Louis Imershein

Data deduplication and compression can increase the capacity of storage allowing users to make more efficient use of what they have and to make new, high-performance devices more affordable. Traditionally these data reduction technologies have been available in proprietary storage systems, but more recently they've made their way into open source with projects such as zfs and btrfs as well as with backup-specific applications such as OpenDedupe. Each of these has issues, either they are resource hogs, or they are designed to address limited use cases. The Virtual Data Optimizer (dm-vdo) is a new open source project from Red Hat that provides inline data reduction for primary storage to the Linux block storage stack. Its intended for use in active virtualization and or container workloads with minimal impact to performance and resources. This talk will introduce dm-vdo, describe how it works, resource requirements, effectiveness for different use cases and performance impact.