Cloud Resource Provisioning to Extend the Capacity of Local Resources in the Presence of Failures

Abstract

Abstract-In this paper, we investigate Cloud computing resource provisioning to extend the computing capacity of local clusters in the presence of failures. We consider three steps in the resource provisioning including resource brokering, dispatch sequences, and scheduling. The proposed brokering strategy is based on the stochastic analysis of routing in distributed parallel queues and takes into account the response time of the Cloud provider and the local cluster while considering computing cost of both sides. Moreover, we propose dispatching with probabilistic and deterministic sequences to redirect requests to the resource providers. We also incorporate checkpointing in some well-known scheduling algorithms to provide a fault-tolerant environment. We propose two cost-aware and failure-aware provisioning policies that can be utilized by an organization that operates a cluster managed by virtual machine technology and seeks to use resources from a public Cloud provider. Simulation results demonstrate that the proposed policies improve the response time of users' requests by a factor of 4.10 under a moderate load with a limited cost on a public Cloud.

Cost-aware request routing in multi-geography cloud data centres using software-defined networking

“…Note that the value of each parameter in Table 4 denotes the corresponding mean value. As shown in other works (Javadi, Thulasiraman, and Buyya 2012), infrastructures in each CDC are interconnected through a high-performance network, therefore, the time to transfer the essential software and its data for each application can be negligible. However, our work considers this dimension and gives the realistic setting of parameters including ADsize n and dtr cn in Table 4.…”

Capacity planning for data center to support green computing

“…The constraint in (13) ensures the growing size of data center over the periods. The constraints in (14) and (15) ensure that performance requirements in terms of job waiting time and blocking probability are met in all states ω . The constraint in (16) ensures that the number of servers in the data center is a positive integer and bounded by n m .…”

“…However, all works in the literature related to capacity planning ignored the optimization with the uncertainty of workload demand. Although some works proposed the optimization algorithms for capacity planning of distributed system (e.g., [13], [14]), the impact of power and workload management was not jointly considered. In [15], the capacity planning optimization model with the power and workload management scheme was formulated to determine the optimal number of servers required in new data centers, to locate the new data centers to appropriate areas, and to distribute computing jobs to the data centers where energy costs can be minimized.…”

A Visualized Framework of Automatic Orchestration Engine Supporting Hybrid Cloud Resources

“…Thirdly, they cannot deal with the scenarios in which users deploy systems on top of both VMs and physical servers. This paper is inspired by [15][16][17][18][19]. References [15,16] offer pre-existing common VM images deploying whole application service stacks in several OS versions, such as: different versions of MySQL installed in different OS releases.…”

“…When users select one OS release and a certain software version, our system will install the related software in real-time. References [17][18][19] offer an approach to dynamically provisioning Cloud resource for extending the capacity of local resources. In this paper, we propose a framework of AOE system in a hybrid Cloud environment in which the resources consist of local physical servers, local private Cloud resources, and other private or public Cloud resources.…”