Skip to content
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.

Commit 8412870

Browse files
committedJul 8, 2021
add cache monitoring article
1 parent 27e6b10 commit 8412870

File tree

7 files changed

+80
-11
lines changed

7 files changed

+80
-11
lines changed
 

‎articles/hpc-cache/hpc-cache-manage.md

Lines changed: 2 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -4,7 +4,7 @@ description: How to manage and update Azure HPC Cache using the Azure portal or
44
author: ekpgh
55
ms.service: hpc-cache
66
ms.topic: how-to
7-
ms.date: 06/28/2021
7+
ms.date: 07/08/2021
88
ms.author: v-erkel
99
---
1010

@@ -245,14 +245,6 @@ $
245245

246246
---
247247

248-
## Cache metrics and monitoring
249-
250-
The overview page shows graphs for some basic cache statistics - cache throughput, operations per second, and latency.
251-
252-
![screenshot of three line graphs showing the statistics mentioned above for a sample cache](media/hpc-cache-overview-stats.png)
253-
254-
These charts are part of Azure's built-in monitoring and analytics tools. Additional tools and alerts are available from the pages under the **Monitoring** heading in the portal sidebar. Learn more in the portal section of the [Azure Monitoring documentation](../azure-monitor/essentials/monitor-azure-resource.md#monitoring-in-the-azure-portal).
255-
256248
## View warnings
257249

258250
If the cache goes into an unhealthy state, check the **Warnings** page. This page shows notifications from the cache software that might help you understand its state.
@@ -269,5 +261,5 @@ Kinds of warnings you might see here include:
269261

270262
## Next steps
271263

272-
* Learn more about [Azure metrics and statistics tools](../azure-monitor/index.yml)
264+
* [Monitor the cache with statistics](metrics.md)
273265
* Get [help with your Azure HPC Cache](hpc-cache-support-ticket.md)

‎articles/hpc-cache/index.yml

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -39,8 +39,10 @@ landingContent:
3939
url: hpc-cache-add-storage.md
4040
- text: Mount clients
4141
url: hpc-cache-mount.md
42-
- text: Manage and monitor the cache
42+
- text: Manage the cache
4343
url: hpc-cache-manage.md
44+
- text: View cache metrics
45+
url: metrics.md
4446
- linkListType: concept
4547
links:
4648
- text: Plan the aggregated namespace
188 KB
Loading
198 KB
Loading
24.6 KB
Loading

‎articles/hpc-cache/metrics.md

Lines changed: 73 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,73 @@
1+
---
2+
title: Azure HPC Cache metrics and monitoring
3+
description: How to see statistics about your Azure HPC Cache
4+
author: ekpgh
5+
ms.service: hpc-cache
6+
ms.topic: how-to
7+
ms.date: 07/08/2021
8+
ms.author: v-erkel
9+
---
10+
11+
# Cache metrics and monitoring
12+
13+
The Azure Portal has several built-in visualizations that you can use to monitor the performance of your Azure HPC Cache.
14+
15+
This article explains where to find the statistics visualizations, and what they show.
16+
17+
There are four HPC Cache portal pages that show cache performance data:
18+
19+
* **Overview**
20+
* **Metrics**
21+
* **Cache report**
22+
* **Client status**
23+
24+
Open the **Metrics**, **Cache report**, and **Client status** pages from the **Monitoring** menu in the sidebar. The cache's **Overview** page is the top item on the sidebar menu.
25+
26+
![screenshot of the Monitoring menu on the left of the Azure portal interface for HPC Cache.](media/monitoring-menu.png)
27+
28+
## Overview page
29+
30+
The cache's **Overview** page shows graphs for some basic cache statistics - cache throughput, operations per second, and latency. The graphs are at the bottom of the page, underneath the **Essentials** section.
31+
32+
![screenshot of three line graphs showing the statistics mentioned above for a sample cache](media/hpc-cache-overview-stats.png)
33+
34+
These charts are part of Azure's built-in monitoring and analytics tools. Learn more about these capabilities from the [Azure Monitoring documentation](../azure-monitor/essentials/monitor-azure-resource.md#monitoring-in-the-azure-portal).
35+
36+
## Metrics page
37+
38+
The HPC Cache **Metrics** page is another standard Azure feature. Follow the tips on the page to create your own charts and to learn more about the features.
39+
40+
## Cache report
41+
42+
The **Cache report** is a customized set of charts that show your cache's used space, free space, and cache recycle rate.
43+
44+
By default, the page shows aggregated data for all of the cache's storage targets, but you can use the **Storage target** control to show the cache consumption data for one particular storage target instead. You also can customize the time range to show.
45+
46+
![screenshot of the Cache report page.](media/cache-report.png)
47+
48+
* **Cache used space** is the amount of space in your cache storage that's in use storing files that client machines have requested. If you selected a single storage target, it shows only the space used for files from that back-end storage system.
49+
50+
* **Cache free space** is the available capacity remaining in cache storage.
51+
52+
* **Cache recycle rate** represents the rate at which old files are being removed from the cache to make room for newly requested files. For a single storage target, this calculation only includes files stored on that target.
53+
54+
## Client status
55+
56+
The **Client status** page lists the client machines that are connected to your Azure HPC Cache.
57+
58+
![screenshot of the Client status page.](media/client-status.png)
59+
60+
You can customize the table with filters to show a specific client address or address range, a specific storage target, or an individual cache mount address.
61+
62+
The connection types shown in the table have to do with the client connections to the storage target:
63+
64+
* `azure_https_backend` is a secure client connection to a standard Azure Blob storage system.
65+
* `mount_backend` is an NFS connection to either a hardware NFS system or an NFS-enabled blob container.
66+
* `nfs_backend` is similar to the `mount_backend` connection, but represents a separate connection used by an NFS-internal remote procedure call instead of one initiated by the client mount operation.
67+
68+
These values can be useful when debugging connection issues between clients and the HPC Cache.
69+
70+
## Next steps
71+
72+
* Learn more about [Azure metrics and statistics tools](../azure-monitor/index.yml)
73+
* [Manage the cache](hpc-cache-manage.md)

‎articles/hpc-cache/toc.yml

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -51,6 +51,8 @@ items:
5151
href: hpc-cache-mount.md
5252
- name: Manage the cache
5353
href: hpc-cache-manage.md
54+
- name: Metrics and monitoring
55+
href: metrics.md
5456
- name: Use optional settings
5557
items:
5658
- name: Configure optional settings

0 commit comments

Comments
 (0)
Please sign in to comment.