Saved in:
Bibliographic Details
Main Authors: Wang, Xin, Rizzini, Pietro Lodi, Medya, Sourav, Lan, Zhiling
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.11111
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915618120269824
author Wang, Xin
Rizzini, Pietro Lodi
Medya, Sourav
Lan, Zhiling
author_facet Wang, Xin
Rizzini, Pietro Lodi
Medya, Sourav
Lan, Zhiling
contents The Dragonfly network, with its high-radix and low-diameter structure, is a leading interconnect in high-performance computing. A major challenge is workload interference on shared network links. Parallel discrete event simulation (PDES) is commonly used to analyze workload interference. However, high-fidelity PDES is computationally expensive, making it impractical for large-scale or real-time scenarios. Hybrid simulation that incorporates data-driven surrogate models offers a promising alternative, especially for forecasting application runtime, a task complicated by the dynamic behavior of network traffic. We present \ourmodel, a surrogate model that combines graph neural networks (GNNs) and large language models (LLMs) to capture both spatial and temporal patterns from port level router data. \ourmodel outperforms existing statistical and machine learning baselines, enabling accurate runtime prediction and supporting efficient hybrid simulation of Dragonfly networks.
format Preprint
id arxiv_https___arxiv_org_abs_2511_11111
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SMART: A Surrogate Model for Predicting Application Runtime in Dragonfly Systems
Wang, Xin
Rizzini, Pietro Lodi
Medya, Sourav
Lan, Zhiling
Machine Learning
Distributed, Parallel, and Cluster Computing
The Dragonfly network, with its high-radix and low-diameter structure, is a leading interconnect in high-performance computing. A major challenge is workload interference on shared network links. Parallel discrete event simulation (PDES) is commonly used to analyze workload interference. However, high-fidelity PDES is computationally expensive, making it impractical for large-scale or real-time scenarios. Hybrid simulation that incorporates data-driven surrogate models offers a promising alternative, especially for forecasting application runtime, a task complicated by the dynamic behavior of network traffic. We present \ourmodel, a surrogate model that combines graph neural networks (GNNs) and large language models (LLMs) to capture both spatial and temporal patterns from port level router data. \ourmodel outperforms existing statistical and machine learning baselines, enabling accurate runtime prediction and supporting efficient hybrid simulation of Dragonfly networks.
title SMART: A Surrogate Model for Predicting Application Runtime in Dragonfly Systems
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2511.11111