Cloudlet network design optimization

Abstract

Major interest is currently given to the integration of clusters of virtualization servers, also referred to as 'cloudlets', into the access network to allow higher performance and reliability in the access to mobile cloud services. We tackle the cloudlet network design problem for mobile access networks. The model is such that virtual machines are associated with mobile users and are allocated to cloudlets. Designing a cloudlet network implies first determining where to install cloudlet facilities among the available sites, then assigning sets of access points such as basestations to cloudlets, while supporting virtual machine migrations and taking into account partial user mobility information, as well as the satisfaction of service-level agreements. We present linkpath formulations supported by heuristics to compute solutions in reasonable time. We qualify the advantage in considering mobility for both users and virtual machines as up to 40% less cloudlet facilities to install and 40% less virtual machine migrations to execute. We compare two migration modes, bulk and live migration, as a function of mobile cloud service requirements, determining that a high preference should be given to bulk migrations for delay-stringent services such as augmented reality support, while for applications with less stringent delay requirements, live migration appears as largely preferable.

Multi-access edge computing: open issues, challenges and future perspectives

“…However, cloudlet performance relies upon user mobility [41]. In [42], two migration models have been compared cloudlet network design i.e. VM bulk migration and VM live migration.…”

On Uncoordinated Service Placement in Edge-Clouds

“…Related work in the area of resource allocation and request routing for edge-clouds have considered optimal allocation of services and dispatching of requests to edge-cloud nodes, e.g., [9], [10], [11]. In general, these proposals attempt to minimize the average response time of the edge-clouds without considering individual QoS requirements of applications, which we model in the form of deadlines on response delays.…”

Seamless Support of Low Latency Mobile Applications with NFV-Enabled Mobile Edge-Cloud

“…In Section III, we first show that the problem formulated is NP-hard and then, we detail our solution that consists of (1) a capacity violation detection (CVD) mechanism to estimate the time when ALB cannot cope with service elasticity and (2) an online adaptive greedy (OAG) heuristic algorithm to dynamically choose the locations of service-hosting nodes and associated network paths based on the most current servicehosting node deployment. In Section IV, using real mobility traces [16] and a three-level metropolitan scale cellular network [12], we present our evaluation results obtained from a packet-level simulator. We show that our approach satisfies the service-level response time requirement while achieving a cost saving of up to 40% in comparison to current practices.…”

“…On the other hand, work on the resource allocation in an online MEC system has focused on the dynamic routing of user requests to fixed clouds [12], [13], [14], [15]. Locations of physical micro-cloud hardware are first fixed (e.g., after solving the static placement problem), and the dynamic problem studied is the mapping of user requests to these predetermined locations where VMs are hosted.…”

“…The placement of service-hosting nodes in MEC environments has been investigated as an offline static network planning problem 1 on how to optimally place a fixed number of micro-clouds in the network to minimize the network access latency [12], [13] assuming known / predicted unchanged load. On the other hand, work on the resource allocation in an online MEC system has focused on the dynamic routing of user requests to fixed clouds [12], [13], [14], [15].…”