selected publications
-
conference proceeding
-
ReCycle: Resilient Training of Large DNNs using Pipeline Adaptation.
211-228.
2024
- High-throughput and Flexible Host Networking for Accelerated Computing 2024
- RecD: Deduplication for End-to-End Deep Learning Recommendation Model Training Infrastructure 2023
- Tectonic-Shift: A Composite Storage Fabric for Large-Scale ML Training 2023
-
Understanding data storage and ingestion for large-scale deep recommendation model training.
1042-1057.
2022
-
ShEF: shielded enclaves for cloud FPGAs.
1070-1085.
2022
-
Llama.
1-17.
2021
-
HyperFlow.
1583-1600.
2018
-
FPGA-Based Remote Power Side-Channel Attacks.
229-244.
2018
-
ReCycle: Resilient Training of Large DNNs using Pipeline Adaptation.
211-228.
2024
-
journal article
-
Remote Power Side- Channel Attacks on FPGAs.
IEEE Design and Test.
13-19.
2025
-
cedar; : Optimized and Unified Machine Learning Input Data Pipelines.
Proceedings of the VLDB Endowment.
488-502.
2024
-
Remote Power Side- Channel Attacks on FPGAs.
IEEE Design and Test.
13-19.
2025