Design and evaluation of a scalable hierarchical application component placement algorithm for cloud resource allocation

Barshan, Maryam; Moens, Hendrik; Turck, Filip De

doi:10.1109/cnsm.2014.7014155

Cited by 12 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work is the extension of our previous work [48] and [49]. In [48], we presented an optimal ILP-based solution which offers limited scalability, making it only suitable for small datacenters.…”

Section: Related Workmentioning

confidence: 90%

See 1 more Smart Citation

Algorithms for network-aware application component placement for cloud resource allocation

Barshan

Moens

Latré³

et al. 2017

J. Commun. Netw.

Self Cite

View full text Add to dashboard Cite

Due to the soaring popularity of cloud-based services over the last years, the size and the complexity of cloud environments has been growing quickly. In the context of cloud systems, mapping a number of application components to a set of physical servers and assigning cloud resources to those components is challenging. Traditional resource allocation systems which rely on a centralized management paradigm suffer from scalability issues, making them inappropriate for large-scale cloud environments. Therefore, there is a need for providing new management solutions that scale well to large size cloud systems. In this article, we present optimal and heuristic solutions for network-aware placement of multi-component applications with differing component characteristics. The optimal ILP-based solution minimizes the application rejection rate and the cost of mapping while respecting application component requirements and physical network limitations. As the execution time of the optimal model scales exponentially, we also offer scalable heuristic solutions for centralized and hierarchical application placement, which are thoroughly explained and evaluated and compared to the optimal solution. Our evaluations show that while the proposed centralized heuristic is near-optimal, the hierarchical approach is much faster and offers higher scalability compared to a centralized approach, e.g. mapping 2.7 million application components onto 512k servers. Moreover, the percentage of servers used and fully placed applications remain close to that of the centralized and optimal solutions.

show abstract

Section: Related Workmentioning

confidence: 90%

“…This optimal solution is also useful for benchmarking real-time heuristic algorithms. In [49], a decentralized algorithm was designed and evaluated which can be applied to large-scale cloud environments. In this article, we have updated the ILP-based model in 3 ways.…”

Section: Related Workmentioning

confidence: 99%

Algorithms for network-aware application component placement for cloud resource allocation

Barshan

Moens

Latré³

et al. 2017

J. Commun. Netw.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The VNF placement and the datacenters/clouds component placement are conceptually similar. A number of studies have been published in this domain . A survey on resource management in cloud computing environments is presented by Jennings and Stadler .…”

Section: Related Workmentioning

confidence: 99%

“…Breitgand et al study the problem of placing virtual machine instances on physical containers having an objective of reducing latency and communication overhead. A novel design is proposed for components placement problem of scalable hierarchical applications in the cloud environment. In order to address the scalability problem of centralized algorithms, the authors propose a distributed algorithm that achieves performance similar to the one of centralized algorithms.…”

Section: Related Workmentioning

confidence: 99%

CU placement over a reconfigurable wireless fronthaul in 5G networks with functional splits

et al. 2019

View full text Add to dashboard Cite

Mobile network operators are currently facing a tremendous increase in the level of data traffic. Although cell size reduction is one of the most common ways used to accommodate such traffic demand, densely deployed small cells also dramatically increase the level of intercell interference. By centralizing baseband signal processing at powerful computing infrastructures, called centralized unit (CU) pools, cloud radio access network (C-RAN) enables advanced coordination algorithms to be employed in dense small cell networks. In C-RAN, due to stringent bandwidth and latency requirements at the fronthaul links, the optical fiber, thanks to its bandwidth and latency characteristics, continues to be the most prevalent fronthaul medium option. Nevertheless, the optical fiber is one of the fronthaul options, while C-RAN (physical layer radio frequency [PHY-RF] split) is one of the functional splits that can be defined each coming with different fronthaul requirements.In this paper, we formulate and solve a dynamic CU placement problem for mobile networks as an integer linear programming (ILP) problem. In the considered network, CU pools are placed at the edges of the network, and a reconfigurable millimeter wave (MMW) wireless fronthaul links are used in order to provide decentralized units (DUs) with connectivity. We study the impact of different functional splits on the placement cost and on the acceptance ratio using different substrate networks. Lastly, we propose and evaluate a CU placement heuristic algorithm using a numerical simulator. The results reveal that the optimal functional split selection can lead to significant resource utilization benefits in the RAN.Int J Network Mgmt. 2020;30:e2086. wileyonlinelibrary.com/journal/nem to the distributed radio access network (D-RAN) architecture, in which coaxial cables are substituted by optical fibers, and a digital interface is used to carry the in-phase/quadrature (I/Q) samples between the baseband unit and the radio unit. However, since in the D-RAN architecture a Long-Term Evolution (LTE) base station, called eNB, cannot share its frequency resources with other eNBs. These scarce frequency resources will be underutilized most of the time in a day due to the spatiotemporally fluctuating traffic demand. In order to tackle the aforementioned problem, the cloud RAN (C-RAN) 3 architecture has been recently proposed. C-RAN is capable of reducing the total cost of ownership (TCO) of mobile networks while, at the same time, optimizing network capacity, coverage, and power consumption. Such goals are achieved in C-RAN by separating baseband units, termed centralized units (CUs) from the radio units, termed distributed units (DUs) and combining CUs in powerful computing infrastructures, termed CU pools. * CU pools can run on a number of general-purpose of-the-shelf servers deployed in centralized locations.The separation of the signal processing functionalities in the RAN protocol stack between the CU pool and the DU is known as functional split. 5 Traditionally, in the ...

show abstract

“…978-3-901882-77-7 c 2015 IFIP -VNF Placement: The VNF placement problem is conceptually similar to component placement in data-centers and clouds. The amount of literature in this domain is thus humbling [11], [12], [13]. In [11] the authors study the problem of placing virtual machine instances on physical containers in such a way to reduce communication overhead and latency.…”

Section: Related Workmentioning

confidence: 99%

Virtual network functions orchestration in wireless networks

Riggio

Bradai

Rasheed

et al. 2015

2015 11th International Conference on Network and Service Management (CNSM)

View full text Add to dashboard Cite

International audienceNetwork Function Virtualization (NFV) is emerging as one of the most innovative concepts in the networking landscape. By migrating network functions from dedicated mid-dleboxes to general purpose computing platforms, NFV can effectively reduce the cost to deploy and to operate large networks. However, in order to achieve its full potential, NFV needs to encompass also the radio access network allowing Mobile Virtual Network Operators to deploy custom resource allocation solutions within their virtual radio nodes. Such requirement raises several challenges in terms of performance isolation and resource provisioning. In this work we formalize the Virtual Network Function (VNF) placement problem for radio access networks as an integer linear programming problem and we propose a VNF placement heuristic. Moreover, we also present a proof–of–concept implementation of an NFV management and orchestration framework for Enterprise WLANs. The proposed architecture builds upon a programmable network fabric where pure forwarding nodes are mixed with radio and packet processing nodes leveraging on general computing platforms

show abstract

Design and evaluation of a scalable hierarchical application component placement algorithm for cloud resource allocation

Cited by 12 publications

References 13 publications

Algorithms for network-aware application component placement for cloud resource allocation

Algorithms for network-aware application component placement for cloud resource allocation

CU placement over a reconfigurable wireless fronthaul in 5G networks with functional splits

Virtual network functions orchestration in wireless networks

Contact Info

Product

Resources

About