←back to thread

The Fastest Redis Ever

(redis.io)
82 points lsferreira42 | 9 comments | 18 Nov 24 21:39 UTC | HN request time: 1.243s | source | bottom
Show context
pcthrowaway ◴[21 Nov 24 07:19 UTC] No.42201918[source]▶
The inclusion of Redis timeseries is huge!

This was available for a long time as an extension as part of Redis Stack, but most hosted Redis providers don't make extensions available (I'm assuming due to nuances in Redis's not-quite-open licensing).

If cloud providers which include Redis are now going to include this, it opens up a lot of potential for my use case.

replies(2): >>42202211 #>>42207333 #
1. jeltz ◴[21 Nov 24 08:21 UTC] No.42202211[source]▶
When do you want to store your time series data in Redis and not a database like TimescaleDB or Clickhouse which is optimized for storage on disk and analytics queries?
replies(1): >>42202223 #
2. rcarmo ◴[21 Nov 24 08:24 UTC] No.42202223[source]▶
Likely when it's small enough to keep in RAM and you want to do some sort of on-the fly aggregation/correlation.
replies(1): >>42202269 #
3. jeltz ◴[21 Nov 24 08:31 UTC] No.42202269[source]▶
Then you can usually just store it in the memory of your application. No need to complicate your stack by running another service.
replies(5): >>42202304 #>>42202310 #>>42202952 #>>42205750 #>>42208946 #
4. _joel ◴[21 Nov 24 08:39 UTC] No.42202304{3}[source]▶
Isn't this just RocksDB?
replies(1): >>42202718 #
5. vrnvu ◴[21 Nov 24 10:06 UTC] No.42202718{4}[source]▶
Or DuckDB
6. jeremycarter ◴[21 Nov 24 10:43 UTC] No.42202952{3}[source]▶
Some large IoT systems need ephemeral timeseries.
replies(1): >>42203612 #
7. jeltz ◴[21 Nov 24 12:22 UTC] No.42203612{4}[source]▶
Which you can store just fine in-memory in a normal data structure. And if you need advanced query capabilities or a query planner there is DuckDB. Using Redis seems like you get most of the disadvantages of having to run a whole database with few of the advantages.
8. pcthrowaway ◴[21 Nov 24 16:11 UTC] No.42205750{3}[source]▶
When you need to be able to retrieve the timeseries data for some period of time, storing it in the application memory doesn't really work since the application will restart whenever updates are made.

Also, redis timeseries offers the ability to downsample to some defined period which is really handy (and afaik isn't really provided by other timeseries databases) as well as set a retention policy.

9. rcarmo ◴[21 Nov 24 21:39 UTC] No.42208946{3}[source]▶
your application can consist of multiple processes.