High Performance Computing Systems with Various Checkpointing Schemes

  • Nichamon Naksinehaboon Louisiana Tech University Computer Science Department Louisiana, 71272, USA
  • Mihaela P[un Louisiana Tech University Mathematics and Statistics Department Louisiana, 71272, USA and Finance and Banks Faculty Spiru Haret University, Romania E-mail:
  • Raja Nassar Louisiana Tech University Mathematics and Statistics Department Louisiana, 71272, USA
  • Chokchai Box Leangsuksun Louisiana Tech University Computer Science Department Louisiana, 71272, USA
  • Stephen Scott Oak Ridge National Laboratory Computer Science and Mathematics Division TN 37831-6173


Finding the failure rate of a system is a crucial step in high performance computing systems analysis. To deal with this problem, a fault tolerant mechanism, called checkpoint/ restart technique, was introduced. However, there are additional costs to perform this mechanism. Thus, we propose two models for different schemes (full and incremental checkpoint schemes). The models which are based on the reliability of the system are used to determine the checkpoint placements. Both proposed models consider a balance of between checkpoint overhead and the re-computing time. Due to the extra costs from each incremental checkpoint during the recovery period, a method to find the number of incremental checkpoints between two consecutive full checkpoints is given. Our simulation suggests that in most cases our incremental checkpoint model can reduce the waste time more than it is reduced by the full checkpoint model. The waste times produced by both models are in the range of 2% to 28% of the application completion time depending on the checkpoint overheads.


[1] A.R. Adiga, G Almasi, and et al., An overview of the BlueGene/L supercomputer, Proceedings of Supercomputing, IEEE/ACM Conference, pp. 60-60, 2002.

[2] J.T. Daly, A Model for Predicting the Optimum Checkpoint Interval for Restart Dumps, ICCS 2003, LNCS 2660, Volume 4, pp. 3–12, 2003.

[3] J.T. Daly, A Higher Order Estimate of the Optimum Checkpoint Interval for Restart Dumps, Future Generation Computer Systems, Elsevier, Amsterdam, 2004.

[4] E. Elnozahy, J. Plank, Checkpointing for Peta-Scale Systems: A Look into the Future of Practical Rollback- Recovery, IEEE Transactions on Dependable and Secure Computing, vol.01,no.2, pp. 97-108, 2004.

[5] R. Geist, R. Reynolds, and J. Westall, Selection of a checkpoint interval in a critical-task environment, IEEE Transactions on Reliability, vol.37, no.4, pp. 395-400, 1988.

[6] I. M. Gelfand and S. V. Fomin, Calculus of Variations, Dover Publications, ISBN-10: 0486414485, 2000.

[7] L. Hancu, Data-Mining Techniques for Supporting Merging Decisions,Int. J. of Computers, Communications and Control,Vol. III (2008), pp. 322-326.

[8] D. I. Hunyadi, M. A. Musan, Modelling of the Distributed Databases. A Viewpoint Mechanism of the MVDB Model's Methodology, Int. J. of Computers, Communications and Control, Vol. III (2008),pp. 327-332.

[9] Y. Ling, J. Mi, and X. Lin, A Variational Calculus Approach to Optimal Checkpoint Placement, IEEE Transactions on Computers, vol. 50, no. 7, pp. 699-707, 2001.

[10] A.J. Oliner, L. Rudolph, and R.K. Sahoo, Cooperative Checkpointing: A Robust Approach to Large-scale Systems Reliability, Proceedings of the 20th Annual International Conference on Supercomputing (ICS), Australia, pp. 14-23, 2006.

[11] T. Ozaki, T. Dohi, and H. Okamura, Distribution-Free Checkpoint Placement Algorithms Based on Min-Max Principle, IEEE Transactions on Dependable and Secure Computing, Volume 3, Issue 2, pp. 130 – 140, 2006.

[12] A. C. Palaniswamy, and P. A. Wilsey, An analytical comparison of periodic checkpointing and incremental state saving, Proc. of the Seventh Workshop on Parallel and Distributed Simulation, California, pp. 127-134, 1993.

[13] J.S. Plank, J. Xu, and R.H. Netzer, 1995a. Compressed differences: an algorithm for fast incremental checkpointing, Technical Report CS-95-302, University of Tennessee at Knoxville, 1995.

[14] J.S. Plank, M.A. Thomason, The Average Availability of Parallel Checkpointing Systems and Its Importance in Selecting Runtime Parameters, The 29th International Symposium on Fault-Tolerant Computing, Madison, WI, pp. 250-259, 1999.

[15] S.M. Ross, Stochastic Processes, Wiley; 2nd edition, ISBN-10: 0471120626, 1995.

[16] B. Schroeder, G.A. Gibson, A large-scale study of failures in high-performance computing systems, Proceedings of International Symposium on Dependable Systems and Networks (DSN). IEEE Computer Society, pp. 249–258, 2006.

[17] A. Tikotekar, C. Leangsuksun, S. Scott, On the survivability of standard MPI applications, In Proceedings of 7th LCI International Conference on Linux Clusters: The HPC Revolution 2006.

[18] N. H. Vaidya, A case for two-level distributed recovery schemes, in Proceedings of the 1995 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems, pp. 64–73, 1995.

[19] K. F. Wong, M.A. Franklin, Distributed Computing Systems and Checkpointing, HPDC, pp. 224-233, 1993.

[20] J.W. Young, A first-order approximation to the optimum checkpoint interval, Communications of ACM volume 17, Issue 9, pp. 530-531, 1974.

[21] Y. Liu, R. Nassar, C. B. Leangsuksun, N. Naksinehaboon, M. Paun, and S. L. Scott, An optimal checkpoint/ restart model for a large scale high performance computing system, in Proc. International Parallel and Distributed Processing Symposium, (Miami, Florida, 2008) pp.1-9, 2008.

[22] D. Zmaranda, G. Gabor, Issues on Optimality Criteria Applied in Real-Time Scheduling,Int. J. of Computers, Communications and Control, Issue 3, pp.536-540, 2008.
How to Cite
NAKSINEHABOON, Nichamon et al. High Performance Computing Systems with Various Checkpointing Schemes. INTERNATIONAL JOURNAL OF COMPUTERS COMMUNICATIONS & CONTROL, [S.l.], v. 4, n. 4, p. 386-400, dec. 2009. ISSN 1841-9844. Available at: <http://univagora.ro/jour/index.php/ijccc/article/view/2455>. Date accessed: 05 july 2020. doi: https://doi.org/10.15837/ijccc.2009.4.2455.


Large-scale distributed system, reliability, fault-tolerance, checkpoint/restart model, HPC