SwapKV: A Hotness Aware In-Memory Key-Value Store for Hybrid Memory Systems

被引:3
作者
Cui, Lixiao [1 ]
He, Kewen [1 ]
Li, Yusen [1 ]
Li, Peng [2 ]
Zhang, Jiachen [1 ]
Wang, Gang [1 ]
Liu, Xiaoguang [1 ]
机构
[1] Nankai Univ, Coll Comp Sci, Nankai Baidu Joint Lab, TJ Key Lab NDST, Tianjin 300350, Peoples R China
[2] ByteDance Inc, Beijing 100089, Peoples R China
基金
美国国家科学基金会;
关键词
Key-value system; persistent memory; hybrid memory system;
D O I
10.1109/TKDE.2021.3077264
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In-memory Key-Value (KV) stores are widely deployed in modern data centers. These systems generally use DRAM as their storage medium, causing huge hardware costs. The emerging persistent memory (PMEM) is a potential substitute for DRAM, which has a lower price and larger capacity, but lower access speed and bandwidth. Many prior studies strive to build hybrid memory systems to retain both the advantages of DRAM and PMEM. However, they are either application agnostic or simply take DRAM as a cache, which are both not efficient for in-memory KV stores. In this paper, we propose SwapKV, a well-designed in-memory KV store for hybrid DRAM-PMEM system. SwapKV has several promising properties. First, SwapKV combines DRAM and PMEM to a uniform memory pool and only stores one copy of data, which maximizes capacity utilization. Second, SwapKV maps all writing operations to DRAM and migrates data to PMEM with large blocks asynchronously, which mitigates the intrinsic inefficiency of PMEM for writing operations. Third, SwapKV maintains the hot data in DRAM through an efficient hotness filtering and data swapping mechanism, which ensures high system throughput and responsiveness. We implement SwapKV and evaluate it under various workload patterns. The results demonstrate that SwapKV improves the throughput by 11 similar to 41 percent compared to the state-of-the-art alternatives.
引用
收藏
页码:917 / 930
页数:14
相关论文
共 31 条
  • [1] Hotspot-Aware Hybrid Memory Management for In-Memory Key-Value Stores
    Jin, Hai
    Li, Zhiwei
    Liu, Haikun
    Liao, Xiaofei
    Zhang, Yu
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2020, 31 (04) : 779 - 792
  • [2] NStore: A High-Performance NUMA-Aware Key-Value Store for Hybrid Memory
    Wang, Zhonghua
    Lu, Kai
    Wan, Jiguang
    Jiang, Hong
    Zhao, Zeyang
    Xu, Peng
    Lai, Biliang
    Li, Guokuan
    Xie, Changsheng
    IEEE TRANSACTIONS ON COMPUTERS, 2025, 74 (03) : 929 - 943
  • [3] ChameleonDB: a Key-value Store for Optane Persistent Memory
    Zhang, Wenhui
    Zhao, Xingsheng
    Jiang, Song
    Jiang, Hong
    PROCEEDINGS OF THE SIXTEENTH EUROPEAN CONFERENCE ON COMPUTER SYSTEMS (EUROSYS '21), 2021, : 194 - 209
  • [4] HaCache: A Hybrid Adaptive Cache for Persistent Memory Based Key-Value Systems
    Cui, Lixiao
    Wang, Gang
    Li, Yusen
    Liu, Xiaoguang
    WEB AND BIG DATA, PT I, APWEB-WAIM 2022, 2023, 13421 : 180 - 189
  • [5] A Multicore-Friendly Persistent Memory Key-Value Store
    Wang Q.
    Zhu B.
    Shu J.
    Jisuanji Yanjiu yu Fazhan/Computer Research and Development, 2021, 58 (02): : 397 - 405
  • [6] HyperKV: A High Performance Concurrent Key-Value Store for Persistent Memory
    Sun, Penghao
    Xue, Dongliang
    You, Litong
    Yan, Yan
    Huang, Linpeng
    19TH IEEE INTERNATIONAL SYMPOSIUM ON PARALLEL AND DISTRIBUTED PROCESSING WITH APPLICATIONS (ISPA/BDCLOUD/SOCIALCOM/SUSTAINCOM 2021), 2021, : 125 - 134
  • [7] PetaKV: Building Efficient Key-Value Store for File System Metadata on Persistent Memory
    Zhang, Yiwen
    Zhou, Jian
    Min, Xinhao
    Ge, Song
    Wan, Jiguang
    Yao, Ting
    Wang, Daohui
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2023, 34 (03) : 843 - 855
  • [8] FastStore: A High-Performance RDMA-enabled Distributed Key-Value Store with Persistent Memory
    Xiong, Ziwei
    Jiang, Dejun
    Xiong, Jin
    2023 IEEE 43RD INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS, ICDCS, 2023, : 406 - 417
  • [9] TrieKV: A High-Performance Key-Value Store Design With Memory as Its First-Class Citizen
    Sun, Hui
    Kong, Deyan
    Jiang, Song
    Yue, Yinliang
    Qin, Xiao
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2024, 35 (12) : 2479 - 2496
  • [10] SpacKV: A Pmem-Aware Key-Value Separation Store Based on LSM-Tree
    Ge, Xuran
    Lai, Mingche
    Liu, Yang
    Wu, Lizhou
    Zhuang, Zhutao
    Ou, Yang
    Chen, Zhiguang
    Xiao, Nong
    NETWORK AND PARALLEL COMPUTING, NPC 2022, 2022, 13615 : 327 - 339