- Database purpose built to store time series data
- Data is sequence of data points over a time interval
- Examples:
- Sensor Data
- Weather Data
- System Monitoring Data
- Website Activity data
- Stock prices
- How is time series data unique?
- very high write throughput
- regular and irregular writes
- data needs to be highly compressed
- large range scans of many record
- write to latest time entry only
- native support for summaries, aggregation and rollups
- Look at some queries
- Daily stock price of AMZN
- Daily average temperature in San Francisco
- Unique website visits every month
- Highest revenue months in last 5 years
- Average uber ride prices by city
- Examples
- QuestDB
- Timescale DB
- Cassandra (NoSQL)
- Amazon timestream