作者: Brett Snyder , Jordan Ringenberg , Robert Green , Vijay Devabhaktuni , Mansoor Alam
DOI: 10.1186/S13677-015-0036-6
关键词:
摘要: Cloud computing paradigm has ushered in the need to provide resources users a scalable, flexible, and transparent fashion much like any other utility. This led for developing evaluation techniques that can quantitative measures of reliability cloud system (CCS) efficient planning expansion. paper presents new, scalable algorithm based on non-sequential Monte Carlo Simulation (MCS) evaluate large scale reliability, it develops appropriate performance measures. Also, new iterative is proposed developed leverages MCS method design highly reliable utilized CCSs. The combination these two algorithms allows CCSs be evaluated by providers alike, providing estimating parameters service level agreements (SLAs) designing match those contractual requirements posed SLAs. Results demonstrate methods are effective applicable systems at scale. Multiple insights also provided into nature CCS design.