Algorithm for Detecting Implicitly Faulty Replicas Based on the Power Consumption Model

  • Hazuki IshiiEmail author
  • Ryuji Oma
  • Shigenari Nakamura
  • Tomoya Enokido
  • Makoto Takizawa
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 97)


A system can be fault-tolerant by replicating each application process on multiple servers. Each replica is performed on a host server. According to the advances of hardware and architecture technologies of servers, each server can be considered to be free of fault, i.e. always proper. On the other hand, replicas of application processes easily suffer from faults, e.g. infected with virus. A faulty replica may send a proper reply, e.g. wiretapped reply. A replica which sends a proper reply but does faulty computation is implicitly faulty. Implicitly faulty replicas cannot be detected by checking the replies. It takes a longer or shorter time and a server supporting a faulty replica consumes more or smaller electric energy since the faulty replica does computation different from a proper replica. In this paper, we propose an algorithm to detect implicitly faulty replicas of a process by using the power consumption and computation models of a server in addition to checking replies in a cluster.


Process faults Implicitly faulty replica Power consumption model Computation model Fault detection 


  1. 1.
    Intel xeon processor 5600 series: The next generation of intelligent server processors, white paper (2010).
  2. 2.
    Job scheduling algorithms in Linux virtual server (2010).
  3. 3.
    Linux Operating Systems.
  4. 4.
    Bernstein, P.A., Goodman, N.: The failure and recovery problem for replicated databases. In: Proceedings of the 2nd ACM Symposium on Principles of Distributed Computing, pp. 114–122 (1998)Google Scholar
  5. 5.
    Defago, X., Schiper, A., Sergent, N.: Semi-passive replication. In: Proceedings of IEEE the 17th Symposium on Reliable Distributed Systems, pp. 43–50 (1998)Google Scholar
  6. 6.
    Denning, D.E.R.: Cryptography and Data Security. Addison Wesley, Boston (1982)zbMATHGoogle Scholar
  7. 7.
    Deplanche, A.M., Theaudiere, P.Y., Trinquet, Y.: Implementing a semi-active replication strategy in CHORUS/ClassiX, a distributed real-time executive. In: Proceedings of IEEE the 18th Symposium on Reliable Distributed Systems, pp. 90–101 (1999)Google Scholar
  8. 8.
    Duolikun, D., Aikebaier, A., Enokido, T., Takizawa, M.: Energy-aware passive replication of processes. Int. J. Mob. Multimed. 9(1,2), 53–65 (2013)Google Scholar
  9. 9.
    Duolikun, D., Enokido, T., Takizawa, M.: Dynamic migration of virtual machines to reduce energy consumption in a cluster international journal of grid and utility computing. Int. J. Grid Utility Comput. 9(4), 357–366 (2018)CrossRefGoogle Scholar
  10. 10.
    Duolikun, D., Kataoka, H., Enokido, T., Takizawa, M.: Simple algorithms for selecting an energy-efficient server in a cluster of servers. Int. J. Commun. Netw. Distrib. Syst. 21(1), 1–25 (2018)CrossRefGoogle Scholar
  11. 11.
    Enokido, T., Aikebaier, A., Takizawa, M.: A model for reducing power consumption in peer-to-peer systems. IEEE Syst. J. 4(2), 221–229 (2010)CrossRefGoogle Scholar
  12. 12.
    Enokido, T., Aikebaier, A., Takizawa, M.: Process allocation algorithms for saving power consumption in peer-to-peer systems. IEEE Trans. Ind. Electron. 58(6), 2097–2105 (2011)CrossRefGoogle Scholar
  13. 13.
    Enokido, T., Ailixier, A., Takizawa, M.: An extended simple power consumption model for selecting a server to perform computation type processes in digital ecosystems. IEEE Trans. Ind. Inform. 10(2), 1627–1636 (2014)CrossRefGoogle Scholar
  14. 14.
    Enokido, T., Takizawa, M.: An integrated power consumption model for distributed systems. IEEE Trans. Ind. Electron. 60(2), 824–836 (2013)CrossRefGoogle Scholar
  15. 15.
    Fischer, M.J., Lynch, N.A., Paterson, M.S.: Impossibility of distributed consensus with one faulty process. In: Proceedings of the Second ACM SIGACT-SIGMOD Symposium on Principles of Database Systems, 21–23 March 1983, Colony Square Hotel, Atlanta, Georgia, USA, pp. 1–7 (1983)Google Scholar
  16. 16.
    Hayashibara, N., Takizawa, M.: Design of the notification system for failure detectors. Int. J. High Perform. Comput. Netw. 6(1), 25–34 (2009)CrossRefGoogle Scholar
  17. 17.
    Ishii, H., Oma, R., Nakamura, S., Enokido, T., Takizawa, M.: Fault detection of process replicas on reliable servers. In: Proceedings of the 22nd International Conference on Network-Based Information System, NBiS-2019 (2019)Google Scholar
  18. 18.
    Kataoka, H., Duolikun, D., Enokido, T., Takizawa, M.: Energy-efficient virtualisation of threads in a server cluster. In: Proceedings of the 10th International Conference on Broadband and Wireless Computing, Communication and Applications, BWCCA-2015, pp. 288–295 (2015)Google Scholar
  19. 19.
    Kataoka, H., Nakamura, S., Duolikun, D., Enokido, T., Takizawa, M.: Multi-level power consumption model and energy-aware server selection algorithm. Int. J. Grid Utility Comput. (IJGUC) 8(3), 201–210 (2017)CrossRefGoogle Scholar
  20. 20.
    Kataoka, H., Sawada, A., Duolikun, D., Enokido, T., Takizawa, M.: Energy-aware server selection algorithm in a scalable cluster. In: Proceedings of IEEE the 30th International Conference on Advanced Information Networking and Applications, AINA-2016, pp. 565–572 (2016)Google Scholar
  21. 21.
    Lamport, L., Shostak, R., Pease, M.: The Byzantine generals problems. ACM Trans. Program. Lang. Syst. 4(3), 382–401 (1992)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Hazuki Ishii
    • 1
    Email author
  • Ryuji Oma
    • 1
  • Shigenari Nakamura
    • 1
  • Tomoya Enokido
    • 1
  • Makoto Takizawa
    • 1
  1. 1.Hosei UniversityTokyoJapan

Personalised recommendations