@inproceedings{wang2025cmpi,title={{cMPI}: Using {CXL} Memory Sharing for {MPI} One-Sided and Two-Sided Inter-Node Communications},author={Wang, Xi and Ma, Bin and Kim, Jongryool and Koh, Byungil and Kim, Hoshik and Li, Dong},booktitle={Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},series={SC '25},year={2025},pages={2216--2232},publisher={Association for Computing Machinery},doi={10.1145/3712285.3759816},url={https://doi.org/10.1145/3712285.3759816},}
IPDPS 2025
Performance Characterization of CXL Memory and Its Use Cases
Xi Wang, Jie Liu, Jianbo Wu, Shuangyan Yang, Jie Ren, Bhanu Shankar, and 1 more author
In 2025 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2025
@inproceedings{wang2025characterization,title={Performance Characterization of {CXL} Memory and Its Use Cases},author={Wang, Xi and Liu, Jie and Wu, Jianbo and Yang, Shuangyan and Ren, Jie and Shankar, Bhanu and Li, Dong},booktitle={2025 IEEE International Parallel and Distributed Processing Symposium (IPDPS)},year={2025},pages={1048--1061},publisher={IEEE},doi={10.1109/IPDPS64566.2025.00097},url={https://doi.org/10.1109/IPDPS64566.2025.00097},}
arXiv
HybridGen: Efficient LLM Generative Inference via CPU-GPU Hybrid Computing
Mao Lin, Xi Wang, Guilherme Cox, Dong Li, and Hyeran Jeon
@misc{lin2026hybridgen,title={{HybridGen}: Efficient {LLM} Generative Inference via {CPU-GPU} Hybrid Computing},author={Lin, Mao and Wang, Xi and Cox, Guilherme and Li, Dong and Jeon, Hyeran},year={2026},month=apr,eprint={2604.18529},archiveprefix={arXiv},primaryclass={cs.PF},doi={10.48550/arXiv.2604.18529},url={https://arxiv.org/abs/2604.18529},}
SC 2026
OCEAN: Open-Source CXL Emulation for Hyperscale Architecture and Networking
Yiwei Yang, Mujahid Al Rafi, Jesun Sahariar Firoz, Zhen Peng, Xi Wang, Kenza Bouasker, and 8 more authors
In International Conference for High Performance Computing, Networking, Storage and Analysis (SC), 2026
@inproceedings{yang2026ocean,title={{OCEAN}: Open-Source {CXL} Emulation for Hyperscale Architecture and Networking},author={Yang, Yiwei and Rafi, Mujahid Al and Firoz, Jesun Sahariar and Peng, Zhen and Wang, Xi and Bouasker, Kenza and Ghosh, Sayan and Li, Jie and Wong, Daniel and Li, Dong and Jeon, Hyeran and Barker, Kevin and Tallent, Nathan and Guo, Luanzheng},booktitle={International Conference for High Performance Computing, Networking, Storage and Analysis (SC)},year={2026},url={https://sc26.conference-program.com/presentation/?id=pap148&sess=sess232},}
SC 2025
mLR: Scalable Laminography Reconstruction based on Memoization
Bin Ma, Viktor Nikitin, Xi Wang, Tekin Bicer, and Dong Li
In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, 2025
@inproceedings{ma2025mlr,title={{mLR}: Scalable Laminography Reconstruction based on Memoization},author={Ma, Bin and Nikitin, Viktor and Wang, Xi and Bicer, Tekin and Li, Dong},booktitle={Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},series={SC '25},year={2025},pages={265--280},publisher={Association for Computing Machinery},doi={10.1145/3712285.3759805},url={https://doi.org/10.1145/3712285.3759805},}
arXiv
TierBPF: Page Migration Admission Control for Tiered Memory via eBPF
Xi Wang, Tal Zussman, Bin Ma, Asaf Cidon, and Dong Li
@misc{wang2026tierbpf,title={{TierBPF}: Page Migration Admission Control for Tiered Memory via {eBPF}},author={Wang, Xi and Zussman, Tal and Ma, Bin and Cidon, Asaf and Li, Dong},year={2026},month=sep,eprint={2604.12300},archiveprefix={arXiv},primaryclass={cs.OS},doi={10.48550/arXiv.2604.12300},url={https://arxiv.org/abs/2604.12300},}
arXiv
Hybrid Adaptive Tuning for Tiered Memory Systems
Xi Wang, Jie Liu, Shuangyan Yang, Jongryool Kim, Pengfei Su, and Dong Li
@misc{wang2026adaptive,title={Hybrid Adaptive Tuning for Tiered Memory Systems},author={Wang, Xi and Liu, Jie and Yang, Shuangyan and Kim, Jongryool and Su, Pengfei and Li, Dong},year={2026},month=apr,eprint={2604.12165},archiveprefix={arXiv},primaryclass={cs.OS},doi={10.48550/arXiv.2604.12165},url={https://arxiv.org/abs/2604.12165},}