Cluster Computing

, Volume 14, Issue 4, pp 325–344 | Cite as

Performance evaluation of a remote memory system with commodity hardware for large-memory data processing

  • Hyuck Han
  • Hyungsoo JungEmail author
  • Sooyong Kang
  • Heon Y. Yeom


The explosion of data and transactions demands a creative approach for data processing in a variety of applications. Research on remote memory systems (RMSs), so as to exploit the superior characteristics of dynamic random access memory (DRAM), has been performed for many decades, and today’s information explosion galvanizes researchers into shedding new light on the technology. Prior studies have mainly focused on architectural suggestions for such systems, highlighting different design rationale. These studies have shown that choosing the appropriate applications to run on an RMS is important in fully utilizing the advantages of remote memory. This article provides an extensive performance evaluation for various types of data processing applications so as to address the efficacy of an RMS by means of a prototype RMS with reliability functionality. The prototype RMS used is a practical kernel-level RMS that renders large memory data processing feasible. The abstract concept of remote memory was materialized by borrowing unused local memory in commodity PCs via a high speed network capable of Remote Direct Memory Access (RDMA) operations. The prototype RMS uses remote memory without any part of its computation power coming from remote computers. Our experimental results suggest that an RMS can be practical in supporting the rigorous demands of commercial in memory database systems that have high data access locality. Our evaluation also convinces us of the possibility that a reliable RMS can satisfy both the high degree of reliability and efficiency for large memory data processing applications whose data access pattern has high locality.


Remote memory system Large memory data processing Main memory databases 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Vishnu, A., Gupta, P., Mamidala, A.R., Panda, D.K.: A software based approach for providing network fault tolerance in clusters with udapl interface: mpi level design and performance evaluation. In: SC’06: Proceedings of the 2006 ACM/IEEE Conference on Supercomputing (2006) Google Scholar
  2. 2.
    Vishnu, A., Gupta, P., Mamidala, A.R., Panda, D.K.: An efficient hardware-software approach to network fault tolerance with infiniband. In: Proceedings of the IEEE Cluster 2009 (2009) Google Scholar
  3. 3.
  4. 4.
  5. 5.
    Transcation Processing Performance Council. TPC-H.
  6. 6.
  7. 7.
    Shin, D.: About SSD. In: Proc. of the USENIX Linux Storage and Filesystem Workshop (LSF’08) (2008) Google Scholar
  8. 8.
    Amza, C., Cox, A.L., Dwarkadas, S., Keleher, P., Lu, H., Rajamony, R., Yu, W., Zwaenepoel, W.: Treadmarks: shared memory computing on networks of workstations. IEEE Comput. 29(2), 18–28 (1996) CrossRefGoogle Scholar
  9. 9.
    Carter, J.: Munin: Efficient Distributed Shared Memory Using Multi-Protocol Release Consistency. Ph.D. Thesis, Rice University (1992) Google Scholar
  10. 10.
    Keleher, P., Cox, A.L., Dwarkadas, S., Zwaenepoel, W.: An evaluation of software based release consistent protocols. J. Parallel Distributed Syst. 29(2), 126–141 (1995) Google Scholar
  11. 11.
    Keleher, P.: Distributed Shared Memory Using Lazy Release Consistency. Ph.D. Thesis, Rice University (1994) Google Scholar
  12. 12.
    Black, D., Gupta, A., Weber, W.D.: Competitive management of distributed shared memory. In: Sprint COMPCON 89 Digest of Papers (1989) Google Scholar
  13. 13.
    Bolosky, W., Scott, M., Fitzgerald, R.: Simple but effective techniques for numa memory management. In: Proc. of ACM SOSP (1989) Google Scholar
  14. 14.
    Holliday, M.: Reference history, page size, and migration daemons in local/remote architectures. In: Proc. of ACM ASPLOS (1989) Google Scholar
  15. 15.
    Bolosky, W., Scott, M., Fitzgerald, R., Fowler, R., Cox, A.: Numa policies and their relationship to memory architecture. In: Proc. of ACM ASPLOS (1991) Google Scholar
  16. 16.
    Markatos, E.P., Dramitinos, G.: Implementation of a reliable remote memory pager. In: Proc. of the 1996 USENIX Technical Conference (1996) Google Scholar
  17. 17.
    Liang, S., Noronha, R., Panda, D.K.: Swapping to remote memory over infiniband: an approach using a high performance network block device. In: Proc. of the 2005 IEEE Cluster Computing (2005) Google Scholar
  18. 18.
    Freeley, M.J., Morgan, W.E., Pighin, F.H., Karlin, A.R., Levy, H.M.: Implementing global memory management in a workstation cluster. In: Proc. of ACM SOSP (1995) Google Scholar
  19. 19.
    Jiang, S., Petrini, F., Ding, X., Zhang, X.: A locality-aware cooperative cache management protocol to improve network file system performance. In: ICDCS’06: Proceedings of the 26th IEEE International Conference on Distributed Computing Systems (2006) Google Scholar
  20. 20.
    Jiang, S., Davis, K., Zhang, X.: Coordinated multilevel buffer cache management with consistent access locality quantification. IEEE Trans. Comput. 56(1) (2007) Google Scholar
  21. 21.
    Comer, D., Griffioen, J.: A new design for distributed systems: the remote memory model. In: Proc. of USENIX Tech. Conf. (1990) Google Scholar
  22. 22.
    Frankling, M.J., Carey, M.J., Livny, M.: Global memory management in client-server DBMS architectures. In: Proc. of VLDB (1992) Google Scholar
  23. 23.
    Dahlin, M.D., Wang, R.Y., Anderson, T.E., Paterson, D.A.: Cooperative caching: using remote client memory to improve file system performance. In: Proc. of USENIX OSDI (1994) Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Hyuck Han
    • 1
  • Hyungsoo Jung
    • 2
    Email author
  • Sooyong Kang
    • 3
  • Heon Y. Yeom
    • 1
  1. 1.School of Computer Science and EngineeringSeoul National UniversitySeoulRepublic of Korea
  2. 2.School of Information TechnologiesUniversity of SydneySydneyAustralia
  3. 3.Division of Computer Science and EngineeringHanyang UniversitySeoulRepublic of Korea

Personalised recommendations