Big Data Research ›› 2018, Vol. 4 ›› Issue (3): 61-80.doi: 10.11959/j.issn.2096-0271.2018031

• STUDY • Previous Articles     Next Articles

Analysis on hybrid memory architecture for big data application

Xin LI1,Xuan CHEN2,Zhiqiu HUANG1   

  1. 1 College of Computer Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China
    2 College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China
  • Online:2018-05-15 Published:2018-05-30
  • Supported by:
    The National High Technology Research and Development Program of China(2015AA01530);Jiangsu Natural Science Foundation(BK20160813)


Due to the limited scalability of DRAM,it is hard to optimize the performance of big data analysis and the big data applications.The new non-volatile memory (NVM) brings the opportunity to improve the performance and efficiency for big data applications,which benefits by the advantages of NVM,including its non-volatile,high storage density,and low power consumption.The PCM/DRAM hybrid memory architecture based on the non-volatile memory was analyzed.The feasibility and advantages of hybrid memory for big data applications through the analysis on the optimization of performance,energy consumption and memory management strategies for hybrid memory architecture were demonstrated.The defects in existing work were summarized and the potential research field in PCM/DRAM hybrid memory architecture was discussed.

Key words: big data, non-volatile memory, phase change memory, performance optimization

CLC Number: 

No Suggested Reading articles found!