Search

Computation offloading manages resource-intensive and mobile collaborative applications (MCA) on mobile devices where much processing is replicated with multiple users in the same environment. In this article, we propose a novel hybrid multicast-based task execution framework for multi-access edge computing (MEC), where a crowd of mobile devices at the network edge leverage network-assisted device-to-device (D2D) collaboration for wireless distributed computing (MDC) and outcome sharing. The framework is socially aware in order to build effective D2D links. A key objective of this framework is to achieve an energy-efficient task assignment policy for mobile users. Specifically, we first introduce the socially aware hybrid computation offloading ( SAHCO ) system model, which combines of MEC offloading and D2D offloading in detail. Then, we formulate the energy-efficient task assignment problem by taking into account the necessary constraints. We next propose a Monte Carlo Tree Search based algorithm, named, TA-MCTS for the task assignment problem. Simulation results show that compared to four alternative benchmark solutions in literature, our proposal can reduce energy consumption up to 45.37 percent.

5G will provide a flexible and programmable infrastructure, allowing different networks to share the same access network. A way to respond to the diverse service requirements of 5G while reducing both CAPEX and OPEX is to adopt cloud-native architectures. In this context, micro-services software design, the corner stone of cloud-native architecture, seems to be ideal for 5G. However, despite its several advantages, micro-services raise new challenges which slow its adoption down in the NFV ecosystem. Indeed, steering the expected 5G traffic between cloud-native network function is extremely challenging and is still under-investigated. In this paper, we address the service function chaining (SFC) in micro-service based network function virtualization (NFV) ecosystem from the view of the traffic steering. Specifically, we design and implement a cloud-native SFC framework offering efficient traffic steering mechanisms while considering the network state of the underlying NFV infrastructure. In this context, an optimized network-aware load balancing strategy is proposed. Based on extensive experiments, the results obtained show that our strategy achieved good results in terms of i) end-to-end latency and ii) deployment time.

Ahstract-5G stakeholders make every effort to efficiently handle the unprecedented services and traffic demand. Particularly, Telco operators aim to deploy a cost effective network services within their next generation mobile networks. By leveraging cloud-native architectures, the stringent 5G services requirements can be satisfied while reducing both CAPEX and OPEX. Concretely, the adoption of microservice software design enables the disaggregation of Telco applications into small independent cloud-native network functions. Theses microservices, deployed as containers, are lightweight and perform in an independent way. However, in spite of its various benefits, microservice architecture raises new challenges to deal with in the network function virtualization (NFV) ecosystem. Typically, the service function chaining (SFC) of cloud-native network functions becomes extremely complex and error-prone due to the high number of microservices. In this paper, we address the problem of 5G traffic steering for cloud-native SFC in microservice based NFV ecosystem. Specifically, we envision and deploy an SDNless SFC framework ensuring efficient traffic steering between cloud-native network functions. Besides, we propose an optimized network-aware load balancing algorithm capable of carrying efficiently the traffic while considering the underlying infrastructure network state. Based on robust experiments conducted in the deployed Kubernetes-based platform, the results show the efficiency of our strategy in terms of deployment time and network performance.

International audience; The massive adoption of Cloud services is entailing a tremendous traffic transiting over data center networks (DCNs). Despite the impressive wired throughput, DCNs are suffering from network congestion. To deal with such resource limitation, first, we put forward a novel Hybrid Data Center Network (HDCN) architecture, based on CISCO's Massively Scalable Data Center (MSDC) model. HDCN enables high throughput wireless (IEEE 802.11ad) and wired (IEEE 802.3) transmissions. Second, we focus on one-hop inter-rack communications in HDCN and propose a novel wireless channel assignment algorithm. Our goal is to maximize the total throughput over the wireless and/or wired infrastructure. We formulate the problem as a minimum graph coloring. To address this NP-hard problem, we propound two novel i) exact and ii) heuristic wireless channel assignment algorithms in HDCN denoted respectively by GC-HDCN and GH-GC-HDCN. The latter make use of i) column generation and ii) branch and price optimization schemes to resolve the Integer Linear Programming assignment problem while reducing computation time. Based on extensive simulations with QualNet simulator considering the full network protocol stack, the results obtained for both: i) uniform and ii) real Facebook's workload, show that our proposals outperform the related prominent strategies in terms of latency and throughput.

International audience; Recently Mobile Edge Computing (MEC) promises a great latency reduction by pushing mobile computing and storage to the network edge. MEC solutions allows the intensive applications to be computed in nearby servers at the edge. In this work, we envision a multi-user WiFi-based MEC architecture. We tackle the problem of joint task assignment and resource allocation. The main objective of our scheme is to minimize the energy consumption on the mobile terminal side under the application latency constraint. Based on extensive simulations conducted in NS3 while considering real input traces, we show that our approach outperforms the related prominent strategies in terms of: i) energy consumption and ii) completion delay.

International audience; To support the drastically increasing data demands, Internet giants are urged to rethink their data center design. Unfortunately, the conventional wired data centers struggle to resist to the huge volume of traffic. In this regard, we investigate a radically new methodology by augmenting the wired Data Center Network (DCN) with wireless communication (60 GHz technology). Heretofore, only few researches have dealt with the optimization of multi-hop communications in such Hybrid DCN (HDCN) infrastructures. In this paper, we address the joint routing and channel allocation issue for batched flow requests within HDCN. We propose a novel strategy named Joint Batch Routing and Channel Assignment Heuristic for HDCN (JBH-HDCN). To do so, we first formulate the problem based on an advanced Multi-Commodity Flow model with interference constraints. Then, we propose i) a heuristic-based solution to find the best sequence to process the batched flow requests, and ii) an advanced Dijkstra algorithm to jointly route and assign channels. We assess the performances of our solution JBH- HDCN under real conditions, within Cisco's MSDC infrastructure, using both: i) Altoona Facebook's DCN workload and ii) uniform traces. To do so, a full protocol stack is implemented within QualNet simulator and extensive simulations are conducted. Obtained results show that our scheme outperforms the related strategies.

International audience; To cope with the unprecedented traffic explosion, Internet giants are urged to rethink their data center design. Unfortunately, the conventional wired data centers struggle to support the impressive growth of both data and online services. In this regard, we resort to augmenting the wired data center network (DCN) with wireless communication (60 GHz technology). Heretofore, only few research work have dealt with the optimization of multi-hop communications in such hybrid DCN (HDCN) infrastructures. In this paper, we investigate, in HDCN, the issue of jointly: 1) routing and 2) channel allocating, in both online and batch arrival modes. First, we formulate the online joint routing and channel assignment problem as a minimum weighted perfect matching problem. We propose a new strategy named joint routing and channel allocation algorithm that makes use of the Blossom algorithm and sequentially computes the optimal routing communication paths. Second, to handle the batched arrivals of flows, we formulate the batch joint routing and channel assignment problem as an advanced multi-commodity flow model. We propose two scalable approaches: 1) a heuristic based solution, named joint routing and channel assignment heuristic and 2) an approximate solution, named scalable joint batch routing and channel allocation, based on the Lagrangian relaxation technique. Based on extensive network simulations conducted in QualNet simulator while considering the full protocol stack, the obtained results for both: 1) real Facebook's data center and 2) uniform, traces, show that our schemes outperform the related strategies.

International audience; Data centers are dealing with a rich panoply of applications which are distributed across thousands of servers. In this context, Cloud providers seek to maximize their revenue by meeting the tremendous traffic demand. Unfortunately, the conventional wired infrastructures struggle to resist to such a traffic explosion and new innovative techniques are required. In this respect, we put forward a Hybrid Data Center Network architecture based on the CISCO's Massively Scalable Data Center model, that leverages both the wired (Fiber/Ethernet) and wireless (IEEE 802.11ad) infrastructures. In this paper, we address the problem of batch-routing jointly to spectrum allocation of intra-data center communication flows. To do so, first, we formulate the problem based on Multi-Commodity Flow problem while considering interference constraints. Then, we propose a new strategy named Joint Batch Routing and Channel assignment approach in HDCN (BR-HDCN). Based on extensive simulations conducted in QualNet simulator while considering the full protocol stack, the obtained results for both: i) real Facebook's DC, and ii) uniform, traces, show that our proposal outperforms the related prominent strategies.

International audience; The explosion of traffic demands within data centers accentuates the conventional infrastructures ossification. Unfortunately, the traditional tree based architectures struggle to resist and suffer from congestion and wiring complexity which deeply deteriorate the network performance. In aim to address the QoS degradation, we put forward a Hybrid Data Center Network (HDCN) architecture based on CISCO's Massively Data Center (MSDC) model. Specifically, it combines both the wired (Fiber and/or Gigabit Ethernet) and wireless communications based on 60 GHz band (IEEE 802.11ad). Note that few related work have dealt with the optimization of multi-hop communications in HDCN infrastructures and hence the problem is still open. In this paper, we tackle the challenge of jointly i) routing and ii) allocating wireless channels for intra data centers flows, while considering beamforming antennas. For that, we propose a new strategy named Joint Routing and Channel Allocation Algorithm (JRCA-HDCN). To do so, first the problem is formulated as a Minimum Weight Perfect Matching. Then, our resolution is based on the Blossom algorithm. JRCA-HDCN aims to maximize the throughput of intra-HDCN communication over the wireless and/or wired infrastructure. Based on extensive simulations, conducted in QualNet simulator while considering the full protocol stack, the obtained results show that our proposal outperforms the related prominent strategies in terms of i) end-to-end delay, ii) throughput and iii) spectrum spatial reuse.

International audience; In this paper, we study the adaptive virtual network embedding problem within Cloud’s backbone. The main idea is to take profit from the unused bandwidth but allocated to virtual networks. Consequently, the acceptance rate of new clients will be maximized. However, the congestion rate of virtual links must be minimized in order to maximize the satisfaction of end-users. To do so, first we formulate the problem as K-supplier optimization problem. Then, we propose a novel virtual network embedding strategy denoted by Adaptive-VNE. It is based on the approximation-algorithm for bottleneck problems and backtracking strategy. The proposal is validated by simulations and experimental testbed. The results obtained show that Adaptive-VNE outperforms the most prominent strategies and reaches a good performance.

International audience; Cloud Computing offers elasticity and enhances resource utilisation. This is why its success strongly depends on the efficiency of the physical resource management. This paper deals with dynamic resource reconfiguration to achieve high resource utilisation and to increase Cloud providers income. We propose a new adaptive virtual network resource reconfiguration strategy named VNR-GA to handle dynamic users' needs and to adapt virtual resource allocation according to the applications' requirements. The proposed algorithm VNR-GA is based on Genetic metaheuristic and takes advantage of resources migration techniques to recompute the resource allocation of instantiated virtual networks. In order to optimally adapt the resource allocation according to customers' needs growth, the main idea behind the proposal is to sequentially generate populations of reconfiguration solutions that minimise both the migration and mapping cost and then select the best reconfiguration solution. VNR-GA is validated by extensive simulations and compared to the most prominent related strategy found in literature (i.e., SecondNet). The results obtained show that VNR-GA reduces the rejection rate of i) virtual networks and ii) resource upgrade requests and thus enhances Cloud Provider revenue and customer satisfaction. Moreover, reconfiguration cost is minimised since our proposal reduces both the amount of migrated resources and their new mapping cost.