big dataYarcData, a NoSQL data engine vendor subsidiary of Cray Inc., has announced a major upgrade to its Urika Big Data analytics appliance, based on graph analytics. It also announced alliances with BI tools vendors Centrifuge Systems and TIBCO as well as Hadoop-based analytics leader Cloudera.

The new version of Urika integrates a list of standard interfaces including W3C industry standards SPARQL and RDF, JDBC (Java DataBase Conectivity), JSON (Java Script Object Notation), and Apache Jena. That plus the new alliances allow non-IT tech business and financial analysts to use Centrifuge Visual Network Analytics and TIBCO Spotfire to submit and refine analysis queries to Urika interactively.

The upgrades include a 400X performance boost to Urika’s ability to respond to analysis queries. They also include new workflow feaatures to support data discovery as an iterative process of hypotheses validation. The new hypothesis validation monitor capability provides users with fine grained detail across the lifecycle of hypotheses and validation results. It allows analysts to reuse queries, analyze performance, and investiage query results to develop important insights faster.

Urika helps automate the discovery of unknown relationships and non-obvious patterns in diverse data sets without the need for pre-modeling, partitioning, or knowing all the questions in advance. Urika includes graphc-optimized hardware providing up to 512 Tbytes of global shared memory, massively multithreaded graph processors supporting 128 threads/processor, highly scalable I/O, and data ingest rates of up to 350 Tbutes/hour, and an RDV/SPARQL database optimized for the underlying hardware.