Yang Hyunsik, Kim Younghan
School of Electronic Engineering, Soongsil University, Seoul 06978, Korea.
Sensors (Basel). 2019 Jul 25;19(15):3276. doi: 10.3390/s19153276.
For many vertical Internet of Things (IoT) applications, the high availability is very important. In traditional cloud systems, services are usually implemented with the same level of availability in which the fault detection and fault recovery mechanisms are not aware of service characteristics. In IoT-cloud, various services are provided with different service characteristics and availability requirements. Therefore, the existing cloud system is inefficient to optimize the availability method and resources to meet service requirements. To address this issue, this paper proposes a high availability architecture that is capable of dynamically optimizing the availability method based on service characteristics. The proposed architecture was verified through an implementation system based on OpenStack, and it was demonstrated that the system was able to achieve the target availability while optimizing resources, in contrast with existing architectures that use predefined availability methods.
对于许多垂直物联网（IoT）应用而言，高可用性非常重要。在传统云系统中，服务通常以相同的可用性级别实现，其中故障检测和故障恢复机制并不了解服务特征。在物联网云环境中，各种服务具有不同的服务特征和可用性要求。因此，现有的云系统在优化可用性方法和资源以满足服务需求方面效率低下。为了解决这个问题，本文提出了一种高可用性架构，该架构能够基于服务特征动态优化可用性方法。通过基于OpenStack的实现系统对所提出的架构进行了验证，结果表明，与使用预定义可用性方法的现有架构相比，该系统在优化资源的同时能够实现目标可用性。
