Multi-stage Scheduling with Scalable Resources for Automated Deployment in Platform as a Service Cloud
2015 Seventh International Symposium on Parallel Architectures, Algorithms and Programming (PAAP) (2015)
Dec. 12, 2015 to Dec. 14, 2015
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/PAAP.2015.46
With the automated deployment services of PaaS (Platform as a Service) cloud, developers can focus only on application code, without concern for the application lifecycle management. The large proportion of the workload of PaaS system comes from the application deployment jobs. In the paper, an application deployment job is modeled as a two-stage flow shop type job with the deadline constraint. We explore the deadline-constraint hybrid flow shop problem for the automated deployment in PaaS Cloud. The scalability of virtual machines are also considered for solving the problem. The objective is to decide the number of scaled-out virtual machines for each stage, and sequence jobs on machines so as to finish jobs before their respective deadlines and minimize the total expenses. Two heuristics are proposed for the considered problem. Computational results show that the proposed algorithms performs well on both effectiveness and efficiency.
Cloud computing, Engines, Automation, Scalability, Virtual machining, Runtime, Scheduling
J. Zhu and C. Sha, "Multi-stage Scheduling with Scalable Resources for Automated Deployment in Platform as a Service Cloud," 2015 Seventh International Symposium on Parallel Architectures, Algorithms and Programming (PAAP), Nanjing, China, 2015, pp. 204-209.