arXiv subjects

Hanlong Liao

Publications and source records attributed to Hanlong Liao.

4 recordsLinked to original sources

arXiv2025

GPUnion: Autonomous GPU Sharing on Campus

A pronounced imbalance in GPU resources exists on campus, where some laboratories own underutilized servers while others lack the compute needed for AI research. GPU sharing can alleviate this disparity, while existing platforms typically rely on centralized oversight and persistent allocation models, conflicting with the voluntary and autonomous nature of academic resource ownership. We present GPUnion, a campus-scale GPU sharing platform enabling voluntary participation while preserving full provider autonomy. GPUnion incorporates three core mechanisms: i) container-based task dispatching and execution, ii) resource provider-first architecture, and iii) resilient execution featuring automatic check-pointing and migration. Case studies across multiple campus scenarios demonstrate 30% more GPU utilization improvement, 40% increase in interactive sessions, and 94% successful workload migration during provider departures.

cs.DC↗

arXiv2025

AIMeter: Measuring, Analyzing, and Visualizing Energy and Carbon Footprint of AI Workloads

The rapid advancement of AI, particularly large language models (LLMs), has raised significant concerns about the energy use and carbon emissions associated with model training and inference. However, existing tools for measuring and reporting such impacts are often fragmented, lacking systematic metric integration and offering limited support for correlation analysis among them. This paper presents AIMeter, a comprehensive software toolkit for the measurement, analysis, and visualization of energy use, power draw, hardware performance, and carbon emissions across AI workloads. By seamlessly integrating with existing AI frameworks, AIMeter offers standardized reports and exports fine-grained time-series data to support benchmarking and reproducibility in a lightweight manner. It further enables in-depth correlation analysis between hardware metrics and model performance and thus facilitates bottleneck identification and performance enhancement. By addressing critical limitations in existing tools, AIMeter encourages the research community to weigh environmental impact alongside raw performance of AI workloads and advances the shift toward more sustainable "Green AI" practices. The code is available at https://github.com/SusCom-Lab/AIMeter.

cs.DC↗

arXiv2025

BandPilot: Toward Performance- and Contention-Aware GPU Dispatching in AI Clusters

Modern multi-tenant AI clusters are increasingly communication-bound, driven by high-volume and multi-round GPU-to-GPU collective communication. Consequently, the GPU dispatcher's choice of a physical GPU subset for each tenant largely determines the job's effective collective bandwidth and thus its performance ceiling. Existing dispatchers predominantly rely on static, topology-aware heuristics that prioritize GPU resource compactness, assuming that minimizing physical distance maximizes communication bandwidth. However, we reveal that this assumption often fails due to complex system-level bottlenecks, such as non-linear NIC saturation and inter-node link heterogeneity. This paper presents BandPilot, a performance- and contention-aware GPU dispatching primitive that optimizes effective collective bandwidth for multi-tenant AI clusters. Specifically, BandPilot learns a data-efficient bandwidth model from sparse NCCL measurements via a hierarchical design. Guided by the model, BandPilot uses an equilibrium-driven heuristic as a fast front end, and invokes a pruned elimination search when a controller predicts that further refinement is worthwhile. To account for multi-tenant interference, BandPilot virtually merges a candidate allocation with co-located cross-host jobs to conservatively estimate shared bottleneck capacity and predict contention-degraded bandwidth. Across a 32-GPU H100 cluster and heterogeneous simulations, BandPilot achieves 90-97% bandwidth efficiency relative to the best-found reference, improving average efficiency by 20-30% over topology-compactness heuristics.

cs.DC↗

arXiv2020

Quantifying Low-Battery Anxiety of Mobile Users and Its Impacts on Video Watching Behavior

People nowadays are increasingly dependent on mobile phones for daily communication, study, and business. Along with this it incurs the low-battery anxiety (LBA). Although having been unveiled for a while, LBA has not been thoroughly investigated yet. Without a better understanding of LBA, it would be difficult to precisely validate energy saving and management techniques in terms of alleviating LBA and enhancing Quality of Experience (QoE) of mobile users. To fill the gap, we conduct an investigation over 2000+ mobile users, look into their feelings and reactions towards LBA, and quantify their anxiety degree during the draining of battery power. As a case study, we also investigate the impact of LBA on user's behavior at video watching, and with the massive collected answers we are able to quantify user's abandoning likelihood of attractive videos versus the battery status of mobile phone. The empirical findings and quantitative models obtained in this work not only disclose the characteristics of LBA among modern mobile users, but also provide valuable references for the design, evaluation, and improvement of QoE-aware mobile applications and services.

cs.HC↗