Non-Markovian Modeling of a BladeCenter Chassis Midplane

详细信息    查看全文

• 作者：Salvatore Distefano (17)
  Francesco Longo (18)
  Marco Scarpa (18)
  Kishor S. Trivedi (19)
  
• 刊名：Lecture Notes in Computer Science
• 出版年：2014
• 出版时间：2014
• 年：2014
• 卷：8721
• 期：1
• 页码：255-269
• 全文大小：291 KB
• 参考文献：1. Resnick, R.I.: A Modern Taxonomy of High Availability. Technical report, Interlog (1996), http://www.generalconcepts.com/resources/reliability/resnick/HA.htm
  2. Cisco Systems Inc.: Data Center High Availability Clusters Design Guide (2006), http://www.ciscocertified.info/en/US/docs/solutions/Enterprise/Data_Center/HA/_Clusters/HA/_Clusters.html
  3. Ghosh, R., Longo, F., Frattini, F., Russo, S., Trivedi, K.: Scalable analytics for iaas cloud availability. IEEE Transactions on Cloud Computing聽2(1), 57鈥?0 (2014) CrossRef
  4. Zheng, J., Okamura, H., Dohi, T.: Component importance analysis of virtualized system. In: 2012 9th International Conference on Ubiquitous Intelligence Computing and 9th International Conference on Autonomic Trusted Computing (UIC/ATC), pp. 462鈥?69 (September 2012)
  5. Muppala, J., Malhotra, M., Trivedi, K.: Markov dependability models of complex systems: Analysis techniques. In: 脰zekici, S. (ed.) Reliability and Maintenance of Complex Systems. NATO ASI Series, vol.聽154, pp. 442鈥?86. Springer, Heidelberg (1996) CrossRef
  6. Fricks, R., Yin, L., Trivedi, K.: Application of semi-markov process and CTMC to evaluation of UPS system availability. In: RAMS 2002, January 28鈥?1, pp. 584鈥?91 (2002)
  7. Mura, I., Bondavalli, A.: Markov regenerative stochastic petri nets to model and evaluate phased mission systems dependability. IEEE Transactions on Computers聽50(12), 1337鈥?351 (2001) CrossRef
  8. Longo, F., Scarpa, M.: Applying symbolic techniques to the representation of non-markovian models with continuous PH distributions. In: Bradley, J.T. (ed.) EPEW 2009. LNCS, vol.聽5652, pp. 44鈥?8. Springer, Heidelberg (2009); Cited by (since 1996)6
  9. Longo, F., Scarpa, M.: Two-layer symbolic representation for stochastic models with phase-type distributed events. International Journal of Systems Science, 1鈥?2
  10. P茅rez-Oc贸n, R., Castro, J.E.R.: Two models for a repairable two-system with phase-type sojourn time distributions. Reliability Engineering & System Safety聽84(3), 253鈥?60 (2004) CrossRef
  11. Smith, W.E., Trivedi, K., Tomek, L., Ackaret, J.: Availability analysis of blade server systems. IBM Systems Journal聽47(4), 621鈥?40 (2008) CrossRef
  12. Bruneo, D., Distefano, S., Longo, F., Puliafito, A., Scarpa, M.: Reliability assessment of wireless sensor nodes with non-linear battery discharge (2010); Cited by (since 1996)5
  13. Bruneo, D., Distefano, S., Longo, F., Puliafito, A., Scarpa, M.: Evaluating wireless sensor node longevity through markovian techniques. Comput. Netw.聽56(2), 521鈥?32 (2012) CrossRef
  14. Sedyakin, N.: On one physical principle in reliability theory. Tekhn. Kibernetika (in Russian - Technical Cybernetics)聽3, 80鈥?7 (1966)
  
• 作者单位：Salvatore Distefano (17)
  Francesco Longo (18)
  Marco Scarpa (18)
  Kishor S. Trivedi (19)
  
  17. Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, 20133, Milano, Italy
  18. Dipartimento di Ingegneria DICIEAMA, Universit脿 di Messina, 98166, Messina, Italy
  19. Department of Electrical and Computer Engineering, Duke University, 27708-0291, Durham, USA
  
• ISSN：1611-3349

文摘

In distributed contexts such as Cloud computing, the reliability and availability of the provided resources and services have to be assured in order to meet user requirements. At the infrastructure level, this specification is translated into tighter ones on the datacenter hosting physical resources. In this paper, starting from a real case study of the IBM BladeCenter, we provide a technique for the quantitative evaluation of datacenter infrastructure availability. The proposed technique allows one to take into account both aging phenomena and multiple operating conditions. In particular, one subsystem of the BladeCenter, the chassis midplane, is studied. Indeed, based on the stochastic characterization of the midplane reliability through statistic measurements, a model dealing with the non-exponential failure time distribution thus obtained is evaluated to demonstrate the suitability and the effectiveness of the proposed technique.