We benchmarked this new implementation on the Amazon Elastic Cloud (EC2) in order to estimate its efficiency and also measure the cost associated with running this type of application on a cloud-environment.
Traditional techniques for auto-scaling are easy to deploy and use a mechanism to manage the number of resources assigned to an application hosted on a cloud platform (e.g., [18]).
Most applications deployed on a cloud infrastructure require a minimum number of virtual machines to operate around the clock.
And since Hadoop is designed to distribute work across hundreds or thousands of computers, the most economical way to deploy applications is on a cloud service.
The capability provided to the consumer is to use the provider's applications running on a cloud infrastructure.
In summary, RAP can take advantage of burstiness characteristics of applications hosted on a cloud to suggest cost-effective resource allocations.
The C-suite and procurement leaders face a common dilemma when choosing procurement software: should the new software system be an "on-premises" application or a cloud solution?
In this paper, we propose a flexible collaboration approach, named CyberLiveApp, to enable live virtual desktop application sharing, based on a cloud and virtualization infrastructure.
Our system can be run stand-alone, embedded in other applications, or deployed on a cloud computing infrastructure for large-scale document processing using the GATE Cloud Parallelizer (GCP).
Both ARVUE and CRAMP provide autonomous shared hosting of third-party Java Servlet applications on an IaaS cloud.
