Clarify availability of ILM and data stream lifecycle #2532
Conversation
🔍 Preview links for changed docs |
manage-data/lifecycle/data-stream.md
Outdated
|
|
||
| * Data stream lifecycle is supported only for data streams and cannot be used with standalone indices. | ||
|
|
||
| * Like [{{ilm-cap}} ({{ilm-init}})](/manage-data/lifecycle/index-lifecycle-management.md), data stream lifecycle is supported in both {{ecloud}} and self-managed environments. Data stream lifecycle is less feature rich than {{ilm-init}} and is focused on simplicity. For a detailed comparison refer to [Data lifecycle](/manage-data/lifecycle.md). |
There was a problem hiding this comment.
I'd avoid "{{ecloud}} and self-managed environments" because we've found that this is not comprehensive or clear enough. to external users, "elastic cloud" could be "elastic cloud on kubernetes", which is technically a self-managed option. I'd indicate either that this is for deployments of the versioned stack, or list out all of the supported deployment types.
There was a problem hiding this comment.
Changed to Data stream lifecycle is supported for all deployment types on the versioned {{stack}} as well as for {{es-serverless}}.
Similarly, for ILM: {{ilm-init}} is available for all deployment types on the versioned {{stack}} but is not available for {{es-serverless}}.
manage-data/lifecycle/data-stream.md
Outdated
|
|
||
| * Data stream lifecycle is supported only for data streams and cannot be used with indices. | ||
|
|
||
| * Data stream lifecycle is supported for all deployment types on the versioned {{stack}} as well as for {{es-serverless}}. Compared with {{ilm-init}}, which is not available for {{serverless-short}}, data stream lifecycle is focused on simplicity, optimized for the most common lifecycle management needs. It enables you to configure the retention duration for your data and to optimize how the data is stored. For a detailed comparison of {{ilm-init}} and data stream lifecycle refer to [Data lifecycle](/manage-data/lifecycle.md). |
There was a problem hiding this comment.
this comparison information doesn't seem to fit here. I think we have that info up here but maybe it makes sense to have this somewhere else on this page too.
There was a problem hiding this comment.
I agree. I've just removed it.
|
|
||
| * **Data streams:** A [data stream](/manage-data/data-store/data-streams.md) acts as a layer of abstraction over a set of indices that contain append-only, time series data. You can configure {{ilm-init}} using a data stream as a single named resource, so that rollover and any other configured actions are performed on the data stream's backing indices automatically. | ||
|
|
||
| * {{ilm-init}} is available for all deployment types on the versioned {{stack}} but is not available for {{es-serverless}}. In a {{serverless-short}} environment, data stream lifecycle (see the following tip) is available as a data lifecycle option. |
There was a problem hiding this comment.
this method of referring to the tip in the body feels weird. I'd keep the link in this bullet body and adjust the note to center the serverless benefit.
There was a problem hiding this comment.
Good idea. I've moved the link into the main text and removed "see the following tip".
|
|
||
| * {{ilm-init}} is available for all deployment types on the versioned {{stack}} but is not available for {{es-serverless}}. In a {{serverless-short}} environment, data stream lifecycle (see the following tip) is available as a data lifecycle option. | ||
|
|
||
| :::{tip} |
There was a problem hiding this comment.
| :::{tip} | |
| :::{admonition} Simpler lifecycle management in Serverless environments |
| :::{tip} | ||
| {{ilm-init}} lets you automatically transition indices through data tiers according to your performance needs and retention requirements. This allows you to balance hardware costs with performance. {{es-serverless}} eliminates this complexity by optimizing your cluster performance for you. In a {{serverless-short}} environment, data stream lifecycle is available as a data management option. | ||
|
|
||
| {applies_to}`stack: ga` {applies_to}`serverless: ga` [Data stream lifecycle](/manage-data/lifecycle/data-stream.md) is a simpler lifecycle management tool optimized for the most common lifecycle management needs. It enables you to configure the retention duration for your data and to optimize how the data is stored, without hardware-centric concepts like data tiers. For a detailed comparison of {{ilm-init}} and data stream lifecycle refer to [Data lifecycle](/manage-data/lifecycle.md). |
There was a problem hiding this comment.
these applies tags don't feel right here. I'd skip them.
the benefit is really much greater for serverless environments and dsl is not powerful enough for all stack cases, so communicating that it's available in stack isn't super critical
can drop the link if it's in the bullet item above
| {applies_to}`stack: ga` {applies_to}`serverless: ga` [Data stream lifecycle](/manage-data/lifecycle/data-stream.md) is a simpler lifecycle management tool optimized for the most common lifecycle management needs. It enables you to configure the retention duration for your data and to optimize how the data is stored, without hardware-centric concepts like data tiers. For a detailed comparison of {{ilm-init}} and data stream lifecycle refer to [Data lifecycle](/manage-data/lifecycle.md). | |
| Data stream lifecycle is a simpler lifecycle management tool optimized for the most common lifecycle management needs. It enables you to configure the retention duration for your data and to optimize how the data is stored, without hardware-centric concepts like data tiers. For a detailed comparison of {{ilm-init}} and data stream lifecycle refer to [Data lifecycle](/manage-data/lifecycle.md). |
There was a problem hiding this comment.
Fixed in new commit
| * {{ilm-init}} is available for all deployment types on the versioned {{stack}} but is not available for {{es-serverless}}. In a {{serverless-short}} environment, data stream lifecycle (see the following tip) is available as a data lifecycle option. | ||
|
|
||
| :::{tip} | ||
| {{ilm-init}} lets you automatically transition indices through data tiers according to your performance needs and retention requirements. This allows you to balance hardware costs with performance. {{es-serverless}} eliminates this complexity by optimizing your cluster performance for you. In a {{serverless-short}} environment, data stream lifecycle is available as a data management option. |
There was a problem hiding this comment.
I think we need to double down on "ILM is not available in serverless BECAUSE" - splitting the info from the note has made that a little harder to parse
| ::: | ||
|
|
||
| ::::{important} | ||
| To use {{ilm-init}}, all nodes in a cluster must run the same version. Although it might be possible to create and apply policies in a mixed-version cluster, there is no guarantee they will work as intended. Attempting to use a policy that contains actions that aren’t supported on all nodes in a cluster will cause errors. |
There was a problem hiding this comment.
is this an availability concern or a prereq?
There was a problem hiding this comment.
I've moved this to the bottom of the intro paragraph.
|
@yetanothertw I applied your comments but some of them may have later been changed or removed as part of the latest review. Just so you don't think I ignored them. :-) |
Co-authored-by: shainaraskas <[email protected]>
3 participants
The scope of ILM and of data stream lifecycle should be called out more clearly, with respect to their supported deployment types and to the objects that each can manage.
This adds an "Availability" section to the top of the main ILM page and the main data stream lifecycle page.
Please see previews:
Rel: #190