ТIME SERIES MODEL FOR PREDICTING AWS SPOT INSTANCE PRICES
DOI:
https://doi.org/10.24867/28BE28PranjkicKeywords:
spot instances, time series, price prediction model, SARIMAAbstract
The growing demand for computing resources has spurred the use of cloud computing providers. This has led to the development of a new dimension where the relationship between resource usage and cost must be considered from an organizational perspective. As part of its EC2 (Elastic Compute Cloud) service, Amazon introduced spot instances as a cheap public infrastructure at the price of service unreliability. The price of spot instances is a key factor in cost management and resource optimization in cloud computing. Dynamic pricing can complicate the process of planning and resource allocation. Therefore, accurate prediction of spot instance prices can greatly benefit cloud users, allowing them to maximize their resources and minimize costs.
References
[1] Wenqiang Liu† , Pengwei Wang*† , Ying Meng, Caihui Zhao and Zhaohui Zhang, Cloud spot instance price prediction using kNN regression
[2] Alkharif S, Lee K, Kim H (2018) Time-series analysis for price prediction of opportunistic Cloud computing resources. 2018 7th international conference on emerging databases. Springer, Singapore, pp 221–229
[3] Dawei Kong, Shijun Liu, Li Pan, Amazon Spot Instance Price Prediction with GRU Network
[4] G.E.P. Box, G. Jenkins, “Time Series Analysis, Forecasting and Control”, Holden-Day, San Francisco, CA, 1970.
[5] Vaia I. Kontopoulou 1, Athanasios D. Panagopoulos 2,* , Ioannis Kakkos 1 and George K. Matsopoulos 1, A Review of ARIMA vs. Machine Learning Approaches for Time Series Forecasting in Data Driven Networks
[6] Paliari, I.; Karanikola, A.; Kotsiantis, S. A comparison of the optimized LSTM, XGBOOST and ARIMA in Time Series forecasting. In Proceedings of the 2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA), Chania Crete, Greece, 12–14 July 2021; pp. 1–7.
[2] Alkharif S, Lee K, Kim H (2018) Time-series analysis for price prediction of opportunistic Cloud computing resources. 2018 7th international conference on emerging databases. Springer, Singapore, pp 221–229
[3] Dawei Kong, Shijun Liu, Li Pan, Amazon Spot Instance Price Prediction with GRU Network
[4] G.E.P. Box, G. Jenkins, “Time Series Analysis, Forecasting and Control”, Holden-Day, San Francisco, CA, 1970.
[5] Vaia I. Kontopoulou 1, Athanasios D. Panagopoulos 2,* , Ioannis Kakkos 1 and George K. Matsopoulos 1, A Review of ARIMA vs. Machine Learning Approaches for Time Series Forecasting in Data Driven Networks
[6] Paliari, I.; Karanikola, A.; Kotsiantis, S. A comparison of the optimized LSTM, XGBOOST and ARIMA in Time Series forecasting. In Proceedings of the 2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA), Chania Crete, Greece, 12–14 July 2021; pp. 1–7.
Downloads
Published
2024-09-05
Issue
Section
Electrotechnical and Computer Engineering
