Dynamic, Latency-Optimal vNF Placement at the Network Edge

Cziva, Richard; Anagnostopoulos, Christos; Pezaros, Dimitrios P.

doi:10.1109/infocom.2018.8486021

Cited by 168 publications

(80 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cziva et al [43] tackle the VNF placement problem at the network edge in order to minimize the total expected latency from all users to their respective VNFs. They further employ the Optimal Stopping Theory to determine when to re-evaluate the optimal placement problem by taking into account the migration cost and random path delay.…”

Section: Delay-aware Vnf/rule Placement and Routing In Edge Clouds Anmentioning

confidence: 99%

“…They propose a heuristic-based incremental allocation mechanism to solve this problem. Different form the work in [9], [43], we propose a general model to quantitatively calculate flow traversing delay of a SFC in edge clouds, and devise a randomized rounding algorithm to jointly solve the delay-aware VNF placement and traffic routing problem in edge clouds. Moreover, there are also some studies about VNF placement in virtual evolved packet core (4G/5G) networks [44], [45], Cloud-Radio Access Networks [46], etc.…”

Section: Delay-aware Vnf/rule Placement and Routing In Edge Clouds Anmentioning

confidence: 99%

See 1 more Smart Citation

Delay-Aware Virtual Network Function Placement and Routing in Edge Clouds

Yang,

Li,

Trajanovski

et al. 2021

IEEE Trans. on Mobile Comput.

View full text Add to dashboard Cite

Mobile Edge Computing (MEC) offers a way to shorten the cloud servicing delay by building the small-scale cloud infrastructures at the network edge, which are in close proximity to the end users. Moreover, Network Function Virtualization (NFV) has been an emerging technology that transforms from traditional dedicated hardware implementations to software instances running in a virtualized environment. In NFV, the requested service is implemented by a sequence of Virtual Network Functions (VNF) that can run on generic servers by leveraging the virtualization technology. Service Function Chaining (SFC) is defined as a chain-ordered set of placed VNFs that handles the traffic of the delivery and control of a specific application. NFV therefore allows to allocate network resources in a more scalable and elastic manner, offer a more efficient and agile management and operation mechanism for network functions and hence can largely reduce the overall costs in MEC. In this paper, we study the problem of how to place VNFs on edge and public clouds and route the traffic among adjacent VNF pairs, such that the maximum link load ratio is minimized and each user's requested delay is satisfied. We consider this problem for both totally ordered SFCs and partially ordered SFCs. We prove that this problem is NP-hard, even for the special case when only one VNF is requested. We subsequently propose an efficient randomized rounding approximation algorithm to solve this problem. Extensive simulation results show that the proposed approximation algorithm can achieve close-to-optimal performance in terms of acceptance ratio and maximum link load ratio.

show abstract

Section: Delay-aware Vnf/rule Placement and Routing In Edge Clouds Anmentioning

confidence: 99%

Section: Delay-aware Vnf/rule Placement and Routing In Edge Clouds Anmentioning

confidence: 99%

Delay-Aware Virtual Network Function Placement and Routing in Edge Clouds

Yang,

Li,

Trajanovski

et al. 2021

IEEE Trans. on Mobile Comput.

View full text Add to dashboard Cite

show abstract

“…Resource allocation and job scheduling is a fundamental problem in edge computing [5], [6], [6]- [11], [13], [22]- [24]. Existing work has shown the importance and challenges of designing efficient application job assignment and scheduling as many edge applications need coordinated computation and communication between multiple components (e.g., containers, micro-services) assigned and scheduled on geo-distributed edge nodes.…”

Section: Related Workmentioning

confidence: 99%

A Deep Reinforcement Learning Approach to Multi-Component Job Scheduling in Edge Computing

Cao

Zhang

Cao

et al. 2019

2019 15th International Conference on Mobile Ad-Hoc and Sensor Networks (MSN)

View full text Add to dashboard Cite

We are interested in the optimal scheduling of a collection of multi-component application jobs in an edge computing system that consists of geo-distributed edge computing nodes connected through a wide area network. The scheduling and placement of application jobs in an edge system is challenging due to the interdependence of multiple components of each job, and the communication delays between the geographically distributed data sources and edge nodes and their dynamic availability. In this paper we explore the feasibility of applying Deep Reinforcement Learning (DRL) based design to address these challenges. We introduce a DRL actor-critic algorithm that aims to find an optimal scheduling policy to minimize average job slowdown in the edge system. We have demonstrated through simulations that our design outperforms a few existing algorithms, based on both synthetic data and a Google cloud data trace.

show abstract

“…In addition, placing service functions nearby the end‐users can reduce the end‐to‐end latency by reducing the network transmission delay between the end‐user and the service function. In NFV, utilizing edge resources for VNFs has been studied by many researchers in recent years …”

Section: Introductionmentioning

confidence: 99%

“…[5][6][7][8] Despite the increased attention of NFV and edge computing, there are limited tools that can verify and evaluate new techniques in the literature. Several proof-of-concept platforms have been proposed by researchers [8][9][10] to show the potential performance and orchestration abilities of NFV over edge computing. However, evaluating a new method in a large-scale is troublesome with the proposed platforms because the empirical system has to be deployed to a large-scale infrastructure in order to evaluate at such scale.…”

Section: Introductionmentioning

confidence: 99%

CloudSimSDN‐NFV: Modeling and simulation of network function virtualization and service function chaining in edge computing environments

Son

Buyya

2019

Softw Pract Exp

View full text Add to dashboard Cite

Summary Software‐defined networking (SDN) has evolved and brought an innovative paradigm shift in computer networks by utilizing a programmable software controller with open protocols. Network functions, previously served on dedicated hardware, have shifted to network function virtualization (NFV) that enabled functions to be virtualized and provisioned dynamically upon generic hardware. In addition to NFV, edge computing utilizes the edge resources close to end‐users, which can reduce the end‐to‐end service delay and the network traffic volume. Although these innovative technologies gained significant attention from both academia and industry, there are limited tools and simulation frameworks for the effectiveness evaluation in a repeatable and controllable manner. Furthermore, large‐scale experimental infrastructures are expensive to setup and difficult to maintain. Even if they are created, they are not available or accessible for the majority of researchers throughout the world. In this paper, we propose a framework for simulating NFV functionalities in both edge and cloud computing environments. In addition to the basic network functionalities supported by SDN in CloudSimSDN, we added new NFV features, such as virtualized network functions allocation, migration, and autoscaling with the support of corresponding network functionalities, such as flow load balancing, rerouting, and service function chaining (SFC) maintenance. We evaluated our simulation framework with autoscaling and placement policies for SFC in the integrated edge and cloud computing environments. The results demonstrate its effectiveness in measuring and evaluating the end‐to‐end delay, response time, resource utilization, network traffic, and power consumption with different algorithms in each scenario.

show abstract

Dynamic, Latency-Optimal vNF Placement at the Network Edge

Cited by 168 publications

References 18 publications

Delay-Aware Virtual Network Function Placement and Routing in Edge Clouds

Delay-Aware Virtual Network Function Placement and Routing in Edge Clouds

A Deep Reinforcement Learning Approach to Multi-Component Job Scheduling in Edge Computing

CloudSimSDN‐NFV: Modeling and simulation of network function virtualization and service function chaining in edge computing environments

Contact Info

Product

Resources

About