Performance-Aware Scheduling and Resource Allocation for Large-Scale Structural Simulations
DOI:
https://doi.org/10.62051/np19jp25Keywords:
Structural simulation; Incremental dynamic analysis; OpenSees; Job scheduling; Resource allocation; Speculative execution; High-throughput computing.Abstract
Seismic performance assessment requires thousands of nonlinear response-history analyses, typically organized as incremental dynamic analysis (IDA) campaigns that share clusters with other users. These campaigns differ from generic batch workloads: every ground motion induces a chain of analyses whose next intensity level is needed only if the current one does not collapse, runtimes are governed by record length, model cost and intensity, and memory requests are sized for the worst case. We measured 810 real OpenSees analyses of four archetype frames under the 22 FEMA P695 far-field records and show that chain parallelism decays sharply as records collapse, that history-based runtime predictors are uncorrelated with actual runtimes, and that static memory requests exceed footprints by a median factor of 2.28. We propose PASS, a cross-task scheduler that combines an intensity-conditioned runtime and memory model, an online collapse-hazard estimator yielding the survival-weighted remaining work of every chain in constant time, a bound-guided critical-chain priority with aged shortest-job-first ordering, probability-gated and preemptible speculative execution of future intensity levels on idle cores, and memory right-sizing with out-of-memory feedback. In trace-driven experiments against FIFO, EASY backfilling, shortest-job-first with history-based and exact runtimes, static partitioning, Tetris and critical-path-first scheduling, PASS shortens the makespan of IDA batches by 33.9% and reduces the mean waiting time of online campaigns by 89.6% relative to the best baseline, raises core utilization from at most 57.8% to 74.6–79.5%, and requires about 142µs per scheduling decision. On 1024-core clusters, where every baseline is bounded by the longest chain, PASS reduces the makespan by 56%. Ablation, sensitivity and robustness experiments quantify the contribution of each component and show that the gains persist under runtime noise, heterogeneous nodes and bursty arrivals.
Downloads
References
[1] Federal Emergency Management Agency. (2009). Quantification of building seismic performance factors (FEMA P695). Applied Technology Council.
[2] Vamvatsikos, D., & Cornell, C. A. (2002). Incremental dynamic analysis. Earthquake Engineering & Structural Dynamics, 31(3), 491–514. DOI: https://doi.org/10.1002/eqe.141
[3] Vamvatsikos, D., & Cornell, C. A. (2004). Applied incremental dynamic analysis. Earthquake Spectra, 20(2), 523–553. DOI: https://doi.org/10.1193/1.1737737
[4] McKenna, F., Scott, M. H., & Fenves, G. L. (2010). Nonlinear finite-element analysis software architecture using object composition. Journal of Computing in Civil Engineering, 24(1), 95–107. DOI: https://doi.org/10.1061/(ASCE)CP.1943-5487.0000002
[5] Zhu, M., McKenna, F., & Scott, M. H. (2018). OpenSeesPy: Python library for the OpenSees finite element framework. SoftwareX, 7, 6–11. DOI: https://doi.org/10.1016/j.softx.2017.10.009
[6] McCallen, D., et al. (2021). EQSIM—a multidisciplinary framework for fault-to-structure earthquake simulations on exascale computers part I: Computational models and workflow. Earthquake Spectra, 37(2), 707–735. https://doi.org/10.1177/8755293020970982 DOI: https://doi.org/10.1177/8755293020970982
[7] Zsarnóczay, A., et al. (2025). An open-source simulation platform to support and foster research collaboration in natural hazards engineering. Frontiers in Built Environment, 11, 1590479. https://doi.org/10.3389/fbuil.2025.1590479 DOI: https://doi.org/10.3389/fbuil.2025.1590479
[8] Lifka, D. A. (1995). The ANL/IBM SP scheduling system. In Job scheduling strategies for parallel processing (Vol. 949, pp. 295–303). Springer. DOI: https://doi.org/10.1007/3-540-60153-8_35
[9] Yoo, A. B., Jette, M. A., & Grondona, M. (2003). SLURM: Simple Linux utility for resource management. In Job scheduling strategies for parallel processing (Vol. 2862, pp. 44–60). Springer. DOI: https://doi.org/10.1007/10968987_3
[10] Vamvatsikos, D. (2011). Performing incremental dynamic analysis in parallel. Computers & Structures, 89(1–2), 170–180. DOI: https://doi.org/10.1016/j.compstruc.2010.08.014
[11] Tsafrir, D., Etsion, Y., & Feitelson, D. G. (2007). Backfilling using system-generated predictions rather than user runtime estimates. IEEE Transactions on Parallel and Distributed Systems, 18(6), 789–803. DOI: https://doi.org/10.1109/TPDS.2007.70606
[12] Li, J., Michelogiannakis, G., Cook, B., Cooray, D., & Chen, Y. (2023). Analyzing resource utilization in an HPC system: A case study of NERSC’s Perlmutter. In High performance computing (pp. 242–257). Springer. https://doi.org/10.1007/978-3-031-32041-5_16 DOI: https://doi.org/10.1007/978-3-031-32041-5_16
[13] Tovar, B., et al. (2018). A job sizing strategy for high-throughput scientific workflows. IEEE Transactions on Parallel and Distributed Systems, 29(2), 240–253. DOI: https://doi.org/10.1109/TPDS.2017.2762310
[14] Bader, J., et al. (2024). Sizey: Memory-efficient execution of scientific workflow tasks. In 2024 IEEE International Conference on Cluster Computing (CLUSTER). https://doi.org/10.1109/CLUSTER59578.2024.00039 DOI: https://doi.org/10.1109/CLUSTER59578.2024.00039
[15] Lehmann, F., et al. (2024). Ponder: Online prediction of task memory requirements for scientific workflows. In 20th IEEE International Conference on e-Science. https://doi.org/10.1109/e-Science62913.2024.10678682 DOI: https://doi.org/10.1109/e-Science62913.2024.10678682
[16] Wang, Y., Zou, X., Lu, Z., Wei, K., & Rhee, M. (2026). PRESAGE: Uncertainty-aware predictive elastic scheduling for cloud analytics and GenAI inference. IEEE Access, 14, 146019–146040. https://doi.org/10.1109/ACCESS.2026.3734498 DOI: https://doi.org/10.1109/ACCESS.2026.3734498
[17] Grandl, R., Ananthanarayanan, G., Kandula, S., Rao, S., & Akella, A. (2014). Multi-resource packing for cluster schedulers. In Proceedings of the 2014 ACM SIGCOMM Conference (pp. 455–466). DOI: https://doi.org/10.1145/2619239.2626334
[18] Topcuoglu, H., Hariri, S., & Wu, M.-Y. (2002). Performance-effective and low-complexity task scheduling for heterogeneous computing. IEEE Transactions on Parallel and Distributed Systems, 13(3), 260–274. DOI: https://doi.org/10.1109/71.993206
[19] Ibarra, L. F., Medina, R. A., & Krawinkler, H. (2005). Hysteretic models that incorporate strength and stiffness deterioration. Earthquake Engineering & Structural Dynamics, 34(12), 1489–1511. DOI: https://doi.org/10.1002/eqe.495
[20] Lignos, D. G., & Krawinkler, H. (2011). Deterioration modeling of steel components in support of collapse prediction of steel moment frames under earthquake loading. Journal of Structural Engineering, 137(11), 1291–1302. DOI: https://doi.org/10.1061/(ASCE)ST.1943-541X.0000376
[21] Haselton, C. B., Liel, A. B., Deierlein, G. G., Dean, B. S., & Chou, J. H. (2011). Seismic collapse safety of reinforced concrete buildings. I: Assessment of ductile moment frames. Journal of Structural Engineering, 137(4), 481–491. DOI: https://doi.org/10.1061/(ASCE)ST.1943-541X.0000318
[22] Baker, J. W. (2015). Efficient analytical fragility function fitting using dynamic structural analysis. Earthquake Spectra, 31(1), 579–599. DOI: https://doi.org/10.1193/021113EQS025M
[23] Guan, X., & Burton, H. V. (2023). Collaborative filtering-based collapse fragility assessment. Earthquake Engineering & Structural Dynamics, 52(11), 3322–3344. https://doi.org/10.1002/eqe.3963 DOI: https://doi.org/10.1002/eqe.3963
[24] Arroyo, O., Feliciano, D., Novoa, D., & Valcárcel, J. (2024). Opseestools: A Python library to streamline OpenSeesPy workflows. SoftwareX, 27, 101832. DOI: https://doi.org/10.1016/j.softx.2024.101832
[25] Boëzennec, R., et al. (2024). Qualitatively analyzing optimization objectives in the design of HPC resource manager. ACM Transactions on Modeling and Performance Evaluation of Computing Systems, 9, 1–28. https://doi.org/10.1145/3701986 DOI: https://doi.org/10.1145/3701986
[26] Merzky, A., Turilli, M., Titov, M., Al-Saadi, A., & Jha, S. (2022). Design and performance characterization of RADICAL-Pilot on leadership-class platforms. IEEE Transactions on Parallel and Distributed Systems, 33(4). DOI: https://doi.org/10.1109/TPDS.2021.3105994
[27] Zaharia, M., Konwinski, A., Joseph, A. D., Katz, R., & Stoica, I. (2008). Improving MapReduce performance in heterogeneous environments. In Proceedings of the 8th USENIX Symposium on Operating Systems Design and Implementation (OSDI) (pp. 29–42).
[28] Zhang, Y., Zhang, T., Zhang, G., & Jacobsen, H.-A. (2023). Lifting the fog of uncertainties: Dynamic resource orchestration for the containerized cloud. In Proceedings of the 2023 ACM Symposium on Cloud Computing (SoCC) (pp. 48–64). https://doi.org/10.1145/3620678.3624646 DOI: https://doi.org/10.1145/3620678.3624646
[29] Graham, R. L. (1969). Bounds on multiprocessing timing anomalies. SIAM Journal on Applied Mathematics, 17(2), 416–429. DOI: https://doi.org/10.1137/0117039
[30] Kaplan, E. L., & Meier, P. (1958). Nonparametric estimation from incomplete observations. Journal of the American Statistical Association, 53(282), 457–481. DOI: https://doi.org/10.1080/01621459.1958.10501452
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Transactions on Environment, Energy and Earth Sciences
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
