Open AccessThe Cloud Resource Forecasting Model Based HMM
Author(s) -
Lixin Han,
Haining Meng,
Tong Zhang,
Yi-Lin Qu
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1043/2/022025
Subject(s) - cloud computing , hidden markov model , computer science , downtime , reliability (semiconductor) , software , key (lock) , markov model , markov chain , distributed computing , data mining , reliability engineering , real time computing , operating system , machine learning , artificial intelligence , engineering , power (physics) , physics , quantum mechanics
One of the most important goals for companies that provide cloud computing services is to maintain high availability on large computer systems. In order to accomplish such objective, it is necessary to discovery the reason of poor availability. Software aging is a main factor in cloud computing services, leading to software failures, poor performances and may result in system downtime. This paper investigates the software aging effects on the OpenStack cloud computing platform and describes a forecasting model based on Hidden Markov Models. The prediction analysis of the Hidden Markov Models on the key performance data of the system shows that the Hidden Markov Models has excellent predictive performance and is suitable for the prediction of the reliability of the cloud server system.