Along with the increase of demand for high definition animation film, when the render system with local computing resources cannot supply enough resources to satisfy the user requirement for time, acquiring additional resources is necessary. The Infrastructure as a service (IaaS) Cloud offers user with computing infrastructures on-demand to be used based on the paradigm of pay-per-use, which provides extra resources with fee to extending the capacity of render system with local cluster. Consequently, the scheduling policy under the hybrid render system should consider the constraints of deadline and budget and billing policy. In this paper, an economy-oriented deadline scheduling policy is proposed, which not only guarantees the deadline for user by the way of employing resources for rendering, but also offers an economic way to hire resources from IaaS Cloud provider reasonably. The experiment with single workload and multi workloads shows that the proposed policy can finish the user’s rendering job before deadline as well as obtain approving cost efficient.
CITATION STYLE
Li, Q., Wu, W., Sun, Z., Wang, L., & Huang, J. (2015). Economy-oriented deadline scheduling policy for render system using IaaS cloud. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9530, pp. 68–78). Springer Verlag. https://doi.org/10.1007/978-3-319-27137-8_6
Mendeley helps you to discover research relevant for your work.