Investing in QuestDB, the fastest open source time series database

There’s some justified skepticism about building a global infrastructural tech company out of Europe. There’s a limited number of local tech giants to acquire infra startups. In general, the tech sector is less developed and contributes much less to the open-source movement. Investors tend to come from financial and consulting rather than tech backgrounds and therefore favour infrastructural startups less.

I do believe, however, that there is an opportunity, at least in the database space. QuestDB is perfectly positioned to capture it for several reasons: the right angle/problem, an authentic team, an adequate business model and engineering approach.

Time series data grows and spreads across various industries and sectors. For example, data from the trading floor and manufacturing lines are organised in this manner. Moreover, both associated sectors are growing. Fintech is everywhere. Coinbase, Robinhood, and others multiply the volume of transactions by opening trading to retail customers. Internet of things is about to climb the slope of enlightenment and become a real thing.

Seeing something before it becomes a real thing is a rare gift from curiosity and doing things that others may consider wired. Nic and Vlad have it. Nic comes from a rare crowd of investment bankers who worked with data and applied quantitative approaches to the M&As. Vlad was tinkering with the tech stack of the largest banks for ages, going as deep as FPGAs. I could not imagine a better team to build a database.

The authenticity and passion of QuestDB founders allowed them to embark on the open-source route successfully (GitHub ⭐️⭐️⭐️ are growing!). Something you could hardly do without enthusiastically enjoying the space you are in. The OSS is the only way to build a global infra company out of Europe, I believe. It is, probably, the best way to speed up the customer development process, that conservative European corporations slow down. The OSS approach is also great for growing faster and going global.

The passion and deep expertise led the team to execute the full-stack approach and develop various database building blocks in-house. The team has built the entirety of the database, such as the storage engine from the ground up, leveraging unconventional principles and techniques from low-latency trading. There is no single third-party dependency in the entire codebase, and the algorithms are optimized to better extract performance from the hardware QuestDB runs onto.

QuestDB stores data in time-based arrays rather than leaning on trees (B-tree or LSM-tree). Here’s an article about the database ingestion performance using the Time-Series Benchmarking Suite against open-source alternatives – InfluxDB, TimescaleDB, and Clickhouse.

I’m sure there are more reasons to love QuestDB and the team behind it. I just did not see them in 2020 when I was making my first angel investments (QuestDB was #2). I’m looking forward to seeing more good news from the team in the years to come.

I’ll continue to share my angel investing memos, to get updates follow me on Twitter 🐦