Skip to main content
Fig. 10 | EURASIP Journal on Wireless Communications and Networking

Fig. 10

From: A novel approach to workload prediction using attention-based LSTM encoder-decoder network in cloud environment

Fig. 10

Overall RMSE of three RNN-based models with different prediction sequence lengths on the Alibaba dataset. Our method: The black line with square symbols is the RMSE when our method is conducted with different prediction sequence lengths. RMSE is calculated over the entire validation set. GRUED: The red line with dot symbols is the RMSE when LSTM is conducted with different prediction sequence lengths. RMSE is calculated over the entire validation set. LSTM: The blue line with triangle symbols is the RMSE when LSTM is conducted with different prediction sequence lengths. RMSE is calculated over the entire validation set

Back to article page