A Time Series Analysis and Persistence Framework for Global Multicloud

Abstract

In the global multicloud environment, time series database is an essential service in large-scale monitoring or IoT data persistence and analysis. However, due to various factors such as huge amounts of data, large numbers of concurrent reading and writing devices and complex network environments, conventional time series databases are often difficult to achieve unified management through global multicloud. This paper tried to put forward a persistence and analysis framework for global multicloud distributed time series databases, which could achieve unified analysis and distributed data persistence in multicloud environments and support scale-out, concurrent writing as well as high performance query and analysis. In addition, the framework presented is able to optimize data lifecycle management, query routing and cost, along with forming a good integration with monitoring ecosystem.