Time-Series Database
A time-series database is a system optimized for data that is indexed by time. Each record has a timestamp, and queries almost always filter or group by time. Examples include server metrics, IoT sensor readings, application events, and financial market ticks.
What makes time-series data special
Time-series data has a few properties that general databases handle poorly. It arrives constantly and in huge volume. It is almost always appended, rarely updated. And queries are nearly always scoped to a time range: the last hour, yesterday, this quarter.
A time-series database is tuned for exactly that shape. It ingests high volumes of timestamped records efficiently, partitions data by time so range queries are fast, and often compresses heavily because consecutive readings are similar. That combination is hard to get from a general-purpose database.
The category spans many use cases: infrastructure monitoring, industrial IoT, observability, and capital markets data all live or die on time-series performance.
How Arc handles Time-Series Database
Arc handles time-series workloads as a columnar analytical database. It ingests time-stamped data at high throughput, stores it as time-partitioned Parquet, and queries it with standard SQL. Unlike older time-series databases, your data stays in an open format you own.
Related terms
Arc is a high-performance columnar database. Open Parquet on storage you own, single Go binary, production-ready in 30 seconds.