Performance analysis of containerized applications on local and remote storage Conference

Xu, Q, Awasthi, M, Malladi, KT et al. (2017). Performance analysis of containerized applications on local and remote storage .

cited authors

  • Xu, Q; Awasthi, M; Malladi, KT; Bhimani, J; Yang, J; Annavaram, M

authors

abstract

  • Docker containers are becoming the mainstay for deploying applications in cloud platforms, having many desirable features like ease of deployment, developer friendliness, and lightweight virtualization. Meanwhile, storage systems have witnessed tremendous performance boost through recent innovations in the industry such as Non-Volatile Memory Express (NVMe) and NVMe Over Fabrics (NVMf) standards. However, the performance of docker containers on these highspeed contemporary SSDs has not yet been investigated. In this paper, we first present a characterization of the performance impact among a wide variety of the available storage options for deploying Docker containers and provide the configuration options to best utilize the high performance SSDs.We then provide the first of its kind characterization results of a Dockerized NoSQL database on an NVMe-over-fabrics prototype and show that its performance matches closely to that of direct attached storage. Finally, we provide experimental results on scaling the performance of NVMf to multiple nodes and present the challenges and projections for future storage system design.

publication date

  • January 1, 2017