The DSS block volume client exposes one or multiple volumes as a block devices on a linux host. These block devices can be formatted and used as if they were locally attached disks. The DSS Block Volume Client includes intelligent caching technologie that caches most frequently accessed data blocks locally on the host. A combination of solid state and hard disk drives can be deployed for the cache, providing the necessary I/O performance for primary data operation.
The client stores all data blocks in the Distributed Storage System backend, which guarantees data availability, scalability and automated management.
By offering this block device client, B-Virtual is the first to combine the required performance for primary storage with the high availability and ease of management of a Distributed Storage System.
Features
Unlimited Snapshots
The DSS block volume client allows you to make an unlimited number of snapshots on any volume. DSS volume snapshots are simple point in time operations, which introduce no overhead and no data copies.
Clones
Any snapshot can be cloned and exposed as an identical volume with full read/write operations.
Clone migration
Any clone can be exposed on a DSS block volume client which is different than the block volume client where the originating volume is exposed. As such, a read/write clone can be created wherever there is access to the distributed storage system backend.
Volume failover
DSS block volume drivers support block replication to eliminate the risk of data loss in case the server where the volume is exposed would fail.
Benefits
Ease of integration
The block volume client exposes standard linux volumes. These can be immediately formatted with your filesystem of choice to integrate in your existing environment. Optionally these volumes can be exposed over iSCSI for IP SAN access.
Granular capacity planning
The block volume client uses the DSS distributed storage system as backend. By default, volumes are thin provisioned as no capacity on the DSS dirstibured storage system is pre-allocated. Only when the volume is being used, capacity in the backend will be consumed for as many blcoks that have been written. This allows for granular capacity planning.
Flexible resource planning
Multiple volumes can be exposed on a single DSS block volume client, sharing a common local cache. If at a certain point the performance of a single block volume client would become insufficient for the workload, you can easily migrate a selection of volumes to other block volume clients. This allows for flexible resource planning based on the system load.
Disaster recovery
The Distributed Storage System allows storing data redundantly across multiple locations. As the block volume client is writing its data to this DSS backend, the volume data will be available across all locations. Using the volume fail over capabilities or the clone roaming feature of the DSS block volume client, you can failover or roam volumes to any location where the DSS storage data is accessible.