Open AccessThe Application and Improvement Discussion of Gossip in NoSQL Databases: Taking Cassandra as an Analysis Example
Author(s) -
Yichen Qian
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1437/1/012001
Subject(s) - gossip , gossip protocol , nosql , computer science , scalability , protocol (science) , flooding (psychology) , fault tolerance , weighting , database , distributed computing , psychology , social psychology , pathology , psychotherapist , radiology , medicine , alternative medicine
The application of gossip in NoSQL has a promising future and has significant academic value. By using gossip, a large storage system can be more scalable and stable with better fault-tolerance. This study takes Cassandra as an example to study the advantages of applying the gossip protocol in NoSQL by comparing with the conditions of using other protocols, such as Flooding algorithm and Structured Broadcast algorithm. To improve the efficiency and scalability when the size of individual gossip populations is limited, the author proposes a reconciliation mechanism named weighting. This study reveals that the gossip protocol in Cassandra contributes to linear scalability with limited individual gossip populations by implementing a simulation of gossip transmission rounds.