DOE OSTI · 3377949
Benchmarking DAOS Filesystem on Aurora
Abstract
We benchmark the DAOS filesystem on Argonne's Aurora supercomputer (127 nodes, 4,064 targets) using fio, IOR, mdtest, and IO500 to characterize I/O and metadata performance across the DFS API and DFuse+POSIX. Single-client fio shows POSIX bandwidth saturating at 1–2 MiB I/O sizes, with write-heavy workloads outperforming reads. Multi-node IOR shows DFS bandwidth scaling well up to ~32 tasks/node, with write latency growing faster than read latency. An 8-node IO500 evaluation shows DFS achieving ~5x higher bandwidth and ~190x higher IOPS than POSIX. Results indicate DAOS is well-suited to read-heavy workloads like AI training data loading, given appropriately sized transfers and concurrency.
Keep this discovery
Explore connections, maps & timelines
George, Rebecca [College of William and Mary, Williamsburg, VA (United States); Thomas Jefferson National Accelerator Facility (TJNAF), Newport News, VA (United States)], Mei, Xinxin [Thomas Jefferson National Accelerator Facility (TJNAF), Newport News, VA (United States)], Ren, Jie [College of William and Mary, Williamsburg, VA (United States)]. 2026-03-27. Benchmarking DAOS Filesystem on Aurora. https://doi.org/10.2172/3377949. https://engineeringpapers.com/record/osti%3A3377949/benchmarking-daos-filesystem-on-aurora
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Engineering Papers by Indiciae: BibTeX · RIS.