Transport For London (TFL) publish data on all journeys made by the 'Boris Bikes'. This data is useful for experimenting with Elasticsearch queries. As there is a high number of journeys, the dataset also makes a good training set for shard-related tasks.
An environment variable called CYCLE_CSV_PATH is required for Logstash to know where to check for source data. The value should be a directory and glob pattern that will match the CSV files from TFL.
Once this environment variable is set, invoke Logstash:
logstash -f tfl-cycle-journey-pipeline.conf
The default output is an Elasticsearch node on 10.0.200.101, as set up in my Elasticsearch training lab repository.
The source data is available from TFL. The bike usage data is in the usage-stats directory.
There is, generally, one file per week. Download as many files as you would like to ingest and place them into the path defined in $CYCLE_CSV_PATH. It may be best to focus on a certain month and year, and download several files for that period.