Skip to content

Commit 6665521

Browse files
jillian-maroketjillian-maroket
authored
Add "Calculation of Resource Metrics" to Monitoring section (#719)
* Add Calculation of Resource Metrics file Signed-off-by: Jillian <[email protected]> * Edit section properties in JSON file Signed-off-by: Jillian <[email protected]> * Edit section properties in other JSON files Signed-off-by: Jillian <[email protected]> --------- Signed-off-by: Jillian <[email protected]>
1 parent 475e0b2 commit 6665521

File tree

8 files changed

+567
-18
lines changed

8 files changed

+567
-18
lines changed

‎docs/monitoring/_category_.json

+2-6
Original file line numberDiff line numberDiff line change
@@ -1,10 +1,6 @@
11
{
22
"position": 12,
33
"label": "Monitoring",
4-
"collapsible": false,
5-
"collapsed": false,
6-
"link": {
7-
"type": "doc",
8-
"id": "harvester-monitoring"
9-
}
4+
"collapsible": true,
5+
"collapsed": true
106
}

‎docs/monitoring/calculation-resource-metrics.md

+187
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,187 @@
1+
---
2+
sidebar_position: 2
3+
sidebar_label: Calculation of Resource Metrics
4+
title: "Calculation of Resource Metrics"
5+
keywords:
6+
- cluster resource metrics
7+
- host resource metrics
8+
- reserved resource
9+
- calculation
10+
---
11+
12+
<head>
13+
<link rel="canonical" href="https://docs.harvesterhci.io/v1.4/monitoring/calculation-resource-metrics"/>
14+
</head>
15+
16+
Harvester calculates resource metrics using data that is dynamically collected from the system. Host-level resource metrics are calculated and then aggregated to obtain the cluster-level metrics.
17+
18+
You can view resource-related metrics on the Harvester UI.
19+
20+
- **Hosts** screen: Displays host-level metrics
21+
22+
![Host-level resource metrics](/img/v1.5/monitoring/host-resource-usage.png)
23+
24+
- **Dashboard** screen: Displays cluster-level metrics
25+
26+
![Cluster-level resource metrics](/img/v1.5/monitoring/cluster-resource-usage.png)
27+
28+
## CPU and Memory
29+
30+
The following sections describe the data sources and calculation methods for CPU and memory resources.
31+
32+
- Resource capacity: Baseline data
33+
- Resource usage: Data source for the **Used** field on the **Hosts** screen
34+
- Resource reservation: Data source for the **Reserved** field on the **Hosts** screen
35+
36+
### Resource Capacity
37+
38+
In Kubernetes, a `Node` object is created for each host. `.status.allocatable.cpu` and `.status.allocatable.memory` represent the available CPU and memory resources of a host.
39+
40+
Example:
41+
42+
```
43+
# kubectl get nodes -A -oyaml
44+
apiVersion: v1
45+
items:
46+
- apiVersion: v1
47+
kind: Node
48+
metadata:
49+
..
50+
management.cattle.io/pod-limits: '{"cpu":"12715m","devices.kubevirt.io/kvm":"1","devices.kubevirt.io/tun":"1","devices.kubevirt.io/vhost-net":"1","memory":"17104951040"}'
51+
management.cattle.io/pod-requests: '{"cpu":"5657m","devices.kubevirt.io/kvm":"1","devices.kubevirt.io/tun":"1","devices.kubevirt.io/vhost-net":"1","ephemeral-storage":"50M","memory":"9155862208","pods":"78"}'
52+
node.alpha.kubernetes.io/ttl: "0"
53+
..
54+
name: harv41
55+
resourceVersion: "2170215"
56+
uid: b6f5850a-2fbc-4aef-8fbe-121dfb671b67
57+
spec:
58+
podCIDR: 10.52.0.0/24
59+
podCIDRs:
60+
- 10.52.0.0/24
61+
providerID: rke2://harv41
62+
status:
63+
addresses:
64+
- address: 192.168.122.141
65+
type: InternalIP
66+
- address: harv41
67+
type: Hostname
68+
allocatable:
69+
cpu: "10"
70+
devices.kubevirt.io/kvm: 1k
71+
devices.kubevirt.io/tun: 1k
72+
devices.kubevirt.io/vhost-net: 1k
73+
ephemeral-storage: "149527126718"
74+
hugepages-1Gi: "0"
75+
hugepages-2Mi: "0"
76+
memory: 20464216Ki
77+
pods: "200"
78+
capacity:
79+
cpu: "10"
80+
devices.kubevirt.io/kvm: 1k
81+
devices.kubevirt.io/tun: 1k
82+
devices.kubevirt.io/vhost-net: 1k
83+
ephemeral-storage: 153707984Ki
84+
hugepages-1Gi: "0"
85+
hugepages-2Mi: "0"
86+
memory: 20464216Ki
87+
pods: "200"
88+
```
89+
90+
### Resource Usage
91+
92+
CPU and memory usage data is continuously collected and stored in the `NodeMetrics` object. Harvester reads the data from `usage.cpu` and `usage.memory`.
93+
94+
Example:
95+
96+
```
97+
# kubectl get NodeMetrics -A -oyaml
98+
apiVersion: v1
99+
items:
100+
- apiVersion: metrics.k8s.io/v1beta1
101+
kind: NodeMetrics
102+
metadata:
103+
...
104+
name: harv41
105+
timestamp: "2024-01-23T12:04:44Z"
106+
usage:
107+
cpu: 891736742n
108+
memory: 9845008Ki
109+
window: 10.149s
110+
```
111+
112+
### Resource Reservation
113+
114+
Harvester dynamically calculates the resource limits and requests of all pods running on a host, and updates the information to the annotations of the `NodeMetrics` object.
115+
116+
Example:
117+
118+
```
119+
management.cattle.io/pod-limits: '{"cpu":"12715m",...,"memory":"17104951040"}'
120+
management.cattle.io/pod-requests: '{"cpu":"5657m",...,"memory":"9155862208"}'
121+
```
122+
123+
For more information, see [Requests and Limits](https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/#requests-and-limits) in the Kubernetes documentation.
124+
125+
## Storage
126+
127+
Longhorn, which is the default Container Storage Interface (CSI) driver of Harvester, provides storage management features such as distributed block storage and tiering.
128+
129+
### Reserved Storage in Longhorn
130+
131+
Longhorn allows you to specify the percentage of disk space that is not allocated to the default disk on each new Longhorn node. The default value is "30". For more information, see [Storage Reserved Percentage For Default Disk](https://longhorn.io/docs/1.8.0/references/settings/#storage-reserved-percentage-for-default-disk) in the Longhorn documentation.
132+
133+
Depending on the disk size, you can modify the default value using the [embedded Longhorn UI](../troubleshooting/harvester.md#access-embedded-rancher-and-longhorn-dashboards).
134+
135+
### Data Sources and Calculation
136+
137+
Harvester uses the following data to calculate metrics for storage resources.
138+
139+
- Sum of the `storageMaximum` values of all disks (`status.diskStatus.disk-name`): Total storage capacity
140+
- Sum of the `storageAvailable` values of all disks (`status.diskStatus.disk-name`): Data source for the **Used** field on the **Hosts** screen
141+
- Sum of the `storageReserved` values of all disks (`spec.disks`): Data source for the **Reserved** field on the **Hosts** screen
142+
143+
Example:
144+
145+
```
146+
# kubectl get nodes.longhorn.io -n longhorn-system -oyaml
147+
148+
apiVersion: v1
149+
items:
150+
- apiVersion: longhorn.io/v1beta2
151+
kind: Node
152+
metadata:
153+
..
154+
name: harv41
155+
namespace: longhorn-system
156+
..
157+
spec:
158+
allowScheduling: true
159+
disks:
160+
default-disk-ef11a18c36b01132:
161+
allowScheduling: true
162+
diskType: filesystem
163+
evictionRequested: false
164+
path: /var/lib/harvester/defaultdisk
165+
storageReserved: 24220101427
166+
tags: []
167+
..
168+
status:
169+
..
170+
diskStatus:
171+
default-disk-ef11a18c36b01132:
172+
..
173+
diskType: filesystem
174+
diskUUID: d2788933-8817-44c6-b688-dee414cc1f73
175+
scheduledReplica:
176+
pvc-95561210-c39c-4c2e-ac9a-4a9bd72b3100-r-20affeca: 2147483648
177+
pvc-9e83b2dc-6a4b-4499-ba70-70dc25b2d9aa-r-4ad05c86: 32212254720
178+
pvc-bc25be1e-ca4e-4818-a16d-48353a0f2f96-r-c7b88c60: 3221225472
179+
pvc-d9d3e54d-8d67-4740-861e-6373f670f1e4-r-f4c7c338: 2147483648
180+
pvc-e954b5fe-bbd7-4d44-9866-6ff6684d5708-r-ba6b87b6: 5368709120
181+
storageAvailable: 77699481600
182+
storageMaximum: 80733671424
183+
storageScheduled: 45097156608
184+
region: ""
185+
snapshotCheckStatus: {}
186+
zone: ""
187+
```

‎static/img/v1.5/monitoring/cluster-resource-usage.png

71.4 KB
Loading

‎static/img/v1.5/monitoring/host-resource-usage.png

102 KB
Loading

‎versioned_docs/version-v1.3/monitoring/_category_.json

+2-6
Original file line numberDiff line numberDiff line change
@@ -1,10 +1,6 @@
11
{
22
"position": 12,
33
"label": "Monitoring",
4-
"collapsible": false,
5-
"collapsed": false,
6-
"link": {
7-
"type": "doc",
8-
"id": "harvester-monitoring"
9-
}
4+
"collapsible": true,
5+
"collapsed": true
106
}

0 commit comments

Comments
 (0)