snb

We are happy to annonunce new audited results for the SNB Interactive workload, achieved by the open-source GraphScope Flex system.

The current audit of the system has broken several records:

  • It achieved 130.1k ops/s on scale factor 100, compared to the previous record of 48.8k ops/s.
  • It achieved 131.3k ops/s on scale factor 300, compared to the previous record of 48.3k ops/s.
  • It is the first system to successfully complete the benchmark on …

LDBC SNB provides a data generator, which produces synthetic datasets, mimicking a social network’s activity during a period of time. Datagen is defined by the charasteristics of realism, scalability, determinism and usability. More than two years have elapsed since my last technical update on LDBC SNB Datagen, in which I discussed the reasons for moving the code to Apache Spark from the MapReduce-based Apache Hadoop implementation and the …

Speeding Up LDBC SNB Datagen

Tags:
DATAGEN , SNB

LDBC’s Social Network Benchmark [4] (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 …

LDBC and Apache Flink

Tags:
FLINK , DATAGEN , SNB

Apache Flink [1] is an open source platform for distributed stream and batch data processing. Flink’s core is a streaming dataflow engine that provides data distribution, communication, and fault tolerance for distributed computations over data streams. Flink also builds batch processing on top of the streaming engine, overlaying native iteration support, managed memory, and program optimization.

Flink offers multiple APIs to process data …

In this post we will look at running the LDBC SNB on Virtuoso.

First, let’s recap what the benchmark is about:

  1. fairly frequent short updates, with no update contention worth mentioning

  2. short random lookups

  3. medium complex queries centered around a person’s social environment

The updates exist so as to invalidate strategies that rely too heavily on precomputation. The short lookups exist for the sake of realism; after all, an …