A method and system for optimizing multi-fingerprint cuckoo filters

By generating multiple fingerprints for each data point and utilizing multiple candidate buckets for the cuckoo filter operation, the problem of high false positive rate in existing cuckoo filters is solved, achieving optimization of the false positive rate and flexible utilization of storage space.

CN123116244APending Publication Date: 2026-11-17HUAZHONG UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611035328.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-13
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The existing Cuckoo filter uses a single fingerprint matching and verification mechanism, which cannot utilize spare storage space to optimize the false judgment rate, and the false judgment probability caused by fingerprint collision is too high.

Method used

The multi-fingerprint cuckoo filter optimization method is adopted. By generating multiple independent bit fingerprints for each data to be inserted and calculating the corresponding candidate bucket index, the insertion, query and deletion operations are performed using multiple candidate buckets to optimize the false positive rate.

Benefits of technology

It enables the flexible conversion of surplus space into false positive rate, reducing the false positive rate, while requiring no modification to the existing storage structure and hardware deployment. It adapts to different space and false positive rate requirements, with low migration cost and low computational complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116244A_ABST
    Figure CN123116244A_ABST
Patent Text Reader

Abstract

This invention proposes a multi-fingerprint cuckoo filter optimization method and system, belonging to the interdisciplinary technical fields of cyberspace security, data structure, and data existence detection. Addressing the core deficiency of existing technologies that determine data existence based on single fingerprint matching, making it difficult to fully utilize storage space to optimize the false positive rate when there is ample storage space, this invention generates multiple independent fingerprints for each data point. During insertion, each fingerprint is stored in either of its two candidate buckets. During querying, all fingerprints must exist in their respective candidate buckets for data existence to be determined. During deletion, any valid copy of each fingerprint of the data is deleted. Through rigorous mathematical derivation, this invention proves that this scheme can flexibly utilize ample storage space to reduce the false positive rate while maintaining reliable dynamic addition and deletion characteristics, while remaining fully compatible with the traditional cuckoo filter storage structure and hash logic. It has low implementation complexity and is suitable for dynamic data existence detection scenarios such as data flow compliance detection and distributed caching.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the interdisciplinary fields of cyberspace security, data structure and data existence detection, and specifically relates to a multi-fingerprint cuckoo filter optimization method and system. Background Technology

[0002] The Cuckoo Filter is a probabilistic data structure that supports dynamic insertion and deletion operations. Since its introduction, it has been widely used in technical scenarios requiring rapid data existence determination, such as data flow compliance verification, network packet filtering, and distributed system caching. The Cuckoo Filter stores a fixed-length digest, called a fingerprint, obtained by hash compression of the data. Compared to traditional Bloom filters, Cuckoo Filters support data deletion and, under certain conditions, exhibit better performance in preventing false positives, making them more widely applicable in dynamic data applications.

[0003] Traditional cuckoo filters rely on XOR operations to calculate candidate bucket indexes, offering advantages such as simple computation and hardware compatibility. Their basic working logic is as follows: For a single piece of data to be stored, a fingerprint and a first candidate bucket index are generated. Then, another candidate bucket index is obtained by XORing the first candidate bucket index with the fingerprint hash value. The fingerprint can be stored in either of the two candidate buckets. During a query, if the target fingerprint matches in any candidate bucket, the data is considered to exist. Deletion is performed by removing any matching fingerprint from either bucket.

[0004] However, existing traditional cuckoo filters have shortcomings: only one fingerprint is generated for a single piece of data, and the query rule is that the data is determined to exist if the single fingerprint is matched. When there is spare storage space in the filter, the existing architecture cannot optimize the probability of false judgment through idle space, and the utilization rate of storage resources is limited.

[0005] To address the aforementioned problems, this invention proposes a multi-fingerprint cuckoo filter optimization method and system. Summary of the Invention

[0006] The purpose of this invention is to propose a multi-fingerprint cuckoo filter optimization method and system to solve the problem that the existing cuckoo filter adopts a single fingerprint matching and verification mechanism, which cannot utilize the spare storage space to optimize the false judgment rate, and the false judgment probability caused by fingerprint collision is too high.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A method for optimizing a multi-fingerprint cuckoo filter includes the following steps: S1. Filter Initialization: Configure filter parameters, including the number of buckets. The number of fingerprints each bucket can hold Maximum number of kicks , and the number of fingerprints generated from each data point. ,in Integers greater than or equal to 2

[0009] S2, Fingerprint Generation: For each piece of data to be inserted... ,generate An independent Bit fingerprint in ; S3, Bucket Search Calculation: For each fingerprint Calculate its candidate bucket index 0. And candidate bucket index 1 ; S4. Data Insertion: For the data to be inserted, try inserting the following data in sequence: Each fingerprint is inserted one by one into its corresponding candidate bucket 0 or candidate bucket 1. If all fingerprints are successfully inserted, the data insertion is considered successful; if any fingerprint fails to be inserted or the number of times it is removed exceeds the limit, the data insertion is considered successful. Then the data insertion is deemed to have failed; S5. Data Query: For the data to be queried, check the following steps sequentially. If a fingerprint exists in the corresponding candidate bucket 0 or candidate bucket 1, the data is considered to exist; if any fingerprint does not exist, the data is considered not to exist. S6. Data Deletion: For data to be deleted, search sequentially. If a fingerprint is found in any valid copy of the corresponding candidate bucket, then the valid copy is deleted and the deletion is considered successful; if any fingerprint does not exist, the deletion is considered unsuccessful.

[0010] Preferably, the bucket search calculation in step S3 satisfies symmetry: Given any bucket index and fingerprints Then the other bucket index is ,in," " represents the XOR operation, and hash() represents the hash function.

[0011] Preferably, the data insertion in step S4 specifically includes: S41, Regarding the first fingerprint Check its candidate bucket index 0 Is it not full? If not, insert. Proceed to process the next fingerprint; S42. If bucket 0 is full, check the index of its candidate bucket 1. Is it not full? If not, insert. Proceed to process the next fingerprint; S43. If both buckets are full, randomly select one bucket, randomly kick out a fingerprint from it, and insert the fingerprint to be inserted into that position. S44. Calculate the index of another candidate bucket for the kicked-out fingerprint. If the bucket is not full, insert the kicked-out fingerprint; if the bucket is full, repeat steps S43-S44 until insertion is successful or the number of kick-outs exceeds the limit. ; S45. If the number of kicks exceeds If all All fingerprints were successfully inserted, indicating successful insertion.

[0012] Preferably, the data query in step S5 specifically includes: S51, Regarding the first fingerprint Check its candidate bucket index 0 Does it exist in If it exists, proceed to process the next fingerprint; S52. If it does not exist in bucket 0, check the index of its candidate bucket 1. Does it exist in If it exists, proceed to process the next fingerprint; S53. If the fingerprint is not found in either of the two buckets, the data is determined to be non-existent; if all... All fingerprints exist, and the determination data exists.

[0013] 5. The multi-fingerprint cuckoo filter optimization method according to claim 1 or 4, characterized in that the data deletion in step S6 specifically includes: S61, Regarding the first fingerprint Following the steps in S5, find any valid copy of the fingerprint and record the bucket index and offset of the copy within the bucket; if no fingerprint is found, the deletion fails. S62, If all If all fingerprints are found, delete all valid copies of the fingerprints sequentially based on the recorded bucket index and offset, and the deletion is considered successful.

[0014] A multi-fingerprint cuckoo filter optimization system includes: The initialization module is used to configure filter parameters and initialize the bucket array; The fingerprint generation module is used to generate fingerprints for the data to be processed. An independent Bit-based valid fingerprint; The bucket search module is used to calculate the indexes of the two candidate buckets corresponding to each fingerprint; The insert module is used to perform data insertion operations. Each fingerprint is inserted into the corresponding candidate bucket; The query module is used to perform data query operations, only when all The system will only return a data existence if all fingerprints exist in the corresponding candidate bucket. The deletion module is used to perform data deletion operations, deleting any valid copy of each fingerprint.

[0015] The present invention further protects a computer device, the computer device including a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the instruction, program, code set or instruction set being loaded and executed by the processor to implement the above-mentioned multi-fingerprint cuckoo filter optimization method.

[0016] The present invention further protects a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the instruction, program, code set, or instruction set is loaded and executed by a processor to implement the above-described multi-fingerprint cuckoo filter optimization method.

[0017] The present invention has the following beneficial effects: (1) It realizes the flexible transformation of surplus space into false positive rate: by adjusting the number of fingerprints This can convert excess storage space into a lower false positive rate.

[0018] (2) Fully compatible with traditional cuckoo filters: No need to modify the existing storage structure, hash logic and hardware deployment, it can directly replace the existing single fingerprint cuckoo filter, only requiring adjustment The parameters can be adapted to different spatial and error rate requirements, and the migration cost is extremely low.

[0019] (3) Low implementation complexity: The core operations only use hash and XOR operations, which are easy to implement in software and hardware. Attached Figure Description

[0020] Figure 1 This is an overall flowchart of a multi-fingerprint cuckoo filter optimization method proposed in this invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0022] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0023] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0024] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0025] Example 1 This embodiment uses specific data operation examples to illustrate in detail the multi-fingerprint cuckoo filter optimization method and system proposed in this invention. This invention can effectively utilize storage space and optimize the query misjudgment rate.

[0026] This embodiment sets the following initial parameters: The filter contains m=8192 buckets, b=4 fingerprints per bucket, k=2 fingerprints generated per data, l=8 bits of fingerprint length, and T=500 maximum number of ejections allowed during data insertion. It should be noted that these parameters are for illustrative purposes only and can be flexibly adjusted according to storage capacity and false positive rate requirements in practical applications.

[0027] Based on the above parameter settings, the specific implementation process is as follows: Step 1, Filter Initialization: Initialize an array containing 8192 buckets, each of which can store 4 8-bit fingerprints.

[0028] Step 2, Data Insertion: For the data to be inserted Perform the following operations: (1) Generate multiple fingerprints: Using a preset hash function, generate two independent 8-bit fingerprints for data x:

[0029]

[0030] (2) Calculate candidate bucket indices: For each generated fingerprint, calculate its indices in the two candidate buckets within the cuckoo filter. Index calculation relies on a hash function and an XOR operation:

[0031]

[0032]

[0033]

[0034] (3) Insert the first fingerprint: Checking buckets 1, 2, 3, and 4, they were found to be not full; insert 0. x 3 A .

[0035] (4) Insert the second fingerprint: Checking bucket #1212, it was found to be full; checking buckets #3456, they were found to be not full, so a 0 was inserted. x 5 F .

[0036] (5) Both fingerprints were successfully inserted, and the data insertion was successful.

[0037] Step 3, Data Query: Query data Does it exist? (1) Generate two fingerprints and candidate bucket indexes in the manner described in step 2; (2) Check if 0x3A exists in buckets 1234; it was found to exist. (3) Check if 0x5F exists in container 1212; it is not found. (4) Check if 0x5F is present in container 3456; it is found to be present. (5) Both fingerprints exist, and the returned data exists.

[0038] Step 4: Data Deletion Delete data : (1) Find two valid copies of the fingerprints in the manner of step 3, and record the second position of the first position of the first and second positions of the second and third positions of the second and third positions of the second and third positions of the second and third positions of the third and fourth positions of the second and third ... respectively. (2) Delete 0x3A in the second position of bucket 1234 and 0x5F in the first position of bucket 3456; (3) All fingerprints have been successfully deleted. The deletion success message is returned.

[0039] Example 2: Based on Embodiment 1, but with a difference, this embodiment uses a specific error rate analysis to illustrate that the multi-fingerprint scheme provided by the present invention can effectively reduce the false judgment rate of the cuckoo filter while ensuring storage efficiency. The specific analysis process is as follows.

[0040] 1. Single fingerprint solution Cuckoo filter false positive rate 1.1 Maximum False Positive Rate The classic paper on traditional cuckoo filters proposes a formula for the maximum false positive rate, which reflects the theoretical upper limit of the false positive rate and is only related to the fingerprint length. And the number of fingerprints each bucket can hold Related to the amount of inserted data And the number of buckets Irrelevant:

[0041] The proof is as follows: The query needs to be compared with a maximum of two buckets. The probability of any fingerprint colliding with the fingerprint being queried is 1 / 2. Therefore all The probability that none of the fingerprints will collide is The probability of a misjudgment is .

[0042] 1.2 Dynamic Misjudgment Rate Traditional maximum false positive rate formulas only reflect the theoretical upper limit and cannot accurately reflect the actual false positive rate during runtime based on the actual amount of inserted data and the number of buckets. This invention derives a dynamic false positive rate formula for the single-fingerprint cuckoo filter:

[0043] in: For the number of buckets, This represents the number of data entries to be inserted.

[0044] The proof is as follows: When determining the existence of data, the Cuckoo Filter needs to calculate the data's fingerprint and the index positions of the two buckets, and check if there exists a fingerprint in either bucket that matches the data to be determined. If so, the data is considered to exist. For the Cuckoo Filter to misjudge, two conditions must be met simultaneously: the fingerprint of the data to be determined must be the same as the fingerprint of the already inserted data, and the index positions of the buckets for the fingerprints of the data to be determined and the fingerprints of the already inserted data must be the same. According to the properties of the Cuckoo Filter, it is impossible for two data sets to have the same fingerprints but only one bucket index position. If two data sets have the same fingerprint and one bucket index position, then the index position of the other bucket must also be the same. Since it cannot distinguish whether the fingerprint already inserted into the Cuckoo Filter is in bucket 0 or bucket 1, the following two situations will lead to a misjudgment of data existence: The first case is: the fingerprint of the data to be determined to exist is the same as that of the inserted data, and the index position of the 0th bucket of the fingerprint of the data to be determined to exist is equal to the index position of the 0th bucket of the fingerprint of the inserted data, and the index position of the 1st bucket of the fingerprint of the data to be determined to exist is equal to the index position of the 1st bucket of the fingerprint of the inserted data.

[0045] The second scenario is: the fingerprint of the data to be determined to exist is the same as that of the inserted data, and the index position of the 1st bucket of the fingerprint of the data to be determined to exist is equal to the index position of the 0th bucket of the fingerprint of the inserted data.

[0046] When the existence of data not inserted into the Cuckoo filter is misjudged due to a single already inserted data, the following conditions must be met: the buckets have the same index position and the fingerprints are identical. The probability that the buckets have the same index position is... The probability of fingerprints being the same is There are two possible scenarios for misjudgment, and the probability of either scenario occurring is 0.25%. Therefore, the probability of misjudging the existence of data without the cuckoo filter due to a single inserted data is... The probability of misjudging a data point not based on a single inserted data point is: , not by The probability of misjudging an already inserted data is: ,because The probability of misjudging an already inserted data is: , which is the dynamic misjudgment rate of the cuckoo filter.

[0047] 1.3 Relationship between maximum false positive rate and dynamic false positive rate The dynamic false positive rate of the cuckoo filter is no greater than the maximum false positive rate, i.e.

[0048] Prove the above equation, that is, prove .

[0049] make That is, proof .

[0050] make ,but At Monotonically increasing. And... l Representing the fingerprint length, which is a positive integer, therefore we have , .therefore At Monotonically increasing. For the cuckoo filter, each fingerprint can be stored in 2 buckets, therefore the cuckoo filter has at least 2 buckets. .because At Monotonically increasing, we can obtain Expanding this formula, we can obtain... .because and It can be known that Therefore, there is .

[0051] When the total number of buckets in the cuckoo filter is Each bucket can hold When there are multiple fingerprints, if it can accommodate... The fingerprint of each piece of data then has Therefore, there is .Depend on , have to This is proven. Therefore, the dynamic false positive rate of the cuckoo filter is no greater than the maximum false positive rate.

[0052] 2. False positive rate of the multi-fingerprint solution's cuckoo filter 2.1 Maximum False Positive Rate Determining the existence of a fingerprint requires comparing it with at most two buckets. Each fingerprint is compared. Therefore, when determining the existence of data without filters, the maximum probability of a single fingerprint existing is... , The maximum probability that all fingerprints exist is: Therefore, the maximum false positive rate of the cuckoo filter in the multi-fingerprint scheme is... .

[0053] 2.2 Dynamic Misjudgment Rate For the cuckoo filter in the multi-fingerprint scheme, data is inserted into the filter each time it is inserted. A fingerprint. Insert. When inserting data, a total of [number] data points are used. There are fingerprints. If the filter has a total of Each bucket has a fingerprint length of [number]. When determining the existence of data without filters, the probability of a single fingerprint existing is: , The probability that all fingerprints exist is: Therefore, the dynamic false positive rate of the cuckoo filter in the multi-fingerprint scheme is... .

[0054] 2.3 Relationship between maximum false positive rate and dynamic false positive rate Under the multi-fingerprint scheme, the dynamic false positive rate of the cuckoo filter is no greater than the maximum false positive rate, i.e.

[0055] Prove the above equation, that is, prove .make That is, proof .

[0056] make ,but At Monotonically increasing. And... Representing the fingerprint length, which is a positive integer, therefore we have , .therefore At Monotonically increasing. For the cuckoo filter, each fingerprint can be stored in 2 buckets, therefore there are .because At Monotonically increasing, we can obtain Expanding this formula, we can obtain... .because It can be known that Therefore, there is .

[0057] When the number of barrels in the multi-fingerprint scheme's cuckoo filter is Each bucket can hold When there are multiple fingerprints, if it can accommodate... Data One fingerprint, then there are Therefore, there is .Depend on , have to This is proven. Therefore, the dynamic false positive rate of the cuckoo filter in the multi-fingerprint scheme is no greater than the maximum false positive rate.

[0058] 3. Comparison of false positive rates between single-fingerprint and multi-fingerprint solutions using the Cuckoo Filter 3.1 Maximum False Positive Rate The maximum false positive rate of the cuckoo filter in the multi-fingerprint scheme is The maximum false positive rate of the cuckoo filter in the single fingerprint scheme is... .because The maximum false positive rate of the multi-fingerprint cuckoo filter is lower than that of the single-fingerprint cuckoo filter.

[0059] 3.2 Dynamic Misjudgment Rate When the number of buckets is And the fingerprint length is When inserting a multi-fingerprint scheme's cuckoo filter, Data When using one fingerprint, the dynamic false positive rate is... As for the cuckoo filter in the single fingerprint scheme, when inserted... Data When using one fingerprint, the dynamic false positive rate is... .

[0060] when When there is a minimum value, It has a minimum value. Let Then it satisfies .make If the cuckoo filter in the multi-fingerprint scheme has a lower dynamic false positive rate, then there exists an integer... , making , , .

[0061] right Taking the derivative, we get .when hour, .therefore, At Monotonically decreasing, Monotonically increasing.

[0062] If the multi-fingerprint scheme's cuckoo filter has a lower dynamic false positive rate, then an integer is required. , making .when , At that time, such a thing did not exist. .when , At that time, such a situation may exist. At this point, it is necessary to make .if but Timely satisfaction , and if This means The minimum point is at There are no integers between them. Make .

[0063] ,Right now .when When, the inequality holds.

[0064] ,therefore This is precisely the dynamic false positive rate formula for the cuckoo filter in the single fingerprint scheme. achievable Therefore, when the dynamic false positive rate of the cuckoo filter in a single fingerprint scheme is lower than... At that time, the dynamic false judgment rate of the multi-fingerprint cuckoo filter was lower than that of the single-fingerprint cuckoo filter.

[0065] The dynamic false positive rate of the cuckoo filter is no greater than its maximum false positive rate. Therefore, when the maximum false positive rate is lower than... At the same time, the dynamic misjudgment rate is also lower than At this point, the dynamic false positive rate of the cuckoo filter in the multi-fingerprint scheme is lower. The formula for the maximum false positive rate of the cuckoo filter in the single-fingerprint scheme is: The number of fingerprints each bucket can hold. The value is usually 4, so as long as the fingerprint length is... This allows the maximum false positive rate of the cuckoo filter to be lower than [a certain value]. At this time, the dynamic false positive rate will also be lower than If the cuckoo filter can accommodate fingerprints under a multi-fingerprint scheme, then the false positive rate can be optimized through a multi-fingerprint scheme.

[0066] In conclusion, by setting multiple fingerprints for each data point, the maximum false positive rate is definitely reduced, and the number of fingerprints that each bucket can hold is also reduced. If the fingerprint length is 4, This also reduces the false positive rate. By setting multiple fingerprints for each data point, the false positive rate is optimized.

[0067] It should be noted that the specific embodiments described above are exemplary. Those skilled in the art can devise various solutions inspired by the disclosure of this invention, and these solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and not intended to limit the scope of the claims. The scope of protection of this invention is defined by the claims and their equivalents. This specification contains multiple inventive concepts; terms such as "preferredly," "according to a preferred embodiment," or "optionally" indicate that the corresponding paragraph discloses an independent concept. The applicant reserves the right to file divisional applications based on each inventive concept.

Claims

1. A method for optimizing a multi-fingerprint cuckoo filter, characterized in that, Includes the following steps: S1. Filter Initialization: Configure filter parameters, including the number of buckets. The number of fingerprints each bucket can hold Maximum number of kicks fingerprint length And the number of fingerprints generated for each piece of data. ,in It is an integer greater than or equal to 2; S2, Fingerprint Generation: For each piece of data to be inserted... ,generate An independent Bit fingerprint ,in ; S3, Bucket Search Calculation: For each fingerprint Calculate its candidate bucket index 0. And candidate bucket index 1 ; S4. Data Insertion: For the data to be inserted, try inserting the following data in sequence: Each fingerprint is inserted one by one into its corresponding candidate bucket 0 or candidate bucket 1. If all fingerprints are successfully inserted, the data insertion is considered successful; if any fingerprint fails to be inserted or the number of times it is removed exceeds the limit, the data insertion is considered successful. Then the data insertion is deemed to have failed; S5. Data Query: For the data to be queried, check the following steps sequentially. If a fingerprint exists in the corresponding candidate bucket 0 or candidate bucket 1, the data is considered to exist; if any fingerprint does not exist, the data is considered not to exist. S6. Data Deletion: For data to be deleted, search sequentially. If a fingerprint is found in any valid copy of the corresponding candidate bucket, then the valid copy is deleted and the deletion is considered successful; if any fingerprint does not exist, the deletion is considered unsuccessful.

2. The multi-fingerprint cuckoo filter optimization method according to claim 1, characterized in that, The bucket search calculation described in step S3 satisfies symmetry: Given any bucket index and fingerprints Then the other bucket index is ,in," " represents the XOR operation, and hash(·) represents the hash value calculation.

3. The multi-fingerprint cuckoo filter optimization method according to claim 1, characterized in that, The data insertion in step S4 specifically includes: S41, Regarding the first fingerprint Check its candidate bucket index 0 Is it not full? If not, insert. Proceed to process the next fingerprint; S42. If bucket 0 is full, check the index of its candidate bucket 1. Is it not full? If not, insert. Proceed to process the next fingerprint; S43. If both buckets are full, randomly select one bucket, randomly kick out a fingerprint from it, and insert the fingerprint to be inserted into that position. S44. Calculate the index of another candidate bucket for the kicked-out fingerprint. If the bucket is not full, insert the kicked-out fingerprint; if the bucket is full, repeat steps S43-S44 until insertion is successful or the number of kick-outs exceeds the limit. ; S45. If the number of kicks exceeds If all All fingerprints were successfully inserted, indicating successful insertion.

4. The multi-fingerprint cuckoo filter optimization method according to claim 1, characterized in that, The data query described in step S5 specifically includes: S51, Regarding the first fingerprint Check its candidate bucket index 0 Does it exist in If it exists, proceed to process the next fingerprint; S52. If it does not exist in bucket 0, check the index of its candidate bucket 1. Does it exist in If it exists, proceed to process the next fingerprint; S53. If the fingerprint is not found in either of the two buckets, the data is determined to be non-existent; if all... All fingerprints exist, and the determination data exists.

5. The multi-fingerprint cuckoo filter optimization method according to claim 1 or 4, characterized in that, The data deletion described in step S6 specifically includes: S61, Regarding the first fingerprint Following the steps in S5, find any valid copy of the fingerprint and record the bucket index and offset of the copy within the bucket; if no fingerprint is found, the deletion fails. S62, If all If all fingerprints are found, delete all valid copies of the fingerprints sequentially based on the recorded bucket index and offset, and the deletion is considered successful.

6. A multi-fingerprint cuckoo filter optimization system for implementing the method described in any one of claims 1-5, characterized in that, include: The initialization module is used to configure filter parameters and initialize the bucket array; The fingerprint generation module is used to generate fingerprints for the data to be processed. An independent Bit-based valid fingerprint; The bucket search module is used to calculate the indexes of the two candidate buckets corresponding to each fingerprint; The insert module is used to perform data insertion operations. Each fingerprint is inserted into the corresponding candidate bucket; The query module is used to perform data query operations, only when all The system will only return a data existence if all fingerprints exist in the corresponding candidate bucket. The deletion module is used to perform data deletion operations, deleting any valid copy of each fingerprint.

7. A computer device, characterized in that, The computer device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set, or instruction set, and the instruction, program, code set, or instruction set is loaded and executed by the processor to implement the multi-fingerprint cuckoo filter optimization method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, at least one program, code set, or instruction set, which is loaded and executed by a processor to implement the multi-fingerprint cuckoo filter optimization method as described in any one of claims 1-7.