Chorograph are pleased to reveal that they have completed the first LDBC SNB Audited Benchmark (Linked Data Benchmark Council, Social Network Benchmark).The team at Chorograph audited FMA Technologies flagship product TuGraph 1.10 (formerly known as Lightgraph) using the SNB Interactive Workload
Ben Steer of Chorograph, also a member of the LDBC SNB Taskforce, played a leading role in the development of the most successful and enduring industry standard benchmarks from the LDBC. LDBC SNB is an industrial and academic initiative, formed by principal actors in the field of graph-like data management. Its goal is to define a framework where different graph-based technologies can be fairly tested and compared, that can drive the identification of systems’ bottlenecks and required functionalities, and can help researchers open new frontiers in high-performance graph data management.
The result of the LDBC SNB Benchmark can be found in full on the LDBC website: http://ldbcouncil.org/benchmarks/snb
We carried out the tests using the LDBC SNB Interactive Workload in addition to the first extensive ACID tests for graph databases. Chorograph managed the tests end to end utilising Amazon Web services, this was led by certified and independent Benchmark Auditor Marton Bur. All programs and scripts used in the process are public. The final result was reviewed and approved by the LDBC Executive Committee which is chaired by many key organisations in the Graph Technology space.
The LDBC SNB Interactive Workload test process starts with data generation, and goes through data loading, program execution and result verification. SNB simulates relationship maintenance, personal information maintenance, post and reply operations, and forum maintenance in a social networking site. At the same time, it also performs analysis tasks such as relationship queries and relationship recommendations. During the test, multiple clients send requests to the server concurrently. The server needs to process the request and return the result within a given time. Requests that do not respond within the given time are considered timeout. If the proportion of timeout requests is greater than 5%, the test result is deemed invalid.
The SNB also contains read and write tasks, tests for simple and complex queries, and specifies the response time of the system, so it can better reflect the system The overall performance. Coupled with strict auditing of the test process, the credibility of the results is guaranteed to the utmost extent. At the same time, SNB also announced the procedures and scripts used in the testing process, as well as the detailed results produced during the testing process, further ensuring the reproducibility of the test.
If you are currently a graph database vendor that would like more information on the LDBC SNB Audited Benchmark process then feel free to reach out to email@example.com