[Paper Review] Next Generation Cloud Computing: New Trends and Research Directions

Blesson Varghese, Rajkumar Buyya|arXiv (Cornell University)|Jul 24, 2017

Cloud Computing and Resource Management83 references9 citations

TL;DR

This paper outlines the evolution of cloud computing toward multi-cloud, edge-based, and decentralized architectures, proposing new models like cloudlets, fog computing, and serverless execution to enhance scalability, resilience, and energy efficiency. It identifies key research challenges in security, application expressiveness, resource management, and sustainability, emphasizing the shift from centralized data centers to distributed, heterogeneous infrastructures for next-generation systems.

ABSTRACT

The landscape of cloud computing has significantly changed over the last decade. Not only have more providers and service offerings crowded the space, but also cloud infrastructure that was traditionally limited to single provider data centers is now evolving. In this paper, we firstly discuss the changing cloud infrastructure and consider the use of infrastructure from multiple providers and the benefit of decentralising computing away from data centers. These trends have resulted in the need for a variety of new computing architectures that will be offered by future cloud infrastructure. These architectures are anticipated to impact areas, such as connecting people and devices, data-intensive computing, the service space and self-learning systems. Finally, we lay out a roadmap of challenges that will need to be addressed for realising the potential of next generation cloud systems.

Motivation & Objective

• To analyze the transformation of cloud infrastructure from centralized data centers to decentralized, multi-provider, and edge-based models.
• To identify emerging computing architectures—such as cloudlets, fog computing, and serverless computing—that support low-latency, scalable, and resilient applications.
• To examine the impact of these new models on data-intensive workloads, IoT integration, and self-learning systems.
• To outline critical research challenges in security, application expressiveness, efficient resource management, and sustainability in next-generation cloud systems.
• To propose a roadmap for advancing cloud computing toward greener, more resilient, and interoperable infrastructures across heterogeneous providers.

Proposed method

• Analyzes trends in cloud infrastructure evolution, including multi-cloud, micro-cloud, cloudlets, ad hoc clouds, and heterogeneous cloud deployments.
• Maps infrastructure changes across nine abstraction layers (network to application), identifying where architectural changes are required.
• Proposes serverless computing as a model to replace traditional VM-based provisioning, reducing idle resource costs.
• Introduces software-defined and resilient cloud models to improve dynamic resource orchestration and fault tolerance.
• Proposes energy-aware provisioning and carbon footprint-aware scheduling algorithms to reduce environmental impact.
• Advocates for integrated management of servers, networks, and cooling systems using IoT-enabled feedback for dynamic power state control.

Experimental results

Research questions

• RQ1 How can cloud infrastructure evolve beyond single-provider data centers to support decentralized, multi-cloud, and edge-based computing?
• RQ2 What new computing architectures are needed to support low-latency, data-intensive, and IoT-driven workloads in next-generation cloud systems?
• RQ3 How can energy efficiency and sustainability be integrated as first-class QoS metrics in cloud resource provisioning and management?
• RQ4 What are the key challenges in securing, expressing, and managing applications across heterogeneous, multi-cloud, and decentralized environments?
• RQ5 How can resilience and dynamic resource orchestration be enhanced through software-defined and self-healing cloud architectures?

Key findings

• Multi-cloud and edge-based models such as cloudlets and fog computing are emerging as critical for reducing latency and improving responsiveness in mobile and IoT applications.
• Serverless computing is gaining traction as a model that eliminates idle VM costs by charging only for actual execution time.
• Energy consumption in data centers remains a major concern, with cooling and networking contributing significantly to total power use.
• Carbon footprint-aware and energy-aware provisioning algorithms can reduce environmental impact while meeting QoS requirements, though trade-offs with performance exist.
• The integration of IoT-enabled systems for real-time monitoring and control of server, network, and cooling systems enables dynamic, adaptive power management.
• A shift toward decentralized, heterogeneous, and multi-provider cloud infrastructures is inevitable, driving the need for new research in security, interoperability, and sustainable system design.