Journal of Cloud Computing

Advances, Systems and Applications

Table 11 Summary of lowest MAPE values of our proposed models compared to the models in [44]

From: VTGAN: hybrid generative adversarial networks for cloud workload prediction

Method	Best tested	Best tested	Lower
	window size	split ratio	MAPE value
ARIMA [44]	N/A	80:20	37.031
LSTM [44]	120	65:35	11.7246
GRU [44]	90	70:30	11.9765
Bi-LSTM [44]	60	65:35	11.7046
BHyPreC [44]	120	70:30	10.7738
VTGAN (LSTM-based)	30	80:20	9.0233
VTGAN (GRU-based)	60	80:20	8.1686

Back to article page