An Iterative Approach to Trustable Systems Management Automation and Fault Handling

• Published in: Journal of network and systems management Vol. 22; no. 3; pp. 366 - 395
• Main Authors: McLarnon, Barry, Robinson, Philip, Milligan, Peter, Sage, Paul
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01.07.2014; Springer Nature B.V
• Subjects: Automated; Automation; Cloud computing; Clouds; Communications Engineering; Computer centers; Computer Communication Networks; Computer Science; Computer Systems Organization and Communication Networks; Expenditures; Faults; Information Systems and Communication Service; Management; Networks; Operations Research/Decision Theory; Studies; Systems management; Tasks; Cloud computing; Automation; Systems management; Fault handling; Trustability; Enterprise systems; Data center
• ISSN: 1064-7570; 1573-7705
• DOI: 10.1007/s10922-013-9295-z

Automated systems management solutions aim to reduce the pressure on the administrators of complex, large-scale, distributed systems by enabling the automation of many of the common tasks of management. However, this creates a level of abstraction, which can act as a barrier between the administrator and the elements being controlled. This can impede the transition to new management paradigms required by the increase of off-premise resources and hybrid cloud systems. The resulting loss of control of the managed environment can contribute to a loss of trust in automated systems management solutions and affect their broader use. This paper proposes a novel approach where the administrator can control the automation level on a per task basis. Administrators define a management task as they would perform it directly and allow the solution to identify the triggers that cause the task to be enacted. The solution also allows administrators to define relevant task output that can be analyzed for fault states and enable error recovery without manual intervention. The impact of this approach leads to reduced management effort for the administrator, while retaining controllability and keeping automation costs low, along with reducing the incidence of errors.