This repository contains the CSV data of SRBench [1]
Check out the release section to download the latest dataset version. Two different serializations are available:
- CSV: the folder structure resembles the original data format.
charley/
└── extracted/
├── sensor_X/
│ └── AirTemperature/
│ └──file.csv
├── sensor_Y/
└── sensor_Z/
- JSON-PG [2]: the data is organized in JSON files. Each file is named with a UNIX timestamp and contains the JSON-PG serialization of the events that occurred at the specified time instant.
charley/
└── pg-json/
├── 1091923500.json
└── 1091923800.json
To generate JSON-PG representation:
- Build docker image:
docker build -t json-pg .
- Run the docker image mounting the input and output volumes as follows:
docker run \
-v ${PWD}/input/charley:/data/input \
-v ${PWD}/output/charley:/data/output \
json-pg
- The output folder contains the PG-JSON representation of the data.
[1] Zhang Y., Duc P.M., Corcho O., Calbimonte JP. (2012) SRBench: A Streaming RDF/SPARQL Benchmark. In: Cudré-Mauroux P. et al. (eds) The Semantic Web – ISWC 2012. ISWC 2012. Lecture Notes in Computer Science, vol 7649. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-35176-1_40
[2] Chiba, H., Yamanaka, R. and Matsumoto, S., 2019. Property Graph Exchange Format. arXiv preprint https://arxiv.org/abs/1907.03936.