Graph management and algorithmics are two of the hottest topics in top international data management conferences. As a by-product of some of the findings from those research topics, graph management technologies have flourished in the last 5 years. Sparksee (formerly called DEX) was born as such more than 8 years ago, being one of the precursors of graph management technologies. At the time, the seminal research work was done by DAMA-UPC research group at BarcelonaTech, which spun out to Sparsity Technologies a few years later.
Sparksee was born as the product of intense research on the representation of graphs for high performance management. The seminal paper on Sparksee dates back to 2007, where an analysis and comparison of Sparksee was done (read the seminal paper here). In line with this, Sparksee was designed to have a small software footprint and high data compression, allowing for the execution of complex graph queries on databases with billions of objects in off-the-shelve computing devices. Sparksee makes an intensive use of bitmaps allowing for the use of simple logic operations and high data locality to solve graph analytics. All the resulting technology from our research has been patented internationally.
As a consequence of the interesting field that the seminal paper unwrapped, Sparksee is currently supporting research on related areas that will surely help us improve & secure our product on the following versions, such as community search algorithms, benchmarking, performance analysis, graph applications, query languages, graph algebras and query optimisation.
Keep tuned to the next chapters where we are going to talk a little bit more about all the research areas that Sparsity is working right now with very promising results.
Image courtesy of David Castillo Dominici / FreeDigitalPhotos.net