@InProceedings{oldfield:2007:modeling_checkpoints, author = {Ron A. Oldfield and Sarala Arunagiri and Patricia J. Teller and Seetharami Seelam and Rolf Riesen and Maria Ruiz Varela and Philip C. Roth}, title = {Modeling the Impact of Checkpoints on Next-Generation Systems}, booktitle = {Proceedings of the 24th IEEE Conference on Mass Storage Systems and Technologies}, year = {2007}, month = {September}, address = {San Diego, CA}, DOI = {10.1109/MSST.2007.4367962}, URL = {http://dx.doi.org/10.1109/MSST.2007.4367962}, keywords = {performance modeling, optimal checkpoint interval, I/O performance, fault-tolerance, checkpointing, LWFS, pario-bib}, abstract = {The next generation of capability-class, massively parallel processing (MPP) systems is expected to have hundreds of thousands of processors. For application-driven, periodic checkpoint operations, the state-of-the-art does not provide a solution that scales to next-generation systems. We demonstrate this by using mathematical modeling to compute a lower bound of the impact of these approaches on the performance of applications executed on three massive-scale, in-production, DOE systems and a theoretical petaflop system. We also adapt the model to investigate a proposed optimization that makes use of ``lightweight'' storage architectures and overlay networks to overcome the storage system bottleneck. Our results indicate that (1) as we approach the scale of next-generation systems, traditional checkpoint/restart approaches will increasingly impact application performance, accounting for over 50\% of total application execution time; (2) although our alternative approach improves performance, it has limitations of its own; and (3) there is a critical need for new approaches to checkpoint/restart that allow continuous computing with minimal impact on the scalability of applications.} }