summaryrefslogtreecommitdiffstats
path: root/database/README.md
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2023-05-08 16:27:08 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2023-05-08 16:27:08 +0000
commit81581f9719bc56f01d5aa08952671d65fda9867a (patch)
tree0f5c6b6138bf169c23c9d24b1fc0a3521385cb18 /database/README.md
parentReleasing debian version 1.38.1-1. (diff)
downloadnetdata-81581f9719bc56f01d5aa08952671d65fda9867a.tar.xz
netdata-81581f9719bc56f01d5aa08952671d65fda9867a.zip
Merging upstream version 1.39.0.
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'database/README.md')
-rw-r--r--database/README.md42
1 files changed, 27 insertions, 15 deletions
diff --git a/database/README.md b/database/README.md
index becd4165f..eb7081620 100644
--- a/database/README.md
+++ b/database/README.md
@@ -1,13 +1,3 @@
-<!--
-title: "Database"
-description: "The Netdata Agent leverages multiple, user-configurable time-series databases that use RAM and/or disk to store metrics on any type of node."
-custom_edit_url: "https://github.com/netdata/netdata/edit/master/database/README.md"
-sidebar_label: "Database"
-learn_status: "Published"
-learn_topic_type: "Tasks"
-learn_rel_path: "Setup"
--->
-
# Database
Netdata is fully capable of long-term metrics storage, at per-second granularity, via its default database engine
@@ -42,13 +32,13 @@ The default mode `[db].mode = dbengine` has been designed to scale for longer re
for parent Agents in the _Parent - Child_ setups
The other available database modes are designed to minimize resource utilization and should only be considered on
-[Parent - Child](https://github.com/netdata/netdata/blob/master/docs/metrics-storage-management/how-streaming-works.mdx) setups at the children side and only when the
+[Parent - Child](https://github.com/netdata/netdata/blob/master/docs/metrics-storage-management/enable-streaming.md) setups at the children side and only when the
resource constraints are very strict.
So,
- On a single node setup, use `[db].mode = dbengine`.
-- On a [Parent - Child](https://github.com/netdata/netdata/blob/master/docs/metrics-storage-management/how-streaming-works.mdx) setup, use `[db].mode = dbengine` on the
+- On a [Parent - Child](https://github.com/netdata/netdata/blob/master/docs/metrics-storage-management/enable-streaming.md) setup, use `[db].mode = dbengine` on the
parent to increase retention, a more resource efficient mode like, `dbengine` with light retention settings, and
`save`, `ram` or `none` modes for the children to minimize resource utilization.
@@ -68,7 +58,7 @@ Metrics retention is controlled only by the disk space allocated to storing metr
CPU required by the agent to query longer timeframes.
Since Netdata Agents usually run on the edge, on production systems, Netdata Agent **parents** should be considered.
-When having a [**parent - child**](https://github.com/netdata/netdata/blob/master/docs/metrics-storage-management/how-streaming-works.mdx) setup, the child (the
+When having a [**parent - child**](https://github.com/netdata/netdata/blob/master/docs/metrics-storage-management/enable-streaming.md) setup, the child (the
Netdata Agent running on a production system) delegates all of its functions, including longer metrics retention and
querying, to the parent node that can dedicate more resources to this task. A single Netdata Agent parent can centralize
multiple children Netdata Agents (dozens, hundreds, or even thousands depending on its available resources).
@@ -144,8 +134,30 @@ Put the above lines in your boot sequence (`/etc/rc.local` or equivalent) to hav
### Monitoring Kernel Memory de-duplication performance
-Netdata will create charts for kernel memory de-duplication performance, like this:
+Netdata will create charts for kernel memory de-duplication performance, the **deduper (ksm)** charts can be seen under the **Memory** section in the Netdata UI.
+
+#### KSM summary
+
+The summary gives you a quick idea of how much savings (in terms of bytes and in terms of percentage) KSM is able to achieve.
+
+![image](https://user-images.githubusercontent.com/24860547/199454880-123ae7c4-071a-4811-95b8-18cf4e4f60a2.png)
+
+#### KSM pages merge performance
+
+This chart indicates the performance of page merging. **Shared** indicates used shared pages, **Unshared** indicates memory no longer shared (pages are unique but repeatedly checked for merging), **Sharing** indicates memory currently shared(how many more sites are sharing the pages, i.e. how much saved) and **Volatile** indicates volatile pages (changing too fast to be placed in a tree).
+
+A high ratio of Sharing to Shared indicates good sharing, but a high ratio of Unshared to Sharing indicates wasted effort.
+
+![image](https://user-images.githubusercontent.com/24860547/199455374-d63fd2c2-e12b-4ddf-947b-35371215eb05.png)
+
+#### KSM savings
+
+This chart shows the amount of memory saved by KSM. **Savings** indicates saved memory. **Offered** indicates memory marked as mergeable.
+
+![image](https://user-images.githubusercontent.com/24860547/199455604-43cd9248-1f6e-4c31-be56-e0b9e432f48a.png)
-![image](https://cloud.githubusercontent.com/assets/2662304/11998786/eb23ae54-aab6-11e5-94d4-e848e8a5c56a.png)
+#### KSM effectiveness
+This chart tells you how well KSM is doing at what it is supposed to. It does this by charting the percentage of the mergeable pages that are currently merged.
+![image](https://user-images.githubusercontent.com/24860547/199455770-4d7991ff-6b7e-4d96-9d23-33ffc572b370.png)