CometCloud: An Autonomic Cloud Engine

1.1 Introduction Clouds typically have highly dynamic demands for resources with highly heterogeneous and dynamic workloads. For example, the workloads associated with the application can be quite dynamic, both in terms of the number of tasks processed as well as computation requirements of each task. Furthermore, different applications may have very different and dynamic QoS requirements, for example, one application may require high throughput while another may be constrained by a budget, and a third may have to balance both throughput and budget. The performance of a cloud service can also vary based on these varying loads as well as failures, network conditions, etc., resulting in different quality of service to the application. Combining public cloud platforms and integrating them with existing grids and data centers can support on-demand scale-up, scale-down as well as scale-out. Users may want to use resources in their private cloud (or datacenter or grid) first before scaling out onto a public cloud, and may have preference for a particular cloud or may want to combine multiple clouds. However, such integration and interoperability is currently non-trivial. Furthermore, integrating these public cloud platforms with exiting computational grids provides opportunities for on-demand scale-up and scale-down, i.e., cloudbursts. In this chapter, we present the CometCloud autonomic cloud engine. The overarching goal of CometCloud is to realize a virtual computational cloud with resizable computing capability, which integrates local computational environments and public cloud services on-demand, and provide abstractions and mechanisms to support a range of programming paradigms and applications requirements. Specifically, CometCloud enables policy-based autonomic cloudbridging and cloudbursting. Autonomic cloudbridging enables on-the-fly integration of local computational environments (datacenters, grids) and public cloud services (such as Amazon EC2 and Eucalyptus), and autonomic cloudbursting enables dynamic application scale-out to address dynamic workloads, spikes in demands, and other extreme requirements. CometCloud is based on a decentralized coordination substrate, and supports highly heterogeneous and dynamic cloud/Grid infrastructures, integration of public/private clouds and cloudbursts. The coordination substrate is also used to support a decentralized and scalable task space that coordinates the scheduling of task, submitted by a dynamic set of users, onto sets of dynamically provisioned workers on available private and/or public cloud resources based on their QoS constraints such as cost or performance. These QoS constraints along with policies, performance history and the state of resources are used to determine the appropriate size and mix of the public and private clouds that should be …