A Stochastic Model for File Lifetime and Security in Data Center Networks

  • Quan-Lin LiEmail author
  • Fan-Qi Ma
  • Jing-Yu Ma
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11280)


Data center networks are an important infrastructure in various applications of modern information technologies. Note that each data center always has a finite lifetime, thus once a data center fails, then it will lose all its storage files and useful information. For this, it is necessary to replicate and copy each important file into other data centers such that this file can increase its lifetime of staying in a data center network. In this paper, we describe a large-scale data center network with a file d-replication policy, which is to replicate each important file into at most \(d-1\) other data centers such that this file can maintain in the data center network under a given level of data security in the long-term. To this end, we develop three relevant Markov processes to propose two effective methods for assessing the file lifetime and data security. By using the RG-factorizations, we show that the two methods are used to be able to more effectively evaluate the file lifetime of large-scale data center networks. We hope the methodology and results given in this paper are applicable in the file lifetime and data security study of more general data center networks with replication mechanism.


Data center Replication mechanism File lifetime Data security Markov process RG-factorization 


  1. 1.
    Aghajani, R., Robert, P., Sun, W.: A large scale analysis of unreliable stochastic networks. Ann. Appl. Prob. 28(2), 851–887 (2018)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Blake, C., Rodrigues, R.: High availability, scalable storage, dynamic peer networks: pick two. In: The 9th Workshop on Hot Topics in Operating Systems, vol. 3, pp. 18–21 (2003)Google Scholar
  3. 3.
    Chun, B.G., et al.: Efficient replica maintenance for distributed storage systems. In: The 3rd Symposium on Networked Systems Design & Implementation, vol. 6, pp. 45–58 (2006)Google Scholar
  4. 4.
    Feuillet, M., Robert, P.: A scaling analysis of a transient stochastic network. Adv. Appl. Probab. 46(2), 516–535 (2014)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Kersch, P., Szabo, R.: Mathematical modeling of routing in DHTs. In: Shen, X., Yu, H., Buford, J., Akon, M. (eds.) Handbook of Peer-to-Peer Networking. Springer, Boston (2010). Scholar
  6. 6.
    Kniesburges, S., Koutsopoulos, A., Scheideler, C.: CONE-DHT: a distributed self-stabilizing algorithm for a heterogeneous storage system. In: Afek, Y. (ed.) DISC 2013. LNCS, vol. 8205, pp. 537–549. Springer, Heidelberg (2013). Scholar
  7. 7.
    Lian, Q., Chen, W., Zhang, Z.: On the impact of replica placement to the reliability of distributed brick storage systems. In: The 25th IEEE International Conference on Distributed Computing Systems, pp. 187–196. IEEE (2005)Google Scholar
  8. 8.
    Li, Q.L.: Constructive Computation in Stochastic Models with Applications: The RG-Factorizations. Springer, Berlin (2010). Scholar
  9. 9.
    Li, Q.L., Cao, J.: Two types of RG-factorizations of quasi-birth-and-death processes and their applications to stochastic integral functionals. Stoch. Models 20(3), 299–340 (2004)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Pace, A., Quema, V., Schiavoni, V.: Exploiting node connection regularity for DHT replication. In: The 30th IEEE Symposium on Reliable Distributed Systems, pp. 111–120. IEEE (2011)Google Scholar
  11. 11.
    Picconi, F., Baynat, B., Sens, P.: An analytical estimation of durability in DHTs. In: Janowski, T., Mohanty, H. (eds.) ICDCIT 2007. LNCS, vol. 4882, pp. 184–196. Springer, Heidelberg (2007). Scholar
  12. 12.
    Picconi, F., Baynat, B., Sens, P.: Predicting durability in DHTs using Markov chains. In: The 2nd International Conference on Digital Information Management, vol. 2, pp. 532–538. IEEE (2007)Google Scholar
  13. 13.
    Pinheiro, E., Weber, W.D., Barroso, L.A.: Failure trends in a large disk drive population. In: The 5th USENIX Conference on File and Storage Technologies, vol. 7, No. 1, pp. 17–23 (2007)Google Scholar
  14. 14.
    Ramabhadran, S., Pasquale, J.: Analysis of long-running replicated systems. In: INFOCOM 2006, pp. 1–9 (2006)Google Scholar
  15. 15.
    Ramabhadran, S., Pasquale, J.: Durability of replicated distributed storage systems. ACM SIGMETRICS Perform. Eval. Rev. 36(1), 447–448 (2008)CrossRefGoogle Scholar
  16. 16.
    Ramabhadran, S., Pasquale, J.: Analysis of durability in replicated distributed storage systems. In: IEEE International Symposium on Parallel & Distributed Processing, pp. 1–12. IEEE (2010)Google Scholar
  17. 17.
    Sun, W., Feuillet, M., Robert, P.: Analysis of large unreliable stochastic networks. Ann. Appl. Probab. 26(5), 2959–3000 (2016)MathSciNetCrossRefGoogle Scholar
  18. 18.
    Utard, G., Vernois, A.: Data durability in peer to peer storage systems. In: IEEE International Symposium on CLUSTER Computing and the Grid, pp. 90–97. IEEE (2004)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.School of Economics and Management SciencesYanshan UniversityQinhuangdaoChina

Personalised recommendations