Graph Data Council

SemBench: A Benchmark for Semantic Query Processing Engines. arxiv.org/abs/2511.01716

We are hosting our 21st TUC meeting in Boston on Friday, September 4, co-located with VLDB. Join us for a day of talks on graph processing, including ontologies, graph query languages, and text to query generation. 🎟️ The registration and CfP are open: ldbcouncil.org/event/twenty-f…

📼 ➡️ ☁️ We migrated most of our datasets from tape storage to Cloudflare: ldbcouncil.org/post/datasets-… 🌊 We also released experimental DuckLake versions of the small SNB datasets (up to scale factor 10). These allow partial querying via DuckLake clients such as DuckDB.

Using this setup, the system also completed the audit on SF100 and 1000, with more than 125k operations per second on both scale factors. See the detailed results on our new website at ldbcouncil.org/benchmarks/snb…

Congratulations to the Huawei Cloud, whose Graph Engine Service achieved record-breaking results on the LDBC SNB Interactive workload's scale factor 300 data set. It delivered a performance of more than 139k operations per second on a single server, powered by Kunpeng 920 CPUs.

We have a few new domains! See our new blog post:

We also rolled out a new website. Head to ldbcouncil.org to check out our fresh design and content, including a podcast with GDC's chair, Henry Gabb.

LDBC has rebranded as the Graph Data Council to reflect our broad interest in graph data management technology. This still includes Linked Data and benchmarks but also extends to graph schema and standard query languages (GQL and SQL/PGQ).

📢 GDC's chair, Henry Gabb, gave an interview to GraphGeeks on our history, rebranding and future plans. Listen to/watch the podcast below! – Apple: podcasts.apple.com/us/podcast/gra… – Spotify: open.spotify.com/episode/0rWwe5… – YouTube: youtube.com/watch?v=LqWa4o…

🎡 Are you in London this Friday, November 21? If so, this is your chance to build new connections in the graph community at the GraphGeeks mixer – an informal community meeting after Connected Data London. Register at luma.com/wy2f5i0d

We have published the programme for our next Technical User Community meeting in London on September 6 (Saturday) after the VLDB conference. If you're in town and interested in graph technology, please register and join us: ldbcouncil.org/event/twentiet…

LDBC will hold the next Technical User Community meeting in London on September 6, after the VLDB conference. The call for talks is open – submit your work on graph data management: ldbcouncil.org/event/twentiet…

🔔 Ready for #VLDB2025? Secure your spot now! 🔗 Register here: vldb.org/2025/?info-reg… 📌 Each research/industrial paper requires an in-person author registration.

Last December, LDBC members met in London to decide on future leadership, governance, and finances. As a result, LDBC switched to a new membership fee structure and made several changes to its board of directors, including a new chair. Read more: ldbcouncil.org/post/ldbc-orga…

We enabled a new scale factor, SF3000, for the LDBC Social Network Benchmark's Interactive workload. The data set is available at #sf3000" target="_blank" rel="nofollow noopener">ldbcouncil.org/data-sets-surf…

We are hosting a half-day LDBC user meeting after Connected Data London #cdl24 See the agenda at: ldbcouncil.org/event/nineteen…

🌟 #NebulaGraph Enterprise v5.0 is here! Enjoy #graph projection for efficient subgraph analysis, multi-level permission management, and advanced query optimization. Secure, fast, and innovative. #GraphDB #NebulaGraph nebula-graph.io/posts/enterpri…

Finding rock-scissors-paper (A beats B beats C beats A) triangles is tricky with #SQL SQL/PGQ in the SQL:2023 standard & Oracle Database 23c simplifies this Use GRAPH_TABLE to find paths: MATCH ( A ) -[ WHERE a_win ]-> ( B ) -[ WHERE b_win ]-> ( C ) -[ WHERE c_win ]-> ( A )

We are holding an LDBC community meeting in London this Friday (2024-12-13) starting at 13.00, near the venue of @Connected_Data London. Check out the programme at ldbcouncil.org/event/nineteen…

📈 New audited LDBC results: TigerGraph broke the record on the Social Network Benchmark's BI workload on the scale factor 1,000 data set using @AMD EPYC 9355 CPUs. See the results at ldbcouncil.org/benchmarks/snb…