Naima Abrar Shami
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Interests
Naima Abrar Shami has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Vasiliki Kalavri
Boston University, Boston, USA
2 shared publications - Po Hao Chen
Brown University
1 shared publication - Yuchen Lu
Brown University, Providence, USA
1 shared publication - Yuhang Song
Boston University
1 shared publication
Latest Publications
- Open Access
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Bridging GNN Inference and Dataflow Stream Processing: Challenges and Opportunities
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessBridging GNN Inference and Dataflow Stream Processing: Challenges and Opportunities
- Open Access
- 9 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-5
In situ neighborhood sampling for large-scale GNN training
Graph Neural Network (GNN) training algorithms commonly perform neighborhood sampling to construct fixed-size mini-batches for weight aggregation on GPUs. State-of-the-art disk-based GNN frameworks compute sampling on the CPU, transferring edge partitions from disk to memory for every mini-batch. We argue that this design incurs significant waste of PCIe bandwidth, as entire neighborhoods are transferred to main memory only to be discarded after sampling. In this paper, we make the first step towards an inherently different approach that harnesses near-storage compute technology to achieve efficient large-scale GNN training. We target a single machine with one or more SmartSSD devices and develop a high-throughput, epoch-wide sampling FPGA kernel that enables pipelining across epochs. When compared to a baseline random-access sampling kernel, our solution achieves up to 4.26× lower sampling time per epoch.
- 9 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Open AccessIn situ neighborhood sampling for large-scale GNN training
Graph Neural Network (GNN) training algorithms commonly perform neighborhood sampling to construct fixed-size mini-batches for weight aggregation on GPUs. State-of-the-art disk-based GNN frameworks compute sampling on the CPU, transferring edge partitions from disk to memory for every mini-batch. We argue that this design incurs significant waste of PCIe bandwidth, as entire neighborhoods are transferred to main memory only to be discarded after sampling. In this paper, we make the first step towards an inherently different approach that harnesses near-storage compute technology to achieve efficient large-scale GNN training. We target a single machine with one or more SmartSSD devices and develop a high-throughput, epoch-wide sampling FPGA kernel that enables pipelining across epochs. When compared to a baseline random-access sampling kernel, our solution achieves up to 4.26× lower sampling time per epoch.