Abstract
Performance of file systems shift during their life cycles. Evaluating this performance change over time is not trivial. Complexity arises in the interplay between external (i.e. application I/O workloads) and internal (i.e. the filesystem state) factors. Many benchmarks can test how a filesystem performs at the current snapshot state, but to observe the change over time necessitates that the filesystem state mutate (age) between benchmark runs. For a large-scale HPC parallel filesystem, the sheer scale and amount of interacting components during I/O operations magnify these challenges.
There have been several approaches that address different aspects of filesystem aging, from creating statistically realistic filesystem images to file age distributions. The common drawbacks are the scale to be evaluated and the time needed to converge; none of the methods in literature targeted network or parallel file systems. Also, none were evaluated with a filesystem image over 300 GiB, most under 50 GiB, yet almost all took between a half hour to 7 h to converge. For a large-scale parallel file system, these methods are impractical as far as time and resources needed (a typical large PFS is in the PB range). Additionally, HPC filesystem I/O workloads are drastically different from local system workloads used in earlier studies.
This paper presents the design, implementation and evaluation of LCIO synthetic filesystem aging benchmark, which aims to address the question of “how will the filesystem perform at different stages of its life cycle?”. As such, being able to answer that question as realistically as feasible in a reasonable time is where LCIO contributes.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Agrawal, N., Arpaci-Dusseau, A.C., Arpaci-Dusseau, R.H.: Generating realistic impressions for file-system benchmarking. Trans. Storage 5(4), 16:1–16:30 (2009). https://doi.org/10.1145/1629080.1629086
Axboe, J.: FIO: flexible I/O tester. https://github.com/axboe/fio
Conway, A., et al.: File systems fated for senescence? Nonsense, says science!. In: 15th USENIX Conference on File and Storage Technologies (FAST 2017), pp. 45–58. USENIX Association, Santa Clara (2017). https://www.usenix.org/conference/fast17/technical-sessions/presentation/conway
IBM: IBM spectrum scale (2018). https://en.wikipedia.org/wiki/IBM_Spectrum_Scale
Kadekodi, S., Nagarajan, V., Ganger, G.R.: Geriatrix: aging what you see and what you don’t see. A file system aging approach for modern storage systems. In: 2018 USENIX Annual Technical Conference (USENIX ATC 2018), pp. 691–704. USENIX Association, Boston (2018). https://www.usenix.org/conference/atc18/presentation/kadekodi
LLNL: IOR HPC benchmark (2017). https://www.nersc.gov/users/computational-systems/cori/nersc-8-procurement/trinity-nersc-8-rfp/nersc-8-trinity-benchmarks/ior/
(NERSC), N.E.R.S.C.C.: MDtest (2013). https://www.nersc.gov/users/computational-systems/cori/nersc-8-procurement/trinity-nersc-8-rfp/nersc-8-trinity-benchmarks/mdtest
OLCF, O.R.L.C.F.: SPIDER storage system (2018). https://www.olcf.ornl.gov/olcf-resources/data-visualization-resources/spider/
OpenSFS: Lustre (2018). http://lustre.org/documentation/
Smith, K.A., Seltzer, M.I.: File system aging - increasing the relevance of file system benchmarks. SIGMETRICS Perform. Eval. Rev. 25(1), 203–213 (1997). https://doi.org/10.1145/258623.258689
Traeger, A., Zadok, E., Joukov, N., Wright, C.P.: A nine year study of file system and storage benchmarking. Trans. Storage 4(2), 5:1–5:56 (2008). https://doi.org/10.1145/1367829.1367831
Vazhkudai, S.S., et al.: The design, deployment, and evaluation of the coral pre-exascale systems. In: Proceedings of the International Conference for High Performance Computing, Networking, Storage, and Analysis, SC 2018, pp. 52:1–52:12. IEEE Press, Piscataway (2018). http://dl.acm.org/citation.cfm?id=3291656.3291726
Wang, F., Sim, H., Harr, C., Oral, S.: Diving into petascale production file systems through large scale profiling and analysis. In: Proceedings of the 2nd Joint International Workshop on Parallel Data Storage & Data Intensive Scalable Computing Systems, PDSW-DISCS 2017, pp. 37–42. ACM, New York (2017). https://doi.org/10.1145/3149393.3149399
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Bachstein, M., Wang, F., Oral, S. (2020). LCIO: Large Scale Filesystem Aging. In: Gao, W., Zhan, J., Fox, G., Lu, X., Stanzione, D. (eds) Benchmarking, Measuring, and Optimizing. Bench 2019. Lecture Notes in Computer Science(), vol 12093. Springer, Cham. https://doi.org/10.1007/978-3-030-49556-5_24
Download citation
DOI: https://doi.org/10.1007/978-3-030-49556-5_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-49555-8
Online ISBN: 978-3-030-49556-5
eBook Packages: Computer ScienceComputer Science (R0)