The Power of Better Choice: Reducing Relocations in Cuckoo Filter

Wang, Fei‐Yue; Chen, Hanhua; Liao, Liangyi; Zhang, Fan; Jin, Hai

doi:10.1109/icdcs.2019.00043

Cited by 7 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A cuckoo filter with a load of 75 percent already has 75 percent of its buckets in use. However, research shows that the higher the load factor of a cuckoo filter, the more false negatives will start appearing [9,10]. During our experimentation, a load factor lower than 65 percent prevented the system to generate false negatives when checking the existence of keys.…”

Section: Cuckoo Filtersmentioning

confidence: 83%

“…Further optimization in this area will be needed for our system. Also, the cuckoo filters in [10][11][12][13] can provide a better performance result than using the cuckoo filters in [20]. This solution can close the breach of performance between cuckoo filters and bloom filters and enhance performance when querying and deleting keys in the LSM tree.…”

Section: Lsm Tree Testing Resultsmentioning

confidence: 99%

“…Findings still show that bloom filters offer on average 1.5x times faster insertion performance than the developed LSM tree implementing cuckoo filters for high and low false positive rates. Other studies [10][11][12][13] showed other extensions of newly developed cuckoo filters that can guarantee better space efficiency and low latency by doing a different set of operations when choosing candidate buckets than the original cuckoo filters [20]. In the original study [20], cuckoo filters choose the bucket candidates randomly, and as stated in [9], this affects the load balance of the filter.…”

Section: Lsm Tree Testing Resultsmentioning

confidence: 99%

“…Research in [10] shows mathematically and experimentally that Cuckoo filters can suffer from performance degradation during element insertion because of its random eviction strategy for the candidate buckets. This random selection also brings loadbalancing problems to the data structure and thus leads to frequent relocations.…”

Section: Novel Cuckoo Filters Related Researchmentioning

confidence: 99%

See 3 more Smart Citations

LSM Tree Optimization Through The Implementation of Cuckoo Filters

Valverde,

Kim,

Kim

et al. 2023

Preprint

View full text Add to dashboard Cite

This study aims to develop an LSM tree that uses two different kinds of probabilistic data structures (PDS). These two data structures are the Bloom Filters and the state-of-the-art Cuckoo Filter released in 2014 by Fan. Cuckoo filters are a perfect choice for saving space and also for deleting keys that are not needed in the algorithm so the filters are synchronized with the keys of all the SSTables. The implementation in this study of Cuckoo filters showed three main advantages (1) Keys that are deleted in the SSTables are immediately reflected in the Cuckoo Filters (2) When SSTables are merged, Cuckoo Filters are also merged by measuring its load factor for availability (3) The performance of Cuckoo Filters compared to Bloom filters after deleting and query shows better performance despite that cuckoo filters are also being updated. This approach is expected to reduce storage space and the amount of time for queries and deletions on the overall data structure. This new optimized LSM tree is expected to be implemented in HBase as future work to check its performance with a real-world database for more empirical results.

show abstract

Section: Cuckoo Filtersmentioning

confidence: 83%

Section: Lsm Tree Testing Resultsmentioning

confidence: 99%

Section: Lsm Tree Testing Resultsmentioning

confidence: 99%

Section: Novel Cuckoo Filters Related Researchmentioning

confidence: 99%

See 2 more Smart Citations

LSM Tree Optimization Through The Implementation of Cuckoo Filters

Valverde,

Kim,

Kim

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Determining whether a particular element belongs to a given set is a common problem in computer science, with widespread applications in bioinformatics, machine learning, computer networks, the Internet of Things, and database systems [39]. Filter data structures such as Bloom filters and Cuckoo filters can approximately determine if an element is part of a specified set and have been extensively applied in network routing [40], information retrieval, file merging [41], spam detection [42], and distributed systems [43].…”

Section: Cuckoo Filtermentioning

confidence: 99%

Enhancing Efficiency and Security in Unbalanced PSI-CA Protocols through Cloud Computing and Homomorphic Encryption in Mobile Networks

Tan,

Du,

Weng

2024

Future Internet

View full text Add to dashboard Cite

Private Set Intersection Cardinality (PSI-CA) is a cryptographic method in secure multi-party computation that allows entities to identify the cardinality of the intersection without revealing their private data.Traditional approaches assume similar-sized datasets and equal computational power, overlooking practical imbalances. In real-world applications, dataset sizes and computational capacities often vary, particularly in Internet of Things and mobile scenarios where device limitations restrict computational types. Traditional PSI-CA protocols are inefficient here, as computational and communication complexities correlate with the size of larger datasets. Thus, adapting PSI-CA protocols to these imbalances is crucial. This paper explores unbalanced scenarios where one party (the receiver) has a relatively small dataset and limited computational power, while the other party (the sender) has a large amount of data and strong computational capabilities.This paper, based on the concept of commutative encryption, introduces Cuckoo filter, cloud computing technology, and homomorphic encryption, among other technologies, to construct three novel solutions for unbalanced Private Set Intersection Cardinality (PSI-CA): an unbalanced PSI-CA protocol based on Cuckoo filter, an unbalanced PSI-CA protocol based on single-cloud assistance, and an unbalanced PSI-CA protocol based on dual-cloud assistance. Depending on performance and security requirements, different protocols can be employed for various applications.

show abstract