CN112395213A - Memory hot-spot data oriented ACEH index structure and method - Google Patents

Memory hot-spot data oriented ACEH index structure and method Download PDF

Info

Publication number
CN112395213A
CN112395213A CN202011296272.2A CN202011296272A CN112395213A CN 112395213 A CN112395213 A CN 112395213A CN 202011296272 A CN202011296272 A CN 202011296272A CN 112395213 A CN112395213 A CN 112395213A
Authority
CN
China
Prior art keywords
data
bucket
key
data bucket
segment
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
Application number
CN202011296272.2A
Other languages
Chinese (zh)
Other versions
CN112395213B (en
Inventor
何水兵
朱彤
曾令仿
段雪豪
银燕龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Zhejiang Lab
Original Assignee
Zhejiang University ZJU
Zhejiang Lab
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU, Zhejiang Lab filed Critical Zhejiang University ZJU
Priority to CN202011296272.2A priority Critical patent/CN112395213B/en
Publication of CN112395213A publication Critical patent/CN112395213A/en
Application granted granted Critical
Publication of CN112395213B publication Critical patent/CN112395213B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
    • G06F12/023—Free address space management
    • G06F12/0238—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
    • G06F12/0246—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory in block erasable memory, e.g. flash memory
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses an ACEH index structure and a method for hot data based on memory, wherein the structure comprises the following steps: directory entries, segments, and buckets; the method comprises the following steps: the directory entries are indexed by a global depth G, a segment corresponds to a set of buckets, the segments are indexed by a local depth L, and L is G-log2 kAnd k represents the number of pointers pointing to the data bucket, the data bucket index adopts an Adjusted-Cuckoo algorithm to position a data bucket inserted by a hash key, the Adjusted-Cuckoo algorithm comprises two hash functions, two insertable data buckets are generated, then an empty data bucket is selected for insertion, the Adjusted-Cuckoo algorithm determines one data bucket, and the second data bucket is directly determined as the next data bucket of the current data bucket, the operation method comprises the following steps: step one, inserting operation; step two, refreshing operation; step three, splitting operation; and step four, deleting operation.

Description

Memory hot-spot data oriented ACEH index structure and method
Technical Field
The invention relates to a computer memory storage structure, in particular to an ACEH index structure and method based on memory hot spot data.
Background
Internet companies have widely used key-value storage technology in actual production environments to improve the performance of data storage. Scholars studied the hot problem in different scenarios and in some scenarios presented an effective solution. However, the hot spot problem in the key-value storage scenario is ignored.
In a conventional extensible hash structure, when a key-value pair (key-value) is inserted, the key is first matched with a directory entry (directory). For example, the key matches the first directory entry "00" of the directory, and is inserted into the Bucket (Bucket) pointed to by the first directory entry. When entering a data bucket, the traditional extensible hash directly adopts a sequential traversal mode until a first empty slot point (slot) is found, and a key value pair is inserted. During searching, the directory entry is found according to the corresponding position of the key, the data bucket is positioned through the pointer of the directory entry, and finally the corresponding key value pair is searched in a sequential traversal mode. The deletion process is the same as the lookup. When refreshing, the traditional scalable hash just does an insert operation again, which wastes space greatly.
When expanded, directory entries are multiplied. When the directory is modified, the pointer to each directory entry is modified, and the partial key-value pairs in the data bucket are moved accordingly. For example, when the directory number is 4, the key-value pair is inserted into the data bucket of directory entry 00, and after expansion, the key-value pair will move to the data bucket corresponding to directory entry 001 (because the first three bits of the key-value pair are 001).
Disclosure of Invention
In order to solve the defects of the prior art and achieve the purposes of increasing the utilization rate of the memory and accelerating the search performance, the invention adopts the following technical scheme:
an ACEH index structure based on memory hotspot-oriented data, comprising: directory entries and buckets, using a middle nodeThe structure is called a segment, a global depth G is used as a segment index between a directory entry and the segment for solving the problem of inserting and searching in a three-layer structure, one segment comprises a group of data buckets, a local depth L is used as the data bucket index between the segment and the data bucket, and L is G-log2 kk denotes the number of pointers to the bucket, L is used because it is not necessary that one Directory entry (Directory) corresponds to one segment after the extensible hash expansion, since the number of directory entries after expansion is greater than the number of segments in the original, new segments will only be created if key-value pairs need to be moved to a non-existing bucket, the data bucket index adopts an Adjusted-Cuckoo algorithm to position a data bucket inserted by a hash key, the Adjusted-Cuckoo algorithm comprises two hash functions to generate two insertable data buckets, then selecting an empty data bucket for insertion, determining one data bucket by the Adjusted-Cuckoo algorithm, directly determining the second data bucket as the next data bucket of the current data bucket, the method is environment-friendly to cacheline, and compared with the traditional cuckoo hash algorithm, the method can utilize the spatial locality to accelerate the search performance and remarkably improve the utilization rate of the memory. This structure provides a significant performance boost for operating on NVM.
A memory hot data oriented ACEH indexing method is characterized in that a directory entry carries out segment indexing through global depth G, one segment corresponds to a group of data buckets, the segment carries out data bucket indexing through local depth L, and L is G-log2 kK represents the number of pointers pointing to the data bucket, L is adopted because one Directory entry (Directory) does not necessarily correspond to one segment after the extensible hash expansion, and because the number of Directory entries after the expansion is more than the number of original segments, a new segment is created only when a key value pair needs to be moved to a nonexistent data bucket, the data bucket index adopts an Adjusted-Cuckoo algorithm to position a data bucket with inserted hash keys, the Adjusted-Cuckoo algorithm comprises two hash functions to generate two insertable data buckets, then an empty data bucket is selected for insertion, the Adjusted-Cuckoo algorithm determines one data bucket, and the second data bucket is directly determined as the next data bucket of the current data bucket, so that the setting is friendly to cacheline, compared with the traditional Cuckoo hash algorithm, the index can be used for utilizing one Directory entry (Directory) to correspond to one segment, and the index is positioned in the data bucket with two insertable data bucketsThe method has the advantages of spatial locality, accelerated search performance, obvious improvement on the utilization rate of the memory, great performance improvement on the operation on the NVM, and the operation method comprises the following steps:
step one, inserting operation;
step two, refreshing operation;
step three, splitting operation;
and step four, deleting operation.
Further, the Adjusted-Cuckoo algorithm selects a multiplex-shift function to determine a data bucket.
Further, in the inserting operation of the first step, a section pointed by a directory entry is found by locating the directory entry at the highest position, then two insertable data buckets are found according to an Adjusted-Cuckoo algorithm, data in the two data buckets are sequentially traversed, and if a key is the same as the inserted key, the value corresponding to the current key is updated; and if the key does not exist and the key is inserted, randomly inserting the hash key if both the two data buckets have free positions, inserting the free data buckets if only one data bucket is free, and if both the two data buckets have no free positions, failing to insert and performing splitting operation.
Furthermore, the refresh operation in the second step adopts in-situ refresh, directly compares the key to be updated with the key stored in each slot in the detection process, and updates the key if the keys are the same. The traditional expandable hash is to directly insert data, so that repeated key value pairs exist, space is wasted, errors in reading data are likely to occur, in-situ refreshing is performed, the repeated key value pairs are removed, and the utilization rate of a memory is improved.
Furthermore, the splitting operation in the third step creates a new segment, and at the same time, a redundant directory entry pointer pointing to the original segment points to the new segment, and the valid key-value pair corresponding to the directory entry pointer is also transferred to the new segment. And deleting the key value pairs in the original segment.
Further, in the deletion operation in the fourth step, after the directory entry is updated, the query for searching the migrated record accesses the new segment, and the query for searching the non-migrated record accesses the old segment, but since the split segments (i.e., the old segments) include all keys, they always successfully find the records for searching key values, which include some unnecessary repeated entries, and the lazy expansion is performed in the expansion process, and when part of key value pairs are migrated in the expansion process, the key value pairs of the migrated original data bucket are not immediately deleted, but when new data are inserted into the original data bucket, the migrated key value pairs are directly replaced, so that the overhead of hash table expansion can be reduced, and the utilization rate of the memory can be improved.
Further, the lazy deletion directly covers the data x2 in the Bucket1 data Bucket over the data x1 when deleting the data x1 in the Bucket0 data Bucket, records the data at the position of x2 as invalid, strives for the data inserted into both the Bucket0 and the Bucket1 to be in the Bucket0, reduces the number of search times, reduces the average length of search data, and improves the data access performance.
The invention has the advantages and beneficial effects that:
the method is characterized in that the method adopts the improved hash index aiming at the hot spot data set based on the extensible hash structure ACEH, and is different from the common extensible hash index, in the secondary indexing process, the ACEH uses a modified cuckoo hash algorithm, so that the position where each datum can be inserted is increased, and the memory utilization rate is increased; the ACEH also provides an in-situ refresh operation, reducing the occupation of memory space by repeated key value pairs. Meanwhile, the operation can also reduce the splitting operation of the ACEH structure and improve the insertion performance.
Drawings
Fig. 1 is a schematic diagram of an ACEH index structure according to the present invention.
FIG. 2 is a schematic diagram of the process of the Adjusted-Cuckoo algorithm in the present invention.
FIG. 3 is a diagram illustrating creation of a new segment in the present invention.
FIG. 4 is a diagram illustrating the expansion of directory entries in the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
An ACEH (Adjusted-Cuckoo extensible Hashing) structure for hotspot data is a hash storage structure for a data set containing hotspots.
First, an ACEH logic structure and algorithm:
as shown in fig. 1, the configuration of the ACEH uses an intermediate structure, called Segment, more than the conventional extensible hash. A Segment consists of N buckets. In order to solve the insertion and search problem in the three-layer structure, G bits (representing the global depth) are used as segment indexes in the structure, and the Adjusted-Cuckoo algorithm is used for positioning which Bucket the hash key is inserted into.
The Adjusted-Cuckoo algorithm: the method is the same as the traditional Cuckoo hash algorithm, the Adjusted-Cuckoo algorithm also comprises two hash functions, so that two pluggable buckets can be generated after a hash key passes through the Adjusted-Cuckoo algorithm, then an empty Bucket is selected for insertion, but the method is different from the Cuckoo hash algorithm in that the Adjusted-Cuckoo algorithm randomly selects a hash-shift function to determine one Bucket, and the second Bucket is directly determined as the next Bucket of the current Bucket.
As shown in fig. 2, it is assumed that x1 can be inserted into Bucket0 and Bucket1 after being calculated by the Adjusted-Cuckoo algorithm, and inserted into Bucket0 after being traversed sequentially, and x2 can be inserted into Bucket0 and Bucket1 after also being calculated by the Adjusted-Cuckoo algorithm, and inserted into Bucket1 through being traversed sequentially, when x1 is deleted, x2 can be directly overlaid on x1, and then the data at the position of x2 is marked as INVALID (INVALID). Therefore, data inserted into both the Bucket0 and the Bucket1 can be strived to be in the Bucket0 as much as possible, so that the number of searching times is reduced, the average length of the searched data is reduced, and the data access performance is improved.
After the Adjusted-Cuckoo algorithm is adopted, the ACEH is obviously improved in the aspect of memory utilization compared with the traditional extensible hash. In the refresh operation, the traditional expandable hash is to insert data directly, so that there is a waste of space due to repeated key-value pairs, and errors in reading data are likely to occur. And the ACEH adopts in-situ refreshing to remove repeated key-value pairs.
II, ACEH operation:
as shown in fig. 3, assume that the given hash key is 00100100110.. 11111110, and since the global depth g (global depth) is 2, the first two bits of the highest order bits are used as Segment indexes and the least significant byte is used as a Bucket index. L represents Local Depth (Local Depth), and L is G-log2 kK represents the number of pointers pointing to the data bucket, and a local depth L is adopted because it is not necessary that one Directory entry (Directory) corresponds to one segment after the extensible hash expansion, and since the number of Directory entries after the expansion is more than the number of original segments, a new segment is created only when a key value pair needs to be moved to a non-existing data bucket, as shown in fig. 4, a Directory entry double expansion (Directory Doubling) operation: and expanding according to the Most Significant Bits, wherein white represents the directory entry before expansion, and gray represents the newly added directory entry after expansion.
Insert operation: when the hash key 00100100100110.. 11111110 is inserted, firstly, the 00 directory item is indexed and positioned through the highest bit (Most signaled Bits), the Segment pointed by the 00 directory item is found, then two insertable buckets are found according to an Adjusted-Cuckoo algorithm, data in the two buckets are sequentially traversed, and if a key is the same as the inserted key, the value corresponding to the current key is updated; if the key does not exist and the key is inserted, if two packets have free positions, the hash key is inserted randomly, if only one packet is free, the free packet is inserted, if two packets do not have free positions, the insertion fails, and Split operation is carried out.
Update operation: for the refresh operation, because the ACEH adopts the linear probing method during the insertion, the key value to be updated can be directly compared with the key stored in each slot in the probing process, and if the key value is the same, the key value is updated.
And (3) Split operation: suppose that the key 00100100100110.. 11111110 is to be inserted into the buckets 2 and 254 after being calculated by the Adjusted-Cuckoo algorithm, but neither of the two buckets has extra space to store data, at this time, the ACEH creates a new Segment4, and at the same time, an extra pointer pointing to Segment3, that is, a pointer of 11 directory entries, points to Segment4 instead, and the valid hash key value pair whose highest bit is 11 in the original Segment3 is also transferred to Segment 4.
Deletion operation: with lazy deletion, migrated records are not deleted immediately. After updating the directory entries, queries searching for migrated records will access the new segment, while queries searching for non-migrated records will access the old segment, but since the split segments (i.e., the old segments) contain all keys, they will always succeed in finding records that search for key values, which contain some unnecessary duplicates. In the process of expanding the inert expansion, when part of key value pairs are migrated, the key value pairs in the original data bucket are not deleted, but when new data are inserted into the original data bucket, the migrated data are directly replaced, and therefore the expenditure of the hash table in the process of expanding can be reduced.
For example, insert a record with a hash key of 1010.. 11111110, then access segment3, assume that the Adjusted-Cuckoo algorithm calculates that to insert Bucket2 and Bucket254, find that Bucket254 is full, the hash key of the first record in Bucket2 is 1001.. 0010 valid, but the hash key of the second record is 1111.. 0010 invalid, and then insert the transaction to replace the second record with the new record. Since the validity of each record is determined by the local depth, the order in which directory entries are updated must be preserved to maintain consistency and fault coverage.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. An ACEH index structure based on memory hotspot-oriented data, comprising: directory entries and data buckets, characterized in thatAn intermediate structure called segment, between the directory entry and the segment by using global depth G as segment index, a segment includes a group of data buckets, between the segment and the data bucket using local depth L as data bucket index, L is G-log2 kAnd k represents the number of pointers pointing to the data bucket, the data bucket index adopts an Adjusted-Cuckoo algorithm to position a data bucket inserted by a hash key, the Adjusted-Cuckoo algorithm comprises two hash functions, two insertable data buckets are generated, then an empty data bucket is selected for insertion, the Adjusted-Cuckoo algorithm determines one data bucket, and the second data bucket is directly determined as the next data bucket of the current data bucket.
2. A memory hot data oriented ACEH indexing method is characterized in that a directory entry carries out segment indexing through global depth G, one segment corresponds to a group of data buckets, the segment carries out data bucket indexing through local depth L, and L is G-log2 kAnd k represents the number of pointers pointing to the data bucket, the data bucket index adopts an Adjusted-Cuckoo algorithm to position a data bucket inserted by a hash key, the Adjusted-Cuckoo algorithm comprises two hash functions, two insertable data buckets are generated, then an empty data bucket is selected for insertion, the Adjusted-Cuckoo algorithm determines one data bucket, and the second data bucket is directly determined as the next data bucket of the current data bucket, the operation method comprises the following steps:
step one, inserting operation;
step two, refreshing operation;
step three, splitting operation;
and step four, deleting operation.
3. The ACEH indexing method based on memory oriented hot spot data as claimed in claim 2, wherein the Adjusted-Cuckoo algorithm determines a data bucket by a multiplex-shift function.
4. The memory hotspot data oriented ACEH indexing method according to claim 2 or 3, wherein the inserting operation of the first step locates the directory entry through the highest bit, finds the segment pointed by the directory entry, then finds two insertable data buckets according to the Adjusted-Cuckoo algorithm, sequentially traverses the data in the two data buckets, and updates the value corresponding to the current key if the key is the same as the insert key; and if the key does not exist and the key is inserted, randomly inserting the hash key if both the two data buckets have free positions, inserting the free data buckets if only one data bucket is free, and if both the two data buckets have no free positions, failing to insert and performing splitting operation.
5. The ACEH indexing method based on memory hot-spot facing data according to claim 2 or 3, wherein the refresh operation in step two is performed by in-situ refresh, and the key to be updated is directly compared with the key stored in each slot in the probing process, and if the key is the same, the key is updated.
6. The method according to claim 2 or 3, wherein the splitting operation in step three creates a new segment, and turns a redundant directory entry pointer pointing to the original segment to point to the new segment, and the valid key-value pair corresponding to the directory entry pointer is also transferred to the new segment.
7. The method according to claim 2 or 3, wherein the deleting operation in step four is a lazy delete, and the key-value pair of the original data bucket that has been migrated is not deleted immediately, but replaced directly when new data is inserted into the original data bucket.
8. The memory-oriented hot-spot data based ACEH indexing method as claimed in claim 7, wherein the lazy deletion, when deleting data x1 in Bucket0 data Bucket, directly overwrites x1 with data x2 in Bucket1 data Bucket and marks data at x2 position as invalid.
CN202011296272.2A 2020-11-18 2020-11-18 ACEH index structure and method based on memory hot spot data Active CN112395213B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011296272.2A CN112395213B (en) 2020-11-18 2020-11-18 ACEH index structure and method based on memory hot spot data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011296272.2A CN112395213B (en) 2020-11-18 2020-11-18 ACEH index structure and method based on memory hot spot data

Publications (2)

Publication Number Publication Date
CN112395213A true CN112395213A (en) 2021-02-23
CN112395213B CN112395213B (en) 2023-05-30

Family

ID=74606497

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011296272.2A Active CN112395213B (en) 2020-11-18 2020-11-18 ACEH index structure and method based on memory hot spot data

Country Status (1)

Country Link
CN (1) CN112395213B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113342706A (en) * 2021-05-13 2021-09-03 武汉大学 Write-optimized extensible hash index structure based on nonvolatile memory and inserting, refreshing and deleting methods
CN113505130A (en) * 2021-07-09 2021-10-15 中国科学院计算技术研究所 Hash table processing method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06231023A (en) * 1993-10-19 1994-08-19 Olympus Optical Co Ltd Information recorder
CN103106144A (en) * 2011-11-15 2013-05-15 北京新媒传信科技有限公司 Compressing method and device of internal storage index
CN110069496A (en) * 2019-03-20 2019-07-30 韶关学院 A kind of Novel chain type Hash table construction method and device
CN111459846A (en) * 2020-03-12 2020-07-28 华中科技大学 Dynamic hash table operation method based on hybrid DRAM-NVM

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06231023A (en) * 1993-10-19 1994-08-19 Olympus Optical Co Ltd Information recorder
CN103106144A (en) * 2011-11-15 2013-05-15 北京新媒传信科技有限公司 Compressing method and device of internal storage index
CN110069496A (en) * 2019-03-20 2019-07-30 韶关学院 A kind of Novel chain type Hash table construction method and device
CN111459846A (en) * 2020-03-12 2020-07-28 华中科技大学 Dynamic hash table operation method based on hybrid DRAM-NVM

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
侯捷: "《STL源码剖析》", 30 June 2002 *
海纳: "《自己动手写Python虚拟机》", 30 June 2019 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113342706A (en) * 2021-05-13 2021-09-03 武汉大学 Write-optimized extensible hash index structure based on nonvolatile memory and inserting, refreshing and deleting methods
CN113505130A (en) * 2021-07-09 2021-10-15 中国科学院计算技术研究所 Hash table processing method
CN113505130B (en) * 2021-07-09 2023-07-21 中国科学院计算技术研究所 Hash table processing method

Also Published As

Publication number Publication date
CN112395213B (en) 2023-05-30

Similar Documents

Publication Publication Date Title
CN110083601B (en) Key value storage system-oriented index tree construction method and system
CN111459846B (en) Dynamic hash table operation method based on hybrid DRAM-NVM
US7539153B1 (en) Method and apparatus for longest prefix matching based on a trie
CN105975587B (en) A kind of high performance memory database index organization and access method
CN112395213A (en) Memory hot-spot data oriented ACEH index structure and method
US8086641B1 (en) Integrated search engine devices that utilize SPM-linked bit maps to reduce handle memory duplication and methods of operating same
US5659739A (en) Skip list data structure enhancements
US7464100B2 (en) Reorganization-free mapping of objects in databases using a mapping chain
CN109189759B (en) Data reading method, data query method, device and equipment in KV storage system
CN107766258B (en) Memory storage method and device and memory query method and device
CN115718819A (en) Index construction method, data reading method and index construction device
Jensen et al. Optimality in external memory hashing
CN113342706A (en) Write-optimized extensible hash index structure based on nonvolatile memory and inserting, refreshing and deleting methods
CN116821127A (en) Method for realizing hash index of kv stored distributed database
CN116701440A (en) Cuckoo filter and data insertion, query and deletion method
CN110928835A (en) Novel file storage system and method based on mass storage
TWI761440B (en) memory access method
CN114531450A (en) Height-based block chain peer-to-peer network data synchronization method
CN111538465B (en) Linux-based high-performance data deduplication method
CN112269784A (en) Hash table structure based on hardware realization and inserting, inquiring and deleting method
CN113779320B (en) Method for solving table entry storage address conflict
CN117743338B (en) Keyword matching method and device based on double hash
CN115599711B (en) Cache data processing method, system, device, equipment and computer storage medium
CN110673952B (en) Data processing method and device for high-concurrency read application
CN117390031B (en) Verification method for validity of secondary index in storage system based on LSM tree

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
GR01 Patent grant
GR01 Patent grant