Next Generation Cloud Computing: New Trends and Research Directions (1707.07452v3)

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.

• The paper highlights a shift from centralized to decentralized multi-cloud and hybrid infrastructures that reduce latency and enhance resource diversity.
• It demonstrates emerging computing models like fog, serverless, and software-defined frameworks that offer agility but require innovative management strategies.
• The study outlines critical research directions in security, application expressivity, economic models, and sustainability for next-generation cloud systems.

Next Generation Cloud Computing: An Overview

The paper by Blesson Varghese and Rajkumar Buyya presents a comprehensive examination of the evolving landscape of cloud computing, highlighting the key trends and research directions anticipated to shape next-generation cloud systems. This work explores emerging infrastructures, computing architectures, and their potential impacts across various domains, identifying significant challenges and proposing forward-looking research avenues.

Evolving Cloud Infrastructure

The paper identifies a shift from traditional, centralized data centers toward a decentralized, multi-provider cloud infrastructure. Key trends include:

• Multi-cloud and Hybrid Models: The integration of resources from multiple providers introduces complexities such as lack of unified APIs and differing resource abstractions. Hybrid clouds combining public and private resources are gaining traction.
• Micro Clouds and Cloudlets: These low-cost, low-power solutions aim to decentralize computing, bringing processing closer to the data source to reduce latency. This trend aligns with the push towards edge computing.
• Ad hoc Clouds: Leveraging underutilized resources from diverse devices, ad hoc clouds offer potential for resource aggregation but face reliability and security challenges.
• Heterogeneous Clouds: Integrating various processor types into cloud infrastructure enhances computing capabilities but introduces challenges in achieving high-level programming abstractions across diverse architectures.

Emerging Computing Architectures

The paper outlines several novel computing models, each addressing distinct demands of future applications:

• Volunteer Computing: Harnessing spare resources, often within social networks, this model presents significant management and security challenges.
• Fog and Mobile Edge Computing: This approach leverages edge devices to process data nearer to the source, supporting IoT applications and minimizing latency.
• Serverless Computing: By abstracting server infrastructure, this model focuses on cost-efficient function execution, necessitating shifts in application development paradigms.
• Software-Defined Computing: Encompassing networks, storage, and computation, this architecture promises greater flexibility and performance but demands new management strategies.

Areas of Impact

These innovations are expected to impact several fields:

• IoT Integration: Enhanced connectivity and processing at the edge facilitate the realization of IoT systems, though security remains a pressing concern.
• Big Data Processing: Emerging architectures offer new avenues for handling the volume and complexity of big data but require novel analytics models.
• Service Diversity: The service landscape is expanding with offerings like Acceleration-as-a-Service and Function-as-a-Service, posing integration challenges.
• Self-Learning Systems: Integrated AI and machine learning can leverage these architectures for scalable self-learning applications.

Challenges and Research Directions

Several challenges must be addressed to fully realize the potential of next-generation cloud systems:

• Security and Privacy: Comprehensive security frameworks are imperative, especially with the increased complexity introduced by edge and ad hoc infrastructures.
• Application Expressivity: New programming paradigms and lightweight algorithms are needed to exploit distributed resources effectively, requiring frameworks that address both modularity and interoperability.
• Economic Models: As distributed systems evolve, new pricing and service models must be devised to accommodate multi-level service agreements.
• Management Strategies: Advanced strategies for resource and service management across heterogeneous environments are necessary to maintain performance and reliability.
• Sustainability: Developing energy-efficient infrastructure and management policies will be critical, given the environmental impact of large-scale computing systems.

The paper concludes that the future of cloud computing lies in the interplay between distributed infrastructures and innovative computing models, fostering environments that support diverse applications while maintaining agility, security, and sustainability. The identified research directions serve as a roadmap for academia and industry, emphasizing collaboration to meet the emerging demands and challenges.