PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 29, 2026Array0 citationsOpen Access

Adaptive weighting hybrid deep learning with Savitzky-Golay filter and intrinsic mode functions for workload prediction in next-generation data centers

View Full Paper
MMMbarek MarwanATAbdelkarim Ait TemghartMLMohamed Lazaar

Key Points

  • This research aims to enhance workload prediction in data centers to improve resource scheduling and energy efficiency under varying workloads.
  • Developed a two-stage prediction framework integrating data cleaning, optimization, and time-series forecasting.
  • Utilized Local Outlier Factor (LOF) for outlier removal, Savitzky-Golay filter for noise smoothing, and Empirical Mode Decomposition (EMD) for signal analysis.
  • Employed a hybrid model combining CNN, BiLSTM, and attention mechanisms for feature extraction and temporal correlation.
  • Achieved a mean squared error (MSE) of 0.030 and a root mean squared error (RMSE) of 0.173, with an R2 of 0.89.
  • Demonstrated superior performance compared to baseline models, with a processing time of 21.5 seconds.
  • Optimizing the selection of 7-9 intrinsic mode functions (IMFs) significantly improved modeling accuracy and reduced noise.

Abstract

ABSTRACT Existing data centers face significant challenges in meeting QoS requirements for resource scheduling while preserving energy efficiency under dynamic and heterogeneous workloads. To address these limitations, we introduce a two-stage prediction framework that integrates a set of complementary techniques, including data cleaning, digital filtering, optimization, and time-series forecasting, to enable proactive resource management and efficient scaling. The contribution of this research is twofold. On the one hand, we design an advanced preprocessing pipeline in which the Local Outlier Factor (LOF) removes abnormal workload spikes, the Savitzky–Golay (SG) filter smooths high-frequency noise, and Empirical Mode Decomposition (EMD) splits nonlinear and non-stationary signals into multi-scale components. On the other hand, we develop a hybrid model that leverages CNN to extract spatial feature correlations, BiLSTM to capture long-term temporal dependencies, and an attention mechanism to highlight the most informative time steps. The Bitbrains fastStorage dataset is used to evaluate the proposed model across multiple time intervals. Overall, the enhanced CNN–BiLSTM–Attention model delivers superior performance (MSE = 0.030, RMSE = 0.173, R 2 = 0.89, Time = 21.5 s), outperforming all baseline models. An ablation study is conducted to assess the contribution of each component to the overall system performance. Importantly, selecting an optimal subset of 7–9 IMFs (Intrinsic Mode Functions) reduces noise, extracts meaningful oscillatory patterns, and enables faster and more accurate modeling.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Marwan et al. (2026) studied this question.

synapsesocial.com/papers/6a192d13fab5b468c4415ea6https://doi.org/10.1016/j.array.2026.100948
Ask AI
Helpful
Bookmark
Share
View Full Paper