Data storage growth is both a reality and a burden for today’s enterprises.
With the move to digital, big data and, data intensive workloads, enterprises are storing an increasing amount of data, growing 30-40 percent per year, according to Vanson Bourne LTD. Much of this growth comes from applications running on cloud platforms such as OpenStack, and these applications have unique requirements that are difficult to meet using traditional storage solutions.
In fact, the Vanson Bourne survey indicates that 70 percent of IT decision makers don’t believe their current storage solutions can handle these next generation workloads. The storage requirements for these new cloud-native applications often include:
Which brings us to the Ceph open source storage project. Ceph was developed by a community of engineers and researchers who needed a distributed object store and file system. Over time, block storage capabilities were added to Ceph which has helped it to become a primary storage option for OpenStack and a valuable software defined storage solution for enterprises.
Red Hat, the primary contributor to the Ceph project, packages a stable version of Ceph and supports it through their Red Hat Ceph Storage offering, which is quickly becoming a leading software defined storage solution in the industry. Rackspace is a big believer in the Ceph project and currently offers a managed Ceph storage offering with both our Rackspace Private Cloud powered by OpenStack, aka RPC-O, and Rackspace Private Cloud powered by Red Hat, or RPC-R.
Recently, Red Hat announced Red Hat Ceph Storage 2, which supports the 10.2 community version release of Ceph, aka the Jewel release. Recognizing the importance of this release for our joint customers, Rackspace supports Red Hat Ceph Storage 2 as part of our Rackspace Private Cloud Powered by Red Hat managed OpenStack offering.
This release is the most stable version yet of Ceph and provides a number of features and benefits to enterprise users.
Exabyte level scalability
- Scale-out architecture that can grow a cluster from a few to thousands of nodes
- Automatic rebalancing designed to seamlessly handle failures and provide data distribution throughout the cluster
- Rolling software upgrades that can be performed in phases with no or minimal downtime
- S3 and Swift compatible for cloud integration with protocols used by Amazon Web Services and the OpenStack Object Storage project
- RESTful for managing cluster and object storage functions programmatically
- Integration with Active Directory, LDAP, and KeyStone v3 for authentication and authorization
- Policies to limit access at the pool, user, bucket or data level
- Encryption that can be implement at the cluster-level
Reliability and Availability
- Striping, erasure coding or replication across nodes to provide data durability, high availability and high performance
- Dynamic block resizing to expand or shrink Ceph block devices with zero downtime
- Storage policies to configure placement of data to reflect SLAs, performance requirements and failure domains
- Snapshots of entire pool or individual block devices
- Global clusters for creating a global namespace for object users with read and write affinity to local clusters
- Multisite replication for disaster recovery or archiving
- Copy-on-write cloning to enable instant provisioning of tens or hundreds of virtual machine images
- In-memory client-side caching to enhance client I/O using a hypervisor cache
- Server-side journaling to accelerate the write performance of data by serializing writes
- Support for commodity hardware to tailor the optimal price/performance mix of standard servers and disk to each workload
- Thin provisioning for creating sparse block images that enables over-provisioning the cluster.
Rackspace will take a careful and phased approach to support for Red Hat Ceph Storage 2 with RPC-R. Much as we are doing with Red Hat OpenStack Platform and with Red Hat CloudForms, Rackspace will support the majority of features in Ceph 2 with the goal of adding support for additional capabilities over time. In this initial release, we will not be supporting erasure coding or global distributed clusters. The goal is to make sure our engineers can offer full managed services of a feature before we officially support that specific capability.
Most importantly, RPC-R customers, with this new 2 release, can gain the key features and benefits outlined earlier and do so without taking on the burden of having to deploy, operate, monitor and troubleshoot their OpenStack or OpenStack storage environment.
RPC-R customers will be able to consume their private cloud storage, be it object or block, with the same power and ease found with public cloud solutions such as AWS EBS and S3 or Rackspace Cloud Block Storage and Cloud Files. This managed solution continues the work that Rackspace and Red Hat are doing together to create the best private cloud and software defined storage offerings for the enterprise.