Sample Header Ad - 728x90

Database recommendation storing a growing set of recordings

-1 votes
1 answer
34 views
Suppose I have a large volume of recordings coming in from various sources. Encoded as JSON-lines, they might look like this:
{ "metric": "temperatureRange", "values": [ "new-york", "2023-02-06T11:42:09", "18.1", "24.6" ] }
{ "metric": "windSpeed", "values": [ "tokyo", "2023-01-06T09:21:30", "7.1" ] }
{ "metric": "humidity", "values": [ "london", "2023-01-06T09:21:30", "0.65" ] }
{ "metric": "windSpeed", "values": [ "london", "2023-02-04T12:54:55", "5.6" ] }
The data is non-relational (in the sense of no foreign keys) and I only ever want to append data. Recordings are never modified after writing. Notice that the values list is not of a fixed size. That said, I think I could choose a reasonable upper-bound of—say—16 values. In terms of querying, I want to be able to do look-ups like: * Find all rows where the metric is windSpeed * Find all rows where the metric is humidity and first value is "new-york" * Find all rows where metric is temperatureRange and the second value is "24.6" I can model this in a traditional RDBMS, however I don't think that is optimized for this use-case. This is because I anticipate a high volume of writes but only very simple queries. **What is a good database system for this type of work load?**
Asked by sdgfsdh (99 rep)
Feb 6, 2023, 11:52 AM
Last activity: Feb 6, 2023, 01:45 PM