Application of Generalized Deduplication Techniques in Edge Computing Environments
Application of Generalized Deduplication Techniques in Edge Computing Environments
复制标题
广义重复数据删除技术在边缘计算环境中的应用
DOI：
10.1007/978-3-031-28694-0_55
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Kurihara Jun
中科院分区：
文献类型：
--
作者：
Watanabe Ryu;Kubota Ayumu;Kurihara Jun
可以考虑在Beyond 5G中使用的移动的网络的用例之一是海量IoT环境，其中连接了许多低功耗和计算能力的IoT（物联网）终端。为了在这种环境下高效利用网络资源，需要压缩和减少大量物联网终端上传的数据量。在这项研究中，我们考虑使用边缘服务器在大规模物联网环境中进行数据压缩，假设多访问边缘计算（MEC）场景。特别是，我们考虑了“广义的压缩（GD）”的应用，这是一种基于重复删除的流数据压缩方法，近年来因其对物联网传感数据的轻量级和高效压缩而备受关注。基本GD算法假设一对一的流传输和接收。在本报告中，我们提出了GD算法的扩展，适用于一对多（边缘服务器和物联网终端）MEC环境，并具有更高效的性能。具体来说，我们调查字典建设GD利用在一对多的环境中，并显示所提出的算法的效率的基本评价。
One of the use cases of mobile networks that can be considered for use in Beyond 5G is a massive IoT environment where many IoT (Internet of Things) terminals with low power consumption and computing power are connected. In order to efficiently use network resources in this environment, it is necessary to compress and reduce the amount of data uploaded by a large number of IoT terminals. In this study, we consider data compression in a Massive IoT environment using edge servers, assuming a Multi-access Edge Computing (MEC) scenario. In particular, we consider the application of “Generalized Deduplication (GD)", a stream data compression method based on duplicate deletion, which has been attracting attention in recent years for its lightweight and efficient compression of IoT sensing data. The basic GD algorithm assumes one-to-one stream transmission and reception. In this report, we propose an extension of the GD algorithm that is suitable for one-to-multi (edge server and IoT terminals) MEC environments and has more efficient performance. Specifically, we investigate dictionary construction for the GD utilization in a one-to-multi environment and show a basic evaluation of the efficiency of the proposed algorithm.