Resource Provisioning in Edge Computing for Latency-Sensitive Applications

Abstract

Low-Latency IoT applications such as autonomous vehicles, augmented/virtual reality devices and security applications require high computation resources to make decisions on the fly. However, these kinds of applications cannot tolerate offloading their tasks to be processed on a cloud infrastructure due to the experienced latency. Therefore, edge computing is introduced to enable low latency by moving the tasks processing closer to the users at the edge of the network. The edge of the network is characterized by the heterogeneity of edge devices forming it; thus, it is crucial to devise novel solutions that take into account the different physical resources of each edge device. In this paper, we propose a resource representation scheme, allowing each edge device to expose its resource information to the supervisor of the edge node through the mobile edge computing application programming interfaces proposed by European Telecommunications Standards Institute. The information about the edge device resource is exposed to the supervisor of the EN each time a resource allocation is required. To this end, we leverage a Lyapunov optimization framework to dynamically allocate resources at the edge devices. To test our proposed model, we performed intensive theoretical and experimental simulations on a testbed to validate the proposed scheme and its impact on different system's parameters. The simulations have shown that our proposed approach outperforms other benchmark approaches and provides low latency and optimal resource consumption.

Service Function Chaining in MEC: A Mean-Field Game and Reinforcement Learning Approach

“…Without loss of generality, we consider the placement of the VNFs within a given EN regardless of the specific assignment between the VNFs and the devices of the corresponding EN. Moreover, in this paper, we leverage our previous works [20], [21] on resource representation to allow the EN to exchange their resource availability status and we consider the overall available resources of the EN. Each VNF requires an amount of resource to perform its tasks in terms of computing, storage, and transmission.…”

“…As mentioned before, the ENs are formed by aggregating different heterogeneous edge devices such as servers, edge routers, small base stations, or even eNodeBs and gNodeBs. In [20], [21], we proposed a resource representation model that allows these heterogeneous devices to exchange information about their resource capabilities in terms of type of the operations that can handle. For instance, the edge routers are more into network operations unlike edge servers that are more into data processing and edge learning.…”

Traffic-Aware Horizontal Pod Autoscaler in Kubernetes-Based Edge Computing Infrastructure

“…Edge computing is a new paradigm that overcomes the inherent limitations of cloud computing by distributing edge nodes with computing resources closer to IoT devices [8], [9]. By processing data at local edge nodes without transferring them to the cloud, edge computing can sufficiently reduce the response time to satisfy the requirements of time-critical applications [10].…”

A Deep Reinforcement Learning Approach for Service Migration in MEC-enabled Vehicular Networks

“…Multi-access edge computing (MEC) is envisioned as a key component for fifth-generation (5G) ultra-reliable low-latency communications (uRLLC) services, alongside software-defined networking (SDN) [5] technology. On the one hand, MEC can be leveraged as an emerging computational paradigm that provides efficient computational capabilities to vehicles deployed in close proximity to MEC servers while ensuring a low latency.…”