Portfolio item number 1
Short description of portfolio item number 1
Short description of portfolio item number 1
Short description of portfolio item number 2
Published in IPDPS (CCF B), 2019
resource scheduling for long-running jobs.
Recommended citation: Heng Wu. (2019). "Aladdin: Optimized Maximum Flow Management for Shared Production Clusters." IPDPS. 696-707. https://github.com/syswu/syswu.github.io/blob/master/files/aladdin_ipdps19.pdf
Published in ICPP (CCF B), 2021
resource recommendation for big data jobs.
Recommended citation: Heng Wu. (2021). "Best VM Selection for Big Data Applications across Multiple Frameworks by Transfer Learning." ICPP. 85(1)-85(11). https://github.com/syswu/syswu.github.io/blob/master/files/vesta_icpp21.pdf
Published in ACM SoCC (CCF B), 2022
resource recommendation for deep learning training jobs.
Recommended citation: Yuewen WU, Heng Wu. (2022). "Serving Unseen Deep Learning Models with Near-Optimal Configurations: a Fast Adaptive Search Approach.." ACM SoCC. 1-16. https://github.com/syswu/syswu.github.io/blob/master/files/falcon_socc22.pdf
Published in DASSFA 23 (CCF B), 2023
task schedling for blockchain.
Recommended citation: Shi, J., Wu, H., Luo, D., Gao, H., Zhang, W. (2023). InstantChain: Enhancing Order-Execute Blockchain Systems for Latency-Sensitive Applications. In: Wang, X., et al. Database Systems for Advanced Applications. DASFAA 2023. Lecture Notes in Computer Science, vol 13943. Springer, Cham. https://doi.org/10.1007/978-3-031-30637-2_31 https://link.springer.com/chapter/10.1007/978-3-031-30637-2_31
Published in ICPP 23 (CCF B), 2023
task schedling for blockchain.
Recommended citation: Diaohan Luo, Yuewen Wu, Tian Yu, Tao Wang, Heng Wu, Wenbo Zhang. (2023). SPLIT: QoS-Aware DNN Inference on Shared GPU via Evenly-Sized Model Splitting. In: 52nd The International Conference on Parallel Processing. ICPP 2023. https://link.springer.com/chapter/10.1007/978-3-031-30637-2_31
Undergraduate course, Nanjing School, University of Chinese Academy of Sciences, 2022
Cloud computing has created a shift from the use of physical hardware and locally managed software-enabled platforms to that of virtualized cloud-hosted services. Cloud assembles large networks of virtual services, including hardware (CPU, storage, and network) and software resources (databases, message queuing systems, monitoring systems, and load-balancers).