Service Level Agreement Complexity: Processing Concerns for Standalone\n and Aggregate SLAs
Christopher C. Lamb, Gregory L. Heileman
Abstract
Open-access reader
Christopher C. Lamb, Gregory L. Heileman
Abstract
Open-access reader
In this paper, we examine the problem of a single provider offering multiple\ntypes of service level agreements, and the implications thereof. In doing so,\nwe propose a simple model for machine-readable service level agreements (SLAs)\nand outline specifically how these machine-readable SLAs can be constructed and\ninjected into cloud infrastructures - important for next-generation cloud\nsystems as well as customers. We then computationally characterize the problem,\nestablishing the importance of both verification and solution, showing that in\nthe general case injecting policies into cloud infrastructure is NP-Complete,\nthough the problem can be made more tractable by further constraining SLA\nrepresentations and using approximation techniques.\n
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper, we examine the problem of a single provider offering multiple\ntypes of service level agreements, and the implications thereof. In doing so,\nwe propose a simple model for machine-readable service level agreements (SLAs)\nand outline specifically how these machine-readable SLAs can be constructed and\ninjected into cloud infrastructures - important for next-generation cloud\nsystems as well as customers. We then computationally characterize the problem,\nestablishing the importance of both verification and solution, showing that in\nthe general case injecting policies into cloud infrastructure is NP-Complete,\nthough the problem can be made more tractable by further constraining SLA\nrepresentations and using approximation techniques.\n
Key concepts: Service-level agreement, Service level, Cloud computing, Computer science, Aggregate (composite), Service (business), Distributed computing, Simple (philosophy)