dpKernels: Harvesting DPU Compute Resources for Data-path Efficiency in Cloud Data Processing
Summary: Introduces dpKernels, portable primitives that expose heterogeneous DPU accelerators and SoC cores to cloud data systems. dpManager optimizes and schedules constrained offloads across DPU generations/vendors, improving datapath throughput while reducing host CPU use. (summarized by gpt-5.6-luna on Aug 28 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Jiasheng Hu (University of Toronto)
- 2. Kaiwen Zheng (University of Toronto)
- 3. Anna Li (University of Toronto)
- 4. Sidharth Sankhe (University of Toronto)
- 5. Philip A. Bernstein (Microsoft)
- 6. Qizhen Zhang (University of Toronto)
BibTeX Citation
@article{hu_vldb26,
title = {{dpKernels: Harvesting DPU Compute Resources for Data-path Efficiency in Cloud Data Processing}},
author = {Hu, Jiasheng and Zheng, Kaiwen and Li, Anna and Sankhe, Sidharth and Bernstein, Philip A. and Zhang, Qizhen},
journal = {PVLDB},
series = {{VLDB} '26},
volume = {19},
number = {10},
pages = {2563--2576},
doi = {10.14778/3828612.3828615},
url = {https://doi.org/10.14778/3828612.3828615},
year = {2026}
}
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 26 of 26 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 4,095 | Distributed Deep Learning on Data Systems: A Comparative Analysis of Approaches | 2021 | VLDB |
| 2 | 2,600 | Robust Query Processing in Co-Processor-accelerated Databases | 2016 | SIGMOD |
| 3 | 3,773 | Hardware-conscious Query Processing in GPU-accelerated Analytical Engines | 2019 | CIDR |
| 4 | 9,464 | DFlush: DPU-Offloaded Flush for Disaggregated LSM-based Key-Value Stores | 2025 | SIGMOD |
| 5 | 9,662 | The End of Moore’s Law and the Rise of The Data Processor | 2021 | VLDB |
| 6 | 2,213 | Towards Cost-Optimal Query Processing in the Cloud | 2021 | VLDB |
| 7 | 8,758 | DDS: DPU-optimized Disaggregated Storage | 2024 | VLDB |
| 8 | 7,959 | Cloud-Native Database Systems and Unikernels: Reimagining OS Abstractions for Modern Hardware | 2024 | VLDB |
| 9 | 11,436 | Analyzing Near-Network Hardware Acceleration with Co-Processing on DPUs | 2025 | VLDB |
| 10 | 9,687 | DPDPU: Data Processing with DPUs | 2025 | CIDR |