[
https://issues.apache.org/jira/browse/CASSANDRA-15618?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17069011#comment-17069011
]
Jordan West commented on CASSANDRA-15618:
-----------------------------------------
Thanks [~rustyrazorblade]. Great to have a doc for this. I appreciate the the
cleanup in the compaction docs as well even if it was a bit separate. Comments
on the new doc below:
Read ahead:
* While the section mentions it, a bit more attention could be paid to the
small reads case. The intro text primarily focuses on hot data set size and
small reads is mentioned only in the 4KB justification.
* A table like the token count section could be a nice way to present the
recommended defaults
* The last sentence of the intro paragraph says “available RAM is greater
than the amount of data on disk”. Should this say “available RAM is greater
than the hot dataset”?
Compression:
* With zStd now included do we want to make a recommendation on the
algorithm to choose? Cc/ [~djoshi]
Compaction:
* This section could use some links since the recommendation is to read up
on the strategies
Encryption:
* Consider explaining why you its painful to switch later to drive this
point home. Also consider a call out or some emphasis.
Racks and Snitch:
* Consider emphasizing (bold/italic/callout) the first paragraph
* “On prem” -> “on-premise”
* Code highlighting of Ec2Snitch
> add docs for production recommendations
> ---------------------------------------
>
> Key: CASSANDRA-15618
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15618
> Project: Cassandra
> Issue Type: Task
> Components: Documentation/Website
> Reporter: Jon Haddad
> Assignee: Jon Haddad
> Priority: Normal
> Labels: pull-request-available
> Time Spent: 10m
> Remaining Estimate: 0h
>
> We have some inline comments in our yaml, but I think it would be helpful if
> we had a more lengthy section in the docs detailing the settings people
> should be using in production. Some of these are unrelated to the yaml
> itself (like reducing read ahead), and putting them in the YAML doesn’t make
> much sense. There are also other settings (like compression chunk length)
> that don’t fit there. Let’s add a page under getting started called
> Production Recommendations.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
