The time data type
Mark Needham
ClickHouse 25.6 introduced the Time data type for storing time values. This video covers how to use the new Time and Time64 data types in your ClickHouse queries and tables.
What we'll cover:
- Creating Time values with casting and toTime() function
- Using Time64 for millisecond/nanosecond precision
- Required settings: enable_time_time64_type
- Practical example with running splits data
- Aggregating Time data (max, min, and workarounds for average/quantiles)
- Converting between Time types and Unix timestamps
We'll work through a real example using running data with moving time and elapsed time columns, showing how Time types make the data more readable compared to storing raw seconds.

User stories, Meetups
ClickHouse at DoorDash
How DoorDash monitors TCP and DNS traffic helping their observability team take care of their kubernetes infrastructure.

Open House
Open House NYC: Keynote
Keynote from Open House Roadshow NYC 2025.

Open House
Open House NYC: ClickHouse for AI/ML
Zach Naimon, Principal Product Manager at ClickHouse
Products
Resources
© 2025 ClickHouse, Inc. HQ in the Bay Area, CA and Amsterdam, NL.