限公司 | Multi-database management system based on base table hash |
CN106339450B (en) * | 2016-08-25 | 2019-08-06 | 成都索贝数码科技股份有限公司 | Index method of tree-shaped data |
CN106527982B (en) * | 2016-10-25 | 2019-04-12 | 西安交通大学 | A kind of object distribution algorithm for the object storage system being made of heterogeneous storage devices |
CN106845926B (en) * | 2016-12-27 | 2020-11-27 | 中国建设银行股份有限公司 | Distributed data stream processing method and system for third-party payment supervision system |
CN107562385B (en) * | 2017-09-13 | 2020-08-04 | 郑州云海信息技术有限公司 | Method, device and equipment for reading data by distributed storage client |
CN109947669A (en) * | 2017-12-21 | 2019-06-28 | 北京忆恒创源科技有限公司 | The address conversion method and system of KV storage equipment |
CN110134678A (en) * | 2018-02-08 | 2019-08-16 | 深圳先进技术研究院 | A kind of indexing means of biological data, system and electronic equipment |
US11061609B2 (en) * | 2018-08-02 | 2021-07-13 | MemVerge, Inc | Distributed memory object method and system enabling memory-speed data access in a distributed environment |
CN109376136A (en) * | 2018-10-19 | 2019-02-22 | 郑州云海信息技术有限公司 | A kind of distributed information log processing system, the network equipment and method |
CN110245129B (en) * | 2019-04-23 | 2022-05-13 | 平安科技(深圳)有限公司 | Distributed global data deduplication method and device |
CN110347886B (en) * | 2019-06-21 | 2022-03-25 | 政采云有限公司 | Distributed parent-child relationship element searching method |
CN110781155B (en) * | 2019-10-18 | 2022-06-24 | 赛尔网络有限公司 | Data storage reading method, system, equipment and medium based on IPFS |
CN111245943A (en) * | 2020-01-14 | 2020-06-05 | 山东汇贸电子口岸有限公司 | Method for obtaining accessible IP address by Redis cluster |
CN111258508B (en) * | 2020-02-16 | 2020-11-10 | 西安奥卡云数据科技有限公司 | Metadata management method in distributed object storage |
CN112511450B (en) * | 2020-11-02 | 2022-05-31 | 杭州迪普信息技术有限公司 | Flow control equipment and method |
CN112445771A (en) * | 2020-12-10 | 2021-03-05 | 阿米华晟数据科技(江苏)有限公司 | Data processing method, device and equipment of network flow and storage medium |
CN112637327B (en) * | 2020-12-21 | 2022-07-22 | 北京奇艺世纪科技有限公司 | Data processing method, device and system |
CN117112549B (en) * | 2023-10-20 | 2024-03-26 | 中科星图测控技术股份有限公司 | Big data merging method based on bloom filter |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102004769A (en) * | 2010-11-12 | 2011-04-06 | 成都市华为赛门铁克科技有限公司 | File management method, equipment and memory system |
CN102591970A (en) * | 2011-12-31 | 2012-07-18 | 北京奇虎科技有限公司 | Distributed key-value query method and query engine system |
CN102929793A (en) * | 2011-08-08 | 2013-02-13 | 株式会社东芝 | Memory system including key-value store |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10558705B2 (en) * | 2010-10-20 | 2020-02-11 | Microsoft Technology Licensing, Llc | Low RAM space, high-throughput persistent key-value store using secondary memory |
US8886827B2 (en) * | 2012-02-13 | 2014-11-11 | Juniper Networks, Inc. | Flow cache mechanism for performing packet flow lookups in a network device |
-
2014
- 2014-03-04 CN CN201410077069.4A patent/CN104156380B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102004769A (en) * | 2010-11-12 | 2011-04-06 | 成都市华为赛门铁克科技有限公司 | File management method, equipment and memory system |
CN102929793A (en) * | 2011-08-08 | 2013-02-13 | 株式会社东芝 | Memory system including key-value store |
CN102591970A (en) * | 2011-12-31 | 2012-07-18 | 北京奇虎科技有限公司 | Distributed key-value query method and query engine system |
Non-Patent Citations (1)
Title |
---|
"一种支持高维数据查询的并行索引机制";王寅峰 等;《华中科技大学学报(自然科学版)》;20110630;第39卷;第156-160页 |
Also Published As
Publication number | Publication date |
---|---|
CN104156380A (en) | 2014-11-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104156380B (en) | A kind of distributed memory hash indexing method and system | |
CN105630955B (en) | A kind of data acquisition system member management method of high-efficiency dynamic | |
CN105320775B (en) | The access method and device of data | |
CN104301360B (en) | A kind of method of logdata record, log server and system | |
JP5996088B2 (en) | Cryptographic hash database | |
CN103902623B (en) | Method and system for the accessing file in storage system | |
CN108319654A (en) | Computing system, cold and hot data separation method and device, computer readable storage medium | |
CN110263014A (en) | Block chain storage system and method towards timing type data | |
CN102314485A (en) | Method and device for adding, searching and deleting hash table | |
CN111552692B (en) | Plus-minus cuckoo filter | |
CN103473298B (en) | Data archiving method and device and storage system | |
CN108287840A (en) | A kind of data storage and query method based on matrix Hash | |
WO2013075306A1 (en) | Data access method and device | |
CN109407985B (en) | Data management method and related device | |
CN110532307A (en) | A kind of date storage method and querying method flowing sliding window | |
CN109189759A (en) | Method for reading data, data query method, device and equipment in KV storage system | |
WO2016119597A1 (en) | Page querying method and data processing node in oltp cluster database | |
CN102542036B (en) | Use the table optimized | |
CN106844491B (en) | Temporary data writing and reading method and device | |
CN103942301A (en) | Distributed file system oriented to access and application of multiple data types | |
CN113867627B (en) | Storage system performance optimization method and system | |
CN107766355B (en) | Hierarchical data management method, hierarchical data management system and instant messaging system | |
CN102650972B (en) | Date storage method, Apparatus and system | |
CN104391992A (en) | Asset data-oriented data processing system | |
WO2024041376A1 (en) | Distributed graph data processing system, method, apparatus and device, and storage medium |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190326 |
|
CF01 | Termination of patent right due to non-payment of annual fee |