Clarify product environment minimum for ECH #3307
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change | ||||
|---|---|---|---|---|---|---|
|
|
@@ -66,3 +66,12 @@ Scaling with {{ecloud}} is easy: | |||||
| * Or, if you prefer manual control, log in to the [{{ecloud}} Console](https://cloud.elastic.co?page=docs&placement=docs-body), select your deployment, select *Edit*, and either increase the number of zones or the size per zone. | ||||||
|
|
||||||
| The recommendation is to scale up the resources within a single zone until the cluster can take the full load (add some buffer to be prepared for a peak of requests), then scale out by adding additional zones depending on your requirements: two zones for High Availability, three zones for Fault Tolerance. | ||||||
|
|
||||||
| ## Minimum size recommendations for production use [ec-minimum-recommendations] | ||||||
|
|
||||||
| To ensure optimal performance and cluster stability in your production environment, we recommend adhering to the following minimum size guidelines. Deviating from these recommendations may lead to performance issues and cluster instability. For an enhanced user experience, consider planning your deployment capacity above these minimum recommendations, and adjust sizing based on your specific use case. | ||||||
|
|
||||||
| * **{{es}} nodes / instances**: For production systems, each {{es}} node / instance in your cluster should have at least 4 GB of RAM. | ||||||
|
||||||
| * **{{es}} nodes / instances**: For production systems, each {{es}} node / instance in your cluster should have at least 4 GB of RAM. | |
| * **{{es}} nodes / instances**: For production systems, each {{es}} node / instance in your cluster must have at least 4 GB of RAM. |
I think must is a stronger verb to use as it relays an obligation, an absolute requirement. Should conveys an expectation and could be interpreted as a lack of obligation.
There was a problem hiding this comment.
Thanks @yetanothertw but here IMHO we'd better say "should" because there's no legal or contractual obligation...
Customer can choose whatever size including the 1GB technical minimum to use and then run into trouble. (This is by design by cloud PM team. I have talked with them in the past)
I am a bit worried if we say "must" instead of "should", it could bring some unexpected discussion about these "obligations", e.g. customer may ask "why you said 'must use' but I don't see that in my contract?", etc and then it's hard for us to explain, because it's not a hard limit. (as again, by design by cloud PM).
(Also we don't separate production use and test use into different our cloud platform. So for testing purposes, it's really ok to use smaller, e.g. 1GB instances and accept the risk like data loss.)
So @yetanothertw @emrcbrn IMHO we should still say "should" instead of "must" so that the obligation sense is not that strong, and indicates that's a recommendation from Elastic, but not mandatory.
@yetanothertw hope this is clear and please let me know if you still strongly suggest we use "must". Happy to discuss in advance.
There was a problem hiding this comment.
Thank you for explaining that, @kunisen. I thought this was a hard requirement, so was trying to be more clear about it. It makes sense to use should when conveying an expectation or recommendation (and not an obligation).
You can also say something like: **{{es}} nodes / instances**: For production systems, we recommend that each {{es}} node / instance in your cluster has at least 4 GB of RAM.
Uh oh!
There was an error while loading. Please reload this page.