Slides:

• Part1: LDBC/SNB intro + SNB Programming Challenge
• Part2: SNB in depth + Programming Challenge Solutions

The material presented in the EDBT summer school is directly based on a number of papers on graph data management systems bencharking:

• The Linked Data Benchmark Council: a Graph and RDF industry benchmarking effort - SIGMOD Record 2014 - a high-level overview of the LDBC project: methods, goals, results.
• The LDBC Social Network Benchmark: Interactive Workload - SIGMOD 2015 - decribes all relevant aspects of the Interactive Workload of the LDBC Social Network Benchmark.
• Graphalytics: A Big Data Benchmark for Graph-Processing Platforms - GRADES 2015 - high-level description of the benchmark that tests SNB on a collection of graph analysis algorithms.
• Parameter Curation for Benchmark Queries - TPCTC 2014 - introduces the problem of finding stable substitution paramameter for correlated query workloads
• S3G2: a Scalable Structure-correlated Social Graph Generator - TPCTC 2012 - a detailed paper about the graph generator which eventually evolved into the SNB datagen
• How community-like is the structure of synthetically generated graphs? - GRADES 2014 - investigates the quality of certain graph generators (LDBC vs RMAT) by comparing six graph metric distributions with results obtained from real-world graphs.
• TPC-H Analyzed: Hidden Messages and Lessons Learned from an Influential Benchmark - TPCTC 2013 - disclaimer: not about graph database benchmarks. However, very relevant nevertheless and defines and illustrates the concept of choke-points benchmark development in the context of the influential TPC-H data warehousing benchmark.

The EDBT 2015 Summer School course on Graph Benchmarking and the SNB Challenge were developed by Peter Boncz and Hannes Mühleisen from the Database Architectures research group of CWI, specifically for the Amsterdam Data Science initiative.