FLEDA: Forecasting-Based Data Partition Placement for Efficient Deep Learning Recommendation Model Training
Open Access
- 8 August 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 8399-8409
Abstract
In recommendation and advertising systems, model training pipelines rely on high-throughput ingestion of fresh, collected behavioral data. These data are typically organized into time-partitioned, multi-level table paths. The access patterns driven by highly concurrent model training tasks exhibit complex cross-path correlations, multi-scale temporal dependencies, and frequent hot-cold transitions, making effective data placement under limited SSD capacity challenging. In this paper, we propose FLEDA, a forecasting-based data partition placement framework that selectively places daily-level training data partitions onto SSDs under constrained storage capacity. We first model historical data access patterns as multivariate and multiscale time series to capture both temporal dynamics and cross-path dependencies. By leveraging dependencies across two dimensions, we propose a model that extracts multi-scale cross-path temporal features and fuses mixture-of-experts to forecast the future throughput of all table paths. Besides, we design a ranking-based placement policy that balances predicted throughput, recent access history, and SSD residency continuity to maximize coverage of future data accesses while avoiding excessive data migration. Experiments on large-scale, real-world training workloads show that FLEDA consistently outperforms state-of-the-art multivariate time-series forecasting models and significantly improves SSDs' throughput coverage with low data migration overhead.Keywords
Funding Information
- National Natural Science Foundation of China (62502430)
- Zhejiang Provincial Natural Science Foundation of China (LQ26F020004)
- ByteDance Research Project (CT20250218122839)
This publication has 35 references indexed in Scilit:
- TempSched: A Temperature-Aware Storage Scheduler for Time Series Across Cloud-Edge-Device2025
- Leveraging Heterogeneous Experts with Advantageous Pattern Memory Learning for Traffic Prediction2025
- CHROME: Concurrency-Aware Holistic Cache Management Framework with Online Reinforcement Learning2024
- An Efficient Deep Reinforcement Learning-Based Automatic Cache Replacement Policy in Cloud Block Storage SystemsIEEE Transactions on Computers, 2023
- Dynamic Resource Allocation for Deep Learning Clusters with Separated Compute and Storage2023
- Mitigating Data Stalls in Deep Learning with Multi-times Data Loading Rule2023
- iCache: An Importance-Sampling-Informed Cache for Accelerating I/O-Bound DNN Model Training2023
- A Survey on Deep Reinforcement Learning for Data Processing and AnalyticsIEEE Transactions on Knowledge and Data Engineering, 2022
- Applied Machine Learning at Facebook: A Datacenter Infrastructure Perspective2018
- Feature Pyramid Networks for Object Detection2017