arXiv subjects

Yash Kurkure

Publications and source records attributed to Yash Kurkure.

3 recordsLinked to original sources

arXiv2026

MARS: A Monte Carlo Tree Search-based Adaptive and Responsive Scheduler

Modern High Performance Computing systems depend on static heuristics and manual administration for job scheduling and reservation management. Deep Reinforcement Learning (DRL) has shown promising scheduling performance but requires historical training data and fixes the optimization goal at training time, forcing operators to retrain whenever priorities shift. We introduce MARS (Monte Carlo Tree Search-based Adaptive and Responsive Scheduler), a training-free HPC scheduler whose optimization goal is configurable through a reward function rather than baked into a learned model. MARS uses a lightweight discrete-event simulator to explore the future consequences of scheduling decisions within a strict time budget, adapting to the configured reward at each scheduling cycle. We evaluate MARS on year-long production workloads from two systems at Argonne Leadership Computing Facility -- 4,360-node Theta and 560-node Polaris---under two reward functions: wait-time minimization (MARS-CW) and utilization maximization (MARS-CU). Unlike DRL and heuristics, which only react to the current queue or wait for backfill to find holes, MARS exploits look-ahead to proactively drain the system and plan around future reservations, packing the system to avoid the fragmentation and utilization drop that typically precede reservation windows. MARS-CW reduces tail wait time by 64% on Theta and 43% on Polaris over the production WFP heuristic, while MARS-CU recovers utilization in the 48 hours leading into maintenance, demonstrating that MARS can target either objective via reward reconfiguration.

cs.DC↗

arXiv2025

A Real-Time Digital Twin for Adaptive Scheduling

High-performance computing (HPC) workloads are becoming increasingly diverse, exhibiting wide variability in job characteristics, yet cluster scheduling has long relied on static, heuristic-based policies. In this work we present SchedTwin, a real-time digital twin designed to adaptively guide scheduling decisions using predictive simulation. SchedTwin periodically ingests runtime events from the physical scheduler, performs rapid what-if evaluations of multiple policies using a high-fidelity discrete-event simulator, and dynamically selects the one satisfying the administrator configured optimization goal. We implement SchedTwin as an open-source software and integrate it with the production PBS scheduler. Preliminary results show that SchedTwin consistently outperforms widely used static scheduling policies, while maintaining low overhead (a few seconds per scheduling cycle). These results demonstrate that real-time digital twins offer a practical and effective path toward adaptive HPC scheduling.

cs.DC↗

arXiv2025

Faster Algorithms for Fair Max-Min Diversification in $\mathbb{R}^d$

The task of extracting a diverse subset from a dataset, often referred to as maximum diversification, plays a pivotal role in various real-world applications that have far-reaching consequences. In this work, we delve into the realm of fairness-aware data subset selection, specifically focusing on the problem of selecting a diverse set of size $k$ from a large collection of $n$ data points (FairDiv). The FairDiv problem is well-studied in the data management and theory community. In this work, we develop the first constant approximation algorithm for FairDiv that runs in near-linear time using only linear space. In contrast, all previously known constant approximation algorithms run in super-linear time (with respect to $n$ or $k$) and use super-linear space. Our approach achieves this efficiency by employing a novel combination of the Multiplicative Weight Update method and advanced geometric data structures to implicitly and approximately solve a linear program. Furthermore, we improve the efficiency of our techniques by constructing a coreset. Using our coreset, we also propose the first efficient streaming algorithm for the FairDiv problem whose efficiency does not depend on the distribution of data points. Empirical evaluation on million-sized datasets demonstrates that our algorithm achieves the best diversity within a minute. All prior techniques are either highly inefficient or do not generate a good solution.

cs.DB↗