Fine-Tuned Sentence-BERT for HPC Job Outcome Prediction via Textual Feature Embedding
Thanh Hoang Le Hai, Huy Nguyen Tuan, Bao Tran Dang, Bao Vo Thuong, Nam Thoai
HPCC / HCMUT
Research at HPCC
Research pillars for HPC systems, data streaming platforms, and emerging quantum and agent-native technologies.
Research Pillars
Each pillar has a dedicated route for deeper content, selected projects, and related publications.
High Performance Computing Systems for AI
Building scalable computing systems and research software for AI workloads, scientific applications, and data-intensive experiments.
Data Streaming Platform for IoT and Big Data
Developing streaming data platforms that connect IoT systems, real-time analytics, and domain applications at scale.
Emerging Technologies: Quantum and Agent-Native Systems
Exploring emerging computing paradigms where quantum methods and agent-native systems extend the future of HPC, AI, and scientific infrastructure.
Selected Projects by Pillar
Ongoing
Enhancing HPC Systems for AI Applications
A project focused on improving high performance computing systems to better support AI workloads, performance optimization, and advanced computing services.
CompletedStreaming Data Platform for Urban and Rural Applications
A data streaming platform for real-time urban and rural applications, connecting IoT data collection, data processing, and analytics workflows.
CompletedAI- and GIS-based Flood Warning System for Ho Chi Minh City
An applied AI and GIS platform for flood warning in large cities, with Ho Chi Minh City as the target deployment context.
CompletedSmart Village Model from Farmer Associations in Dong Thap
A public-sector innovation project exploring smart village models, data platforms, and digital transformation in agricultural communities.
CompletedHPC Infrastructure Design for Ho Chi Minh City
A project proposing high performance computing infrastructure for Ho Chi Minh City, connecting HPC capacity planning with AI, big data, and scientific computing needs.
CompletedMulti-objective Resource Allocation for HPC Cloud
A research project on multi-objective resource allocation for scientific applications on cloud-enabled high performance computing systems.
Selected Publications
Toward Sustainable and Cost-Efficient HPC Systems: A Deep Reinforcement Learning Job Scheduling Approach
Thanh Hoang Le Hai, Nhan Nguyen Phuc, Mai Nguyen Tran Phuong, Minh Bui Ngoc, Nam Thoai
JREP - A Job Runtime Ensemble Predictor for Improving Scheduling Performance on High Performance Computing Systems
Thanh Hoang Le Hai, Thin Nguyen Manh, Hung Nguyen Quang, Nam Thoai
Deviation Backfilling: A Robust Backfilling Scheme for Improving the Efficiency of Job Scheduling on High Performance Computing Systems
Thanh Hoang Le Hai, Khang Nguyen Duy, Thin Nguyen Manh, Danh Mai Hoang, Nam Thoai