Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
[dupe] InfluxDB v0.9.0 released with developer and production support available (influxdb.com)
16 points by pauldix on June 11, 2015 | hide | past | favorite | 6 comments


InfluxDB CEO here. Happy to answer any questions!


Hi and thanks for the new release!

With the new distinction between tags and fields, the documentation isn't clear about how to use them. I assume tags are indexed, but that you can still query by field, and that there are no limitations?

We're basically just throwing documents (containing nested documents) at InfluxDB, with no predefined schema, so we don't know beforehand what users will be performing ad-hoc queries on; it's completely dynamic in the UI. Someone, perhaps it was you, mentioned in another thread that InfluxDB only supported 255 tag names per series, which we'd probably exceed with our schema. At the same time we'd like to benefit from the performance we'd presumably get from using tags.

PS. This page is giving me an error "NoSuchKey": http://influxdb.com/docs/v0.9/concepts/aggregate_functions.h...


There isn't a limitation on the number of tags per data point. Although I assume at some point it'll become a performance bottleneck.

The limitation you're thinking of is the number of fields. Currently, you can only have up to 255 fields defined in a measurement.


I see. What kind of data structure does you use for indexing tags?


Looks like someone beat you to it Paul, more questions probably over here https://news.ycombinator.com/item?id=9700578


Thank you Paul!




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: