Grid (Scale-Out) Index
Sooner or later, any vendor delivering storage solutions will run into the resource limitations of single-server deployment when trying to expand scale and performance. This issue is resolved by introducing multi-node scale-out approaches to storing data. Many vendors deliver solutions that scale across multiple nodes, but very few can deliver deduplication capabilities that provide inline performance and span these environments.
Permabit’s Grid Index technology allows vendors to take advantage of multiple Deduplication Index instances, each running on an independent server. This scale-out approach enables multi-node scaling of deduplication in terms of both capacity and performance.
When compared to other solutions in use today, Permabit’s Grid Index technology enables the greatest expandability, supporting linear growth of performance and capacity.
How it works
A grid consists of multiple clients that are embedded in each storage ingest point, connected to multiple Deduplication Indexes in a grid configuration over the network. In a grid deployment scenario, each client performs its own content-level segmentation and hashing and then routes the resulting content fingerprints to the appropriate Deduplication Index. Each index server stores a balanced portion of the index.
For multi-tenant environments, the Grid Index supports multiple namespaces; separate independent storage domains (e.g., file systems, logical volumes, or systems) can share a single global Albireo Index. In this way, both unified storage pools and large quantities of independent storage volumes benefit from the performance and scalability of the technology.