Scholarlink

This research investigates the effectiveness of deep learning (DL) for virtual machine (VM) load prediction. We propose hybrid models combining convolutional neural networks (CNN), long short-term memory (LSTM) networks, gated recurrent units (GRU), and Transformer models utilizing self-attention mechanisms for VM workload prediction. Our findings demonstrate that these hybrid DL models show superior performance over standalone models, effectively capturing both spatial and temporal features of workload data. However, Transformer models did not perform well in time series data prediction, indicating the need for further optimization.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparative Analysis of Deep Learning Models for VM Load Prediction

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Lijender Rathour,
  • Bharati Sinha

摘要

This research investigates the effectiveness of deep learning (DL) for virtual machine (VM) load prediction. We propose hybrid models combining convolutional neural networks (CNN), long short-term memory (LSTM) networks, gated recurrent units (GRU), and Transformer models utilizing self-attention mechanisms for VM workload prediction. Our findings demonstrate that these hybrid DL models show superior performance over standalone models, effectively capturing both spatial and temporal features of workload data. However, Transformer models did not perform well in time series data prediction, indicating the need for further optimization.