Benchmarking Summary

This is a summary page of benchmarking results of popular analytics/OLAP databases that I’ve undertaken using the dataset 1.1 Billion Taxi Rides <>_. The dataset has 19 columns and takes up about 130 GB of disk space uncompressed. The dataset has been exported from google BigQuery to a google cloud bucket “yellow_taxi_trips” which gives a static copy of the public dataset (which can change any time on BigQuery) and also easy access it from google cloud VMs using gsutil. The export process breaks down the dataset into compressed csv files numbering ~1000 .gz files and places them on the bucket.

Important

This benchmarking is not promoting any technology as such and is an attempt to compare what is out there. Though these benchmarks used the same dataset, I wouldn’t consider any two as comparing apples to apples. The times recorded don’t necessarily reflect the top performance these systems are capable of. It’s likely each system could have performed faster if more effort was put into setup, configuration and/or better hardware was available. Except cloud databases such as BigQuery, the exercise was constrained by 24 vCPU limit within a zone on google cloud which even things out on CPU front. There are some cases where I re-ran a benchmark with more tuning to achieve a better result and in these cases only the fastest benchmark will be listed below. I’m interested in learning of any optimisations that can improve performance. Please open an issue <>_ if you’ve got any configuration settings to share.

The table is sorted by the average time across all queries:

images/benchmark.png