Note: Multiple patents and patent applications cover the technology and processes described in this section. If you are interested in exploring licensing options—technology or intellectual property—please see our contact page.

ScaleDB’s patented technology is based on innovative data structures and index algorithms allowing performance and scalability to support the largest data sets with a unified approach to support heterogeneous views of the data.

ScaleDB’s architecture supports structured and semi-structured data, providing efficient and unified storage for relational and hierarchical (XML) data. ScaleDB can be licensed as an embedded solution for tools and applications. In 2007, the company will deliver a commercial version of ScaleDB that will replace native data storage engines in open source databases to dramatically upgrade their scalability. For more information please see our products section.


Technology Summary
ScaleDB’s technology is based on Patricia tries. Unlike conventional tries, ScaleDB utilizes layers of tries to allow balanced update and search. Within just 3 index layers, billions of data elements are addressed. The two first layers are less than 10MB in size and therefore in memory. The result is that search and update of the largest data sets requires only a single I/O.

ScaleDB is using the Tries not only to index the data but also to map the relationships making the index a self-describing index supporting open and evolving schema.

More on ScaleDB’s technology


White Papers
▪ The ScaleDB data structure
▪ A fast index to semi-structured data
▪ Indexing open schema
▪ Additional Technical Papers



[ Back to top ]