FusionFlow: Accelerating Data Preprocessing for Machine Learning with CPU-GPU Cooperation

Summary: FusionFlow runs data-augmentation on CPUs and GPUs with GPU-aware allocations inside GPU free space and adaptive scheduling to avoid interfering with model training. Yields 16–285% single-machine throughput gains and cuts CPU needs ~50–60% versus CPU-only or remote preprocessing. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 13761
Venue: VLDB
Year: 2024
Pagerank: 4.5366487e-05
Overall Rank: 8,343 | 42.02%
DOI: 10.14778/3636218.3636238

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
8,731	TensorSocket: Shared Data Loading for Deep Learning Training	2026	SIGMOD	4.4520434e-05
10,776	cedar: Optimized and Unified Machine Learning Input Data Pipelines	2025	VLDB	4.1905499e-05

Outgoing Citations (Sorted by Pagerank)

Showing 5 of 5 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
1,500	Analyzing and Mitigating Data Stalls in DNN Training	2021	VLDB	0.00011636174
2,175	tf.data: A Machine Learning Data Processing Framework	2021	VLDB	9.3745231e-05
4,175	FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline	2023	VLDB	6.3772575e-05
5,332	Heterogeneity-Aware Distributed Machine Learning Training via Partial Reduce	2021	SIGMOD	5.5640779e-05
5,560	GoldMiner: Elastic Scaling of Training Data Pre-Processing Pipelines for Deep Learning	2023	SIGMOD	5.4350242e-05

Semantically Similar Papers

Overall Rank	Paper	Year	Venue	Pagerank
9,776	Data Augmentation for ML-driven Data Preparation and Integration	2021	VLDB	4.2815042e-05
10,224	SVFusion: A CPU-GPU Co-Processing Architecture for Large-Scale Real-Time Vector Search	2026	VLDB	4.1905499e-05
5,811	CGgraph: An Ultra-fast Graph Processing System on Modern Commodity CPU-GPU Co-processor	2024	VLDB	5.3168243e-05
7,180	Coresets over Multiple Tables for Feature-rich and Data-efficient Machine Learning	2023	VLDB	4.8032775e-05
5,560	GoldMiner: Elastic Scaling of Training Data Pre-Processing Pipelines for Deep Learning	2023	SIGMOD	5.4350242e-05
6,361	Galvatron: Efficient Transformer Training over Multiple GPUs Using Automatic Parallelism	2023	VLDB	5.0903244e-05
8,731	TensorSocket: Shared Data Loading for Deep Learning Training	2026	SIGMOD	4.4520434e-05
2,175	tf.data: A Machine Learning Data Processing Framework	2021	VLDB	9.3745231e-05
3,694	Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines	2022	SIGMOD	6.8316905e-05
4,175	FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline	2023	VLDB	6.3772575e-05