A Systematic Mapping Study on Real-Time Cloud Services

摘要

Cloud computing is relatively a new technique to host and use the services and applications from the internet. Although it offers a multitude of advantages like scalability, low operating cost, accessibility and maintainability, etc., they are often not utilized to the fullest due to the lack of timeliness property associated with the cloud. Cloud services are mainly designed to maximize throughput and utilization of resources and hence incorporating predictable execution time properties in to the cloud is arduous. Nevertheless, cloud still remains a highly attractive platform for hosting real-time applications and services owing to features like elasticity, multi-tenancy, ability to survive hardware failures, virtualization support and abstraction layer support which provides flexibility and portability. In order for real-time safety-critical applications to exploit the potential of cloud computing, it is essential to ensure the predictable real-time behavior of cloud services. In this paper, we perform a systematic mapping study on real-time cloud services to identify the current research directions and potential research gaps. Our study focuses on analyzing the current architectures and software techniques that are available at present to incorporate real-time property of the cloud services. We also aim at investigating the current challenges involved in realizing a predictable real-time behavior of cloud services.