A Manifesto for Future Generation Cloud Computing

ACM Computing Surveys, Год журнала: 2018, Номер 51(5), С. 1 - 38

Опубликована: Ноя. 19, 2018

The Cloud computing paradigm has revolutionised the computer science horizon during past decade and enabled emergence of as fifth utility. It captured significant attention academia, industries, government bodies. Now, it emerged backbone modern economy by offering subscription-based services anytime, anywhere following a pay-as-you-go model. This instigated (1) shorter establishment times for start-ups, (2) creation scalable global enterprise applications, (3) better cost-to-value associativity scientific high-performance (4) different invocation/execution models pervasive ubiquitous applications. recent technological developments paradigms such serverless computing, software-defined networking, Internet Things, processing at network edge are creating new opportunities computing. However, they also posing several challenges need approaches research strategies, well re-evaluation that were developed to address issues scalability, elasticity, reliability, security, sustainability, application models. proposed manifesto addresses them identifying major open in emerging trends, impact areas. then offers directions next decade, thus helping realisation Future Generation Computing.

Язык: Английский

---

A Survey on the Placement of Virtual Resources and Virtual Network Functions

IEEE Communications Surveys & Tutorials, Год журнала: 2018, Номер 21(2), С. 1409 - 1434

Опубликована: Дек. 3, 2018

Cloud computing and network slicing are essential concepts of forthcoming 5G mobile systems. Network slices essentially chunks virtual connectivity resources, configured provisioned for particular services according to their characteristics requirements. The success cloud hinges on the efficient allocation resources (e.g., VCPU VMDISK) optimal placement virtualized functions (VNFs) composing slices. In this context, paper elaborates issues that may disrupt VNFs machines (VMs). This classifies existing solutions VM based nature, whether is dynamic or static, objectives, metrics. then proposes a classification VNF approaches, first, regarding general management VNFs, second, target type.

Язык: Английский

---

A survey on load balancing algorithms for virtual machines placement in cloud computing

Concurrency and Computation Practice and Experience, Год журнала: 2017, Номер 29(12)

Опубликована: Март 6, 2017

The emergence of cloud computing based on virtualization technologies brings huge opportunities to host virtual resource at low cost without the need owning any infrastructure. Virtualization enable users acquire, configure and be charged pay-per-use basis. However, Cloud data centers mostly comprise heterogeneous commodity servers hosting multiple machines (VMs) with potential various specifications fluctuating usages, which may cause imbalanced utilization within that lead performance degradation service level agreements (SLAs) violations. To achieve efficient scheduling, these challenges should addressed solved by using load balancing strategies, have been proved NP-hard problem. From perspectives, this work identifies analyzes existing algorithms for allocating VMs PMs in infrastructure Clouds, especially focuses balancing. A detailed classification targeting VM placement is investigated surveyed are classified according classification. goal paper provide a comprehensive comparative understanding literature aid researchers providing an insight future enhancements.

Язык: Английский

---

Resource Allocation and Service Provisioning in Multi-Agent Cloud Robotics: A Comprehensive Survey

IEEE Communications Surveys & Tutorials, Год журнала: 2021, Номер 23(2), С. 842 - 870

Опубликована: Янв. 1, 2021

Robotic applications nowadays are widely adopted to enhance operational automation and performance of real-world Cyber-Physical Systems (CPSs) including Industry 4.0, agriculture, healthcare, disaster management. These composed latency-sensitive, data-heavy, compute-intensive tasks. The robots, however, constrained in the computational power storage capacity. concept multi-agent cloud robotics enables robot-to-robot cooperation creates a complementary environment for robots executing large-scale with capability utilize edge resources. However, such collaborative environment, optimal resource allocation robotic tasks is challenging achieve. Heterogeneous energy consumption rates application execution costs associated computing instances make it even more complex. In addition, data transmission delay between local nodes, centres adversely affects real-time interactions impedes service guarantee. Taking all these issues into account, this paper comprehensively surveys state-of-the-art on provisioning robotics. presents domains through explicit comparison contemporary paradigms identifies specific research challenges. A complete taxonomy presented first time, together discussion pooling, computation offloading, task scheduling efficient provisioning. Furthermore, we highlight gaps from learned lessons, present future directions deemed beneficial further advance emerging field.

Язык: Английский

---

Recent advancements in resource allocation techniques for cloud computing environment: a systematic review

Cluster Computing, Год журнала: 2016, Номер 20(3), С. 2489 - 2533

Опубликована: Дек. 8, 2016

Язык: Английский

---

Energy and Migration Cost-Aware Dynamic Virtual Machine Consolidation in Heterogeneous Cloud Datacenters

IEEE Transactions on Services Computing, Год журнала: 2016, Номер 12(4), С. 550 - 563

Опубликована: Окт. 12, 2016

Energy efficiency has become one of the major concerns for today's cloud datacenters. Dynamic virtual machine (VM) consolidation is a promising approach improving resource utilization and energy However, live migration technology that VM relies on costly in itself, this cost usually heterogeneous as well datacenter. This paper investigates following bi-objective optimization problem: how to pay limited costs save much possible via dynamic To capture these two conflicting objectives, score function designed an overall evaluation basis estimation method upper bound maximal saved power. optimize score, greedy heuristic swap operation are introduced, improved grouping genetic algorithm (IGGA) based them proposed. Lastly, empirical studies performed, results show IGGA outperforms existing methods.

Язык: Английский

---

Approaches for optimizing virtual machine placement and migration in cloud environments: A survey

Journal of Parallel and Distributed Computing, Год журнала: 2017, Номер 111, С. 222 - 250

Опубликована: Сен. 20, 2017

Язык: Английский

---

Container‐based cluster orchestration systems: A taxonomy and future directions

Software Practice and Experience, Год журнала: 2018, Номер 49(5), С. 698 - 719

Опубликована: Ноя. 11, 2018

Summary Containers, enabling lightweight environment and performance isolation, fast flexible deployment, fine‐grained resource sharing, have gained popularity in better application management deployment addition to hardware virtualization. They are being widely used by organizations deploy their increasingly diverse workloads derived from modern‐day applications such as web services, big data, internet of things either proprietary clusters or private public cloud data centers. This has led the emergence container orchestration platforms, which designed manage containerized large‐scale clusters. These systems capable running hundreds thousands jobs across machines. To do so efficiently, they must address several important challenges including scalability, fault tolerance availability, efficient utilization, request throughput maximization among others. paper studies these proposes a taxonomy that identifies different mechanisms can be meet aforementioned challenges. The proposed classification is then applied various state‐of‐the‐art leading identification open research gaps literature intended future directions for researchers.

Язык: Английский

---

Virtual Network Functions Routing and Placement for Edge Cloud Latency Minimization

IEEE Journal on Selected Areas in Communications, Год журнала: 2018, Номер 36(10), С. 2346 - 2357

Опубликована: Сен. 12, 2018

As a new way to design, deploy, and manage network services, functions virtualization (NFV) decouples the functions, from one or more physical infrastructures black boxes so they can run in software. It therefore comes as no surprise that NFV originated service providers, who were looking improve deployment of services support their revenue growth objectives. Within ecosystem, high availability, low latency are key quality (QoS) benefits providers expect 5G Cloud networks make delay-critical such remote surgery reality. Therefore, should be placed, chained, routed through considering users/tenants stringent QoS service-level agreement requirements. To this end, routing placement optimization plays major role improving performance overall cost. In paper, we study problem virtual (VNFs) across hosts minimize defined queuing delay within edge clouds links. respect, paper takes holistic view by not only VNFs chaining but also flows aspect since these two problems inter-related have impact on latency.

Язык: Английский

---

Latency and Availability Driven VNF Placement in a MEC-NFV Environment

2015 IEEE Global Communications Conference (GLOBECOM), Год журнала: 2018, Номер unknown

Опубликована: Дек. 1, 2018

Multi-access Edge Computing (MEC) is gaining momentum as it considered one of the enablers 5G ultra-Reliable Low-Latency Communications (uRLLC) services. MEC deploys computation resources close to end user, enabling reduce drastically end-to-end latency. ETSI has recently leveraged architecture run all entities, including applications, Virtual Network Functions (VNF) in a Virtualization (NFV) environment. This evolution allows taking advantage mature and tools NFV, potential apply variety service-tailored function placement algorithms. However, latter need be carefully designed case applications such uRLLC, where service access latency critical. In this paper, we propose novel scheme applicable NFV particular, formulation problem VNF tailored uRLLC an optimization two conflicting objectives, namely minimizing maximizing availability. To deal with complexity problem, Genetic Algorithm solve it, which compare CPLEX implementation our model. Our numerical results show that heuristic algorithm runs efficiently produces solutions approximate well optimal, reducing providing highly-available service.

Язык: Английский

---