Skip to main content
Splunk Lantern

Capacity utilization runway in Dell Isilon NAS

For most storage systems, you buy a certain amount of storage with sufficient capacity to last for some time in the future. A common future time frame could be as long as three years. Having a shorter time frame can be beneficial because it means less storage space is unused for less time. You might need to need to predict how much storage capacity you have left over time in your Dell Isilon (network attached storage) NAS when doing the following:

Prerequisites 

In order to execute this procedure in your environment, the following data, services, or apps are required:

Example

To properly manage your Isilon NAS, you want to measure capacity as percent used and be able to plot that metric over time. You also want to predict the upper and lower bounds of the percent used value as it trends into the future. This "capacity runway" metric will tell you how much time remains before you have to take action, such as removing unused data or adding capacity. 

To optimize the search shown below, you should specify an index and a time range. 

  1. Run the following search: 
sourcetype="emc:isilon:*" Cluster_Name="<name of cluster>" stats.key="ifs.percent.used"
| timechart span=1d p99(stats.value) AS ifs.percent.used
| predict ifs.percent.used AS forecast lower95=low upper95=high

Search explanation

The table provides an explanation of what each part of this search achieves. You can adjust this query based on the specifics of your environment.

Splunk Search Explanation

index=<isilon index>  

This would be the index name given in inputs. Conf when you set up the TA

sourcetype="emc:isilon:*" 

Search only EMC Isilon data.

Cluster_Name="<name of cluster>" 

Enter the cluster name here to limit the search to a specific cluster. You can also omit this filter entirely and get the metric for all the clusters. 

stats.key="ifs.percent.used"

Search all Isilon data for events containing the “ifs.percent.used” resource attribute, which indicates the percent of the Isilon file system (IFS) used for storage. This is the metric that indicates the capacity used as a percentage. 

| timechart span=1d p99(stats.value) AS ifs.percent.used

Graph the results values for the key “ifs.percent.used” in one day increments.

| predict ifs.percent.used AS forecast lower95=low upper95=high

Predict future CPU usage based on the previous numbers, using the default upper and lower confidence intervals of 95 percent.

Result

This search produces a chart that shows the actual value of ifs.percent.used over time, as well as the forecast of when in the future some value of ifs.percent.used will be seen. If that future value is large enough to take action, you will be able to plan ahead based on the time indicated by the forecast. 

  • Was this article helpful?