基于K-means的动态聚类垃圾回收算法研究  

Research on Dynamic Clustering Garbage Collection Algorithm Based on K-means

作　　者：傅宬彧 严华[1] FU Chengyu;YAN Hua(College of Electronics and Information Engineering,Sichuan University,Chengdu 610065,China)

机构地区：[1]四川大学电子信息学院,四川成都610065

出　　处：《现代信息科技》2025年第7期93-97,共5页Modern Information Technology

摘　　要：在基于NAND闪存的消费电子产品中,垃圾回收显著影响设备性能,而冷热数据分离是优化垃圾回收算法的关键。文章提出了一种创新方法,利用K-means聚类算法自动确定数据热度区间,实现高效的冷热数据分离。同时,设计了一种自适应的影响因子调节策略,用于动态平衡历史数据热度与近期数据热度的影响。在此基础上,对垃圾回收策略进行了优化。仿真结果表明,与现有算法相比,该方法在提升读写性能、实现磨损均衡性方面效果显著,有助于延长设备寿命并提高整体性能。In consumer electronics products based on NAND flash memory,garbage collection significantly affects device performance,and cold-hot data separation is the key to optimizing the garbage collection algorithm.This paper proposes an innovative method that uses the K-means clustering algorithm to automatically determine the data heat intervals and achieve efficient cold-hot data separation.Additionally,an adaptive influence factor adjustment strategy is designed to dynamically balance the influence of historical data heat and recent data heat.On this basis,the garbage collection strategy is optimized.The simulation results show that compared with the existing algorithms,this method has a significant effect on improving the readwrite performance and achieving wear leveling,which is helpful in prolonging the service life of the equipment and improving the overall performance.

关 键 词：闪存 垃圾回收 冷热分离 K-MEANS 聚类 

分 类 号：TP333[自动化与计算机技术—计算机系统结构] TP301.6[自动化与计算机技术—计算机科学与技术]