Open source database startup QuestDB said on Wednesday it raised $12 million in a Series A funding round to help the vendor grow its technology and go-to-market efforts.
Founded in San Francisco in 2019, QuestDB has developed an open source time series database technology. While some time series databases have required users to learn new query languages to use them, QuestDB has focused on using SQL as the primary method to query its database.
In this Q&A, Nicolas Hourcard, CEO and co-founder of QuestDB, explains the use cases for time series databases and where the vendor is headed.
Why are you now raising a Series A?
Nicolas Hourcard: We want to double down on wider developer adoption. We're seeing a lot of usage from all sorts of companies in different industries and a lot of developers are joining our community.
Raising capital enables us to build more features in the open source products. It also helps to build the fully hosted cloud that a lot of people ask for pretty much all the time, because they don't want to manage QuestDB in the cloud themselves.
What makes QuestDB different than other time series databases?
Hourcard: Our differentiation is a combination of performance, simplicity through the use of SQL, and the fact that we're open source. We actually built the entire stack for ourselves and we didn't reuse any code that been available out there. The algorithms have been optimized in a way to extract as much performance as possible from the hardware.
Nicolas HourcardCEO and co-founder, QuestDB
So the background of my co-founder and CTO Vlad Ilyushchenko is high-frequency electronic trading. In that world, you tend to do high performance code with zero garbage collection Java. The code approach looks more like C++ than Java, funnily enough, and you can actually get a lot of performance through this type of code.
What do you see as the use cases for a time series database?
Hourcard: Time series data is really everywhere. Actually, we have discovered more use cases as time has progressed.
So initially, we came from the world of electronic trading, so financial services is an obvious use case, where we get a lot of market data coming in at very high rates. Time series is a really good fit there because you really need performance.
Then we saw a lot of companies doing asset tracking for fleets. So any sort of vehicles such as cars, planes, ships, just tracking the position over time. We do have a geospatial element as well, that we released recently in order to do that very quickly.
Another use case is sensor data coming from any sort of machine or manufacturing at large. We're also seeing an increasing number of data science use cases that involved machine learning models to predict future behaviors based on historical patterns. That is all time series data.
How do you find that organizations typically start using the QuestDB time series database?
Hourcard: We see developers that are already using a time series database and typically they are looking for alternatives because of the performance aspect of things.
Then there are other developers that might be using Oracle or PostgreSQL or some other traditional relational databases. For a startup, at the beginning they might start by using the traditional database as they figure things out and don't have a lot of data.
But at some point, the scale means that it's going to sort of break down, especially if it's time series data because the primary characteristic of time series data is that there is a lot of it being produced. So typically we see companies coming to us saying their existing setup doesn't work and they are looking for alternatives.
There are also users that pull in QuestDB for entirely new projects that involve a lot of data and real-time components.
Editor's note: This interview has been edited for clarity and conciseness.