Hello HN,<p>We previously shared benchmark results showing write speeds of 1.4 million rows per second [1] and how we built this ingestion system. Over the last few months, we've developed functionality that supports geospatial data in our time series database. We decided to add geohashes to our type system along with language features to support handling this type.<p>We put a lot of thought into optimizing these features from a storage, performance, and usability perspective and we've put together a blog post [2] that gives a tour of the features including implementation details. We've also updated our live demo [3] which now includes an example data set of 250k objects and sample queries to run against this data set.<p>Given that it's the first time we support geodata in our database, we learned a lot about usage patterns and characteristics of spatial data. We would be happy to hear your thoughts on how we went about adding support for this, where we could improve, and other types of geospatial data we could including in future.<p>Thanks,
Vlad<p>[1] <a href="https://news.ycombinator.com/item?id=27411307" rel="nofollow">https://news.ycombinator.com/item?id=27411307</a><p>[2] <a href="https://questdb.io/blog/2021/10/04/geospatial-timeseries-demo" rel="nofollow">https://questdb.io/blog/2021/10/04/geospatial-timeseries-dem...</a><p>[3] <a href="https://demo.questdb.io/" rel="nofollow">https://demo.questdb.io/</a>