Providing High Availability in Cloud Storage by Decreasing Virtual Machine Reboot Time

Shehbaz Jaffer, Mangesh Chitnis, and Ameya Usgaonkar, NetApp Inc.

Abstract: 

A Virtual Storage Architecture (VSA) is a storage controller deployed as a virtual machine on a server with a hypervisor. The advantage of VSA is to leverage shared data storage services without procuring additional storage hardware, which is a cost effective solution. In case of VSA, high availability (HA) is achieved by restarting the failed virtual machine on an event of a software failure. Rebooting the VSA is a slow operation thereby reducing the overall service availability. In this paper, we describe the challenges and approaches taken to decrease reboot time of VSA to achieve High Availability. We have been able to reduce the VSA reboot time by 18% using our optimizations. We also ex-plore the changes required in Journal based File systems for efficient operation in the Cloud.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {187156,
author = {Shehbaz Jaffer and Mangesh Chitnis and Ameya Usgaonkar},
title = {Providing High Availability in Cloud Storage by Decreasing Virtual Machine Reboot Time},
booktitle = {10th Workshop on Hot Topics in System Dependability (HotDep 14)},
year = {2014},
address = {Broomfield, CO},
url = {https://www.usenix.org/conference/hotdep14/workshop-program/presentation/jaffer},
publisher = {USENIX Association},
month = oct
}
Download
Jaffer PDF