Conserv started in early 2019 to bring better preventive care to the world’s collections. They serve anyone charged with the long-term preservation of physical objects, from paintings and sculptures to books, architecture, and more. Conserv collects environmental data, sensor health data, data from events, locations, and data from human observations.
Conserv collects tens of thousands of data points each day, and their users think about their metrics and trends over the years, not days. The main database criteria were scale, performance, and the ability to tap into the SQL ecosystem. Conserv’s first proof of concept used Elasticsearch to store reads, but they knew it wouldn’t be a permanent solution. They also looked at InfluxDB and Amazon Timestream.
Moving to Timescale enabled Conserv to spend more time providing end-user value and focusing less on managing the database. Conserv benefits from Timescale’s built-in time-series features like time_bucket, time_bucket_backfill, histograms, last observation carried forward, and a handful of others.
The combination of a performant, scalable repository for our time-series data, the core SQL features we know and trust, and the built-in time-series functionality. This is where we’ve gotten the most value out of Timescale.
Nathan McMinn, CTO and Co-founder at Conserv