Jing Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jing Chen has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Ge Yu
Northeastern University, Shenyang, China
2 shared publications - Qiange Wang
Northeastern University, Shenyang, China
2 shared publications - Biswajit Das
Department of Mathematics, Chhaygaon College, Chhaygaon, Assam, India
1 shared publication - Chuanwen Li
Northeastern University
1 shared publication - Gaowu Qin
Key Laboratory for Anisotropy and Texture of Materials (Ministry of Education) School of Materials Science and Engineering Northeastern University Shenyang People's Republic of China
1 shared publication - Shufeng Gong
School of Computer Science and Engineering, Northeastern University, Shenyang, China
1 shared publication
Latest Publications
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 36 (5) , 961-976
Towards Communication-Efficient Out-of-Core Graph Processing on the GPU
The key performance bottleneck of large-scale graph processing on memory-limited GPUs is the host-GPU graph data transfer. Existing GPU-accelerated graph processing frameworks address this issue by managing the active subgraph transfer at runtime. Some frameworks adopt explicit transfer management approaches based on explicit memory copy with filter or compaction. In contrast, others adopt implicit transfer management approaches based on on-demand accesses with the zero-copy mechanism or unified virtual memory. Having made intensive analysis, we find that as the active vertices evolve, the performance of the two approaches varies in different workloads. Due to heavy redundant data transfers, high CPU compaction overhead, or low bandwidth utilization, adopting a single approach often results in suboptimal performance. Moreover, these methods lack effective cache management methods to address the irregular and sparse memory access pattern of graph processing. In this work, we propose a hybrid transfer management approach that takes the merits of both two transfer approaches at runtime. Moreover, we present an efficient vertex-centric graph caching framework that minimizes CPU-GPU communication by caching frequently accessed graph data at runtime. Based on these techniques, we present HytGraph, a GPU-accelerated graph processing framework, which is empowered by a set of effective task-scheduling optimizations to improve performance. Experiments on real-world and synthetic graphs show that HytGraph achieves average speedups of 2.5×, 5.0×, and 2.0× compared to the state-of-the-art GPU-accelerated graph processing systems, Grus, Subway, and EMOGI, respectively.
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 36 (5) , 961-976
Towards Communication-Efficient Out-of-Core Graph Processing on the GPU
The key performance bottleneck of large-scale graph processing on memory-limited GPUs is the host-GPU graph data transfer. Existing GPU-accelerated graph processing frameworks address this issue by managing the active subgraph transfer at runtime. Some frameworks adopt explicit transfer management approaches based on explicit memory copy with filter or compaction. In contrast, others adopt implicit transfer management approaches based on on-demand accesses with the zero-copy mechanism or unified virtual memory. Having made intensive analysis, we find that as the active vertices evolve, the performance of the two approaches varies in different workloads. Due to heavy redundant data transfers, high CPU compaction overhead, or low bandwidth utilization, adopting a single approach often results in suboptimal performance. Moreover, these methods lack effective cache management methods to address the irregular and sparse memory access pattern of graph processing. In this work, we propose a hybrid transfer management approach that takes the merits of both two transfer approaches at runtime. Moreover, we present an efficient vertex-centric graph caching framework that minimizes CPU-GPU communication by caching frequently accessed graph data at runtime. Based on these techniques, we present HytGraph, a GPU-accelerated graph processing framework, which is empowered by a set of effective task-scheduling optimizations to improve performance. Experiments on real-world and synthetic graphs show that HytGraph achieves average speedups of 2.5×, 5.0×, and 2.0× compared to the state-of-the-art GPU-accelerated graph processing systems, Grus, Subway, and EMOGI, respectively.
- 26 October 2021
- Published by Springer Nature in World Wide Web
- Vol. 25 (3) , 1377-1402
Unified-memory-based hybrid processing for partition-oriented subgraph matching on GPU
- 26 October 2021
- Published by Springer Nature in World Wide Web
- Vol. 25 (3) , 1377-1402
Unified-memory-based hybrid processing for partition-oriented subgraph matching on GPU
- 1 May 2014
- Published by Royal Society of Chemistry (RSC) in RSC Advances
- Vol. 4 (43) , 22660-22668
Fabrication of tunable Au SERS nanostructures by a versatile technique and application in detecting sodium cyclamate
In this work, three typical morphologies of SERS substrates with the compositions of Au nanoparticles, short Au nanowires (NWs) with small Au nanoparticles filled in the nanogaps and long Au NWs intertwined into a self-supporting film, were fabricated by a versatile nanocluster fabrication technique. This physical fabrication technique is easy to implement and cost-effective, and the fabricated SERS substrates have a large-area, good homogeneity and long-term stability. More importantly, the SERS substrates prepared by this physical method will not incorporate reductants and surfactant chemicals, and thus can yield clean surfaces, which are suitable for analytes with a weak affinity to the substrate surface. In this study, the fabricated substrate with the highest enhancement factor (EF) successfully acquired the SERS spectrum of sodium cyclamate, a restricted food additive, with a good signal–noise-ratio. Finally, a set of piece-wise linear equations were produced according to the correlation between SERS intensity and sodium cyclamate concentration with a detection limit of 1.6 × 10−9 M.
- 1 May 2014
- Published by Royal Society of Chemistry (RSC) in RSC Advances
- Vol. 4 (43) , 22660-22668
Fabrication of tunable Au SERS nanostructures by a versatile technique and application in detecting sodium cyclamate
In this work, three typical morphologies of SERS substrates with the compositions of Au nanoparticles, short Au nanowires (NWs) with small Au nanoparticles filled in the nanogaps and long Au NWs intertwined into a self-supporting film, were fabricated by a versatile nanocluster fabrication technique. This physical fabrication technique is easy to implement and cost-effective, and the fabricated SERS substrates have a large-area, good homogeneity and long-term stability. More importantly, the SERS substrates prepared by this physical method will not incorporate reductants and surfactant chemicals, and thus can yield clean surfaces, which are suitable for analytes with a weak affinity to the substrate surface. In this study, the fabricated substrate with the highest enhancement factor (EF) successfully acquired the SERS spectrum of sodium cyclamate, a restricted food additive, with a good signal–noise-ratio. Finally, a set of piece-wise linear equations were produced according to the correlation between SERS intensity and sodium cyclamate concentration with a detection limit of 1.6 × 10−9 M.