Press Releases

Independent Benchmark Demonstrates First Trillion-edge Knowledge Graph – Stardog Recognized for Providing Sub-Second Query Times on Hybrid, Multicloud Data

Jun 8, 2021

New McKnight Consulting Group report proves the significant operational and performance gains users can achieve using Stardog’s massive knowledge graph

ARLINGTON, VA., June 8, 2021 — A new benchmark study from McKnight Consulting Group (MCG) unveiled the first demonstration of a massive knowledge graph that consists of materialized and virtual graphs spanning hybrid multicloud platforms. MCG affirmed that Stardog, the leading Enterprise Knowledge Graph platform provider, enabled users to create a one trillion-edge knowledge graph with sub-second query times without the need to store all the data in a single, centralized location. This is critical because today’s changing and proliferating IT environment requires data connections that don’t rely on centralization and/or the need for data movement.

The benchmark proved performance across vast amounts of distributed, real-world data and is the largest scale benchmark of its kind to date. Stardog’s ability to query data where it is stored, in its original source, without the need to make copies or move data, makes data lineage and traceability straightforward. Using the industry standard Berlin SPARQL Benchmark (BSBM), which does not repeat queries or allow results to be cached, makes the findings more reflective of real-world performance. Stardog’s unique virtualization capabilities also contributed to its unprecedented operational cost benefit.

The Stardog server used to conduct this benchmark costs $6.60/hour to operate. Even when accounting for the on-demand cloud pricing of the data stored in AWS Redshift and Azure SQL Server ($2/hour), materializing the same amount of data in a graph database would cost significantly more. This is because Stardog has made data location irrelevant to the enterprise, giving companies the ability to perform high-volume, realistic queries and accelerate knowledge discovery across a wide range of assets or processes without limitation. The full report is available for download https://info.stardog.com/trillion-triple-benchmark.

“This capability can usher in a new era where the knowledge graph is a powerful component of company profitability and competitive advantage,” said William McKnight, president of MKG. “In this report, we have shown that Stardog allows users to query trillion-edge graphs distributed over multiple data sources in different cloud providers. The combination of materialization and virtualization capabilities gives companies the option to store data in Stardog when needed but leave other data in its existing data store and to be queried on-demand. Average query execution times below one second show that performance at this scale is in line with fully materialized enterprise queries, as is the cost.” 

Conventional relational database management systems worked acceptably well when the enterprise data landscape was itself predominantly structured. But the world has changed. The emergence of IoT, the rise in unstructured data volume, increasing relevance of external data, and the trend towards hybrid multicloud environments are challenges that must be overcome with each new request for data. Data strategies centered around relational data systems are rarely sufficient anymore, especially as the requirements to connect data across the enterprise increase. 

Through a unique combination of graph, virtualization, and inference, Stardog’s knowledge graph technology is the key ingredient to transforming existing data infrastructure into a data fabric – a modern technology approach that is being hailed as the next generation data management solution. Enterprise data fabrics offer a new path forward as they weave together data from internal silos and external sources, create a seamless network of information and support the full gambit of the connected enterprise.

“Until recently, the predecessors to the modern data fabric have been data federation and virtualization technologies; however, a majority of these platforms have failed to deliver true inter-connectedness at scale with performance, because they are hampered by bottlenecks inherent in all databases and data stores in the query chain,” said Evren Sirin, CTO and co-founder of Stardog. “Instead of tackling the problem with another abstraction layer, graph model connects data so organizations can understand the data relationships, determine the prioritization of nodes in the relationship, and leverage visualization so it is easier for users to search, investigate, and analyze data, and expose patterns and trends by connecting diverse forms of connected knowledge.”

Additional Resources:

- Download the benchmark here

- View a recent McKnight Consulting Group blog