Querying Time Series

Query patterns for time series data often use SQL constructs:

SELECT ...      -- Can include aggregate functions (e.g., SUM, AVG)
FROM ...
WHERE ...       -- Often uses BETWEEN for time ranges
ORDER BY ...

Common Query Patterns

• Compare periods: Analyze changes between two time ranges.
• Summarize windows: Compute aggregates for defined intervals (e.g., hourly, daily).
• granularity plays an important role in both.

Windowing Concepts

Sliding Window:

• A continuous set of rows grouped by a specific granularity.
• The window slides by one row, creating overlapping windows.
• Useful for monitoring changes over time. Think moving averages.

Tumbling Window:

• Moves forward by the full window size (no overlap).
• Often used for logical groupings of time (e.g., full days, weeks).

Performance & Optimization

• For older data or longer-range queries, use coarser granularity to improve performance.
• Use EXPLAIN or ANALYZE to optimize SQL queries.