Requests sizes / O HPC applications on a supercomputer

• Gessica Mendonça Azevedo UFRGS
• Jean Luca Bez UFRGS
• Pablo Pavan UFRGS
• Francieli Zanon Boito INRIA
• Philippe Olivier Alexandre Navaux UFRGS

DOI: https://doi.org/10.5753/eradrs.2020.10761

Abstract

This study seeks to identify incoming requests sizes and more common output used for HPC applications in large scale environments. For this, we use data for an entire year of characterization with Darshan tool in Intrepid supercomputer Blue Gene / P. By identifying the different patterns of access and the size of requests, we contribute to that new optimization techniques can be tested and evaluated considering the sizes of similar requests to that found in these environments.

Keywords: High Performance Storage Systems

References

Boito, F. Z., Inacio, E. C., Bez, J., Navaux, P. O. A., Dantas, M. A. R., and Denneulin, Y. (2018). A Checkpoint of Research on Parallel I/O for High-Performance Computing. In 2018 ACM Computing Surveys (ACM).

Carns., P. (2013). ALCF I/O Data Repository. Technical report, Argonne Leadership Computing Facility.

Carns, P., Harms, K., Allcock, W., Bacon, C., Lang, S., Latham, R., and Ross, R. (2011). Understanding and improving computational science storage access through continuous characterization. In 2011 IEEE 27th Symposium on Mass Storage Systems and Technologies, page 7(3):8:1–8:26.

Carns, P., Latham, R., Ross, R., K. Iskra, S. L., and Riley., K. (2009). 24/7 Characterization of petascale I/O workloads. In 2009 IEEE International Conference on Cluster Computing and Workshops, pages 1–10.

Pavan, P. J., Bez, J., Serpa, M. S., Boito, F. Z., and Navaux, P. O. A. (2019). An Unsupervised Learning Approach for I/O Behavior Characterization. In 2019 31st International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD).