Earticle

전술 엣지 클라우드 환경에서, 취약한 링크의 안정성에 따라 중앙 클라우드 또는 엣지 클라우드에서 동작하고 있는 서비스를 클라우드간에 이동하는 서비스 이동성 관리가 필요하다. 특히, 전술망은 일반망에 대해 열악한 링크 성능 을 가지고 있어, 이동성 관리는 빠르고 효율적으로 수행되어야 한다. 서비스의 위치 변경에 일관된 동일한 도달성을 제공하기 위해, 본 논문에서는 Location/ID 분리 프로토콜인 LISP(Location/ID Separation Protocol)을 기반 으로 하는 서비스 이동성 관리 기법을 제안한다. 제안하는 기법은 기존의 중앙화된 매핑 시스템에 대한 접근 지연을 줄이기 위해 계층적인 LISP 매핑 시스템을 구성함으로써, 전술망의 링크 안정성이 취약한 환경에서도 서비스 ID에 대한 매핑 정보를 하위 단말에게 제공할 수 있도록 하였다. 서비스 이동성 관리를 위해, 전술 엣지 클라우드 토폴로 지를 기반으로 수평적 이동성과 수직적 이동성의 두 가지 서비스 이동성 시나리오를 정의하고 이에 따른 세부 이동성 관리 절차를 설계하였다.

In the tactical edge cloud environment, service mobility management which moves a service between clouds is required to overcome the less stability of links. In particular, since the tactical network has very poor link performance with respect to the general internet network, service mobility management should be performed in fast and efficient ways. In order to provide consistent and consistent reachability to service location changes, this paper proposes a service mobility management approach based on the Location/ID Separation Protocol (LISP). Especially, in this paper, we design a hierarchical LISP mapping system to reduce access delays to the existing centralized mapping system, so that user can get service ID from mapping system at the edge cloud even if the link between to backbone cloud were unstable. For service mobility management, we defined two service mobility scenarios, horizontal mobility and vertical mobility and detailed mobility management procedures accordingly.

Recently, with the development of IoT technology, data has increased, and problems with the centralized cloud computing method are appearing. As an alternative to this, Edge Computing, a distributed cloud method that processes data close to the edge of the network where data is generated, is utilized. On the other hand, as the number of containers on one host increases, container management becomes difficult, and accordingly, container orchestration technology capable of configuring and managing a large number of containers is required. In this paper, we measure, compare, and analyze the time to transmit sensor data to the DB server of each Kubernetes cluster by building a multicluster infrastructure using Kubernetes, which is the most used container orchestration tool.

In edge computing, most procedures, including data collection, data processing, and service provision, are handled at edge nodes and not in the central cloud. This decreases the processing burden on the central cloud, enabling fast responses to end-device service requests in addition to reducing bandwidth consumption. However, edge nodes have restricted computing, storage, and energy resources to support computation-intensive tasks such as processing deep neural network (DNN) inference. In this study, we analyze the effect of models with single and multiple local exits on DNN inference in an edge-computing environment. Our test results show that a single-exit model performs better with respect to the number of local exited samples, inference accuracy, and inference latency than a multi-exit model at all exit points. These results signify that higher accuracy can be achieved with less computation when a single-exit model is adopted. In edge computing infrastructure, it is therefore more efficient to adopt a DNN model with only one or a few exit points to provide a fast and reliable inference service.

Cloud computing technology based on centralized high-performance computing has brought about major changes across the information technology industry and led to new paradigms. However, with the rapid development of the industry and increasing need for mass generation and real-time processing of data across various fields, centralized cloud computing is lagging behind the demand. This is particularly critical in emerging technologies such as autonomous driving, the metaverse, and augmented/virtual reality that require the provision of services with ultralow latency for real-time performance. To address existing limitations, distributed and edge cloud computing technologies have recently gained attention. These technologies allow for data to be processed and analyzed closer to their point of generation, substantially reducing the response times and optimizing the network bandwidth usage. We describe distributed and edge cloud computing technologies and explore the latest trends in their standardization.