Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
BOOSTEDSOSA introduces a dual-FPGA ML-assisted scheduling architecture that integrates a machine learning predictor to estimate job runtimes based solely on scheduler parameters available at submission time. This innovation significantly reduces the mean absolute error (MAE) in processing time estimates by up to 63.85% compared to traditional user runtime estimates and achieves a remarkable 17x speedup over existing software solutions. The approach addresses the high variance in scheduling performance in heterogeneous environments, thereby enhancing the competitiveness of stochastic online scheduling algorithms in high-performance computing systems.
A machine learning predictor in BOOSTEDSOSA slashes scheduling error by up to 71.88%, transforming how HPC systems handle job runtimes.
Heterogeneous scheduling in stochastic, online envi- ronments, such as high-performance computing (HPC) systems, presents a significant challenge. Stochastic Online Scheduling Accelerators (SOSAs) offer a promising solution, but their effectiveness is compromised by a reliance on runtime estimates provided by users. These estimates introduce substantial vari- ance into the scheduling process (mean MAE in hundreds of Core-Days), thereby weakening the competitiveness of Stochastic Online Scheduling algorithms as their competitive-ratio bound increases with runtime variability. To address this limitation, we introduce BOOSTEDSOSA, a dual-FPGA ML-assisted Scheduling architecture that integrates a Machine Learning predictor for expected processing times, with a novel temporal-aware training policy. The predictor estimates job runtimes using only scheduler parameters available at submission time, enabling its use in existing HPC systems. Using historical real-world HPC job data (from the Argonne Leadership Comput- ing Facility, MIT Supercloud and UIUC Blue Waters workload datasets), we show that the predictor reduces MAE by up to 63.85% compared to user runtime estimates, and the additive training policy reduces MAE by up to 71.88% compared to a static model. End-to-end, BOOSTEDSOSA achieves an average 17x speedup over an AVX-optimized software baseline and processes up to 1,711 jobs/seconds