According to the OpenStack Foundation’s sixth and most recent user survey released just prior to this week’s OpenStack Summit in Tokyo, 62% of users selected Ceph RBD block storage for their OpenStack use cases, nearly three and more than four times the two closest alternatives, LVM (default) and NetApp, respectively. In production, a full 38% of respondents indicated that their OpenStack deployments depended on Ceph as their Cinder driver, with the same comparisons. A survey of the largest production clouds, those exceeding 1,000 cores, showed similar results, with 37% of users selecting Ceph RBD followed by NetApp at 12%. Interestingly enough, with 9% of respondents using GlusterFS in production, development & quality assurance, or proof of concept across all OpenStack deployments, more than 70% of OpenStack users are relying on block storage championed by Red Hat Storage.
Ceph’s reign in popularity comes as no surprise to us. Ceph’s architecture (including the block storage device) predates OpenStack and was actually the inspiration for its Cinder block storage abstraction layer. Ceph was specifically designed as an optimal, massively-scalable, software-defined storage system for cloud infrastructure. It provides unified infrastructure capabilities to support block, object, and file storage, tight integration with OpenStack’s modular architecture and key components for ephemeral and persistent storage, and massive scalability to manage petabytes of data. These factors along with Ceph’s flexible scale-out architecture make it an ideal option for OpenStack along with enterprise-grade solutions like Red Hat Enterprise Linux OpenStack Platform, Red Hat Cloud Infrastructure, and the Red Hat NFV Platform.
If you’re at the OpenStack Summit, please visit us at booth P7, and be sure to follow our Twitter feed for news and insight throughout the week. Also consider taking a free test drive of Red Hat Ceph Storage or Red Hat Gluster Storage today.