How to use Amazon FSx for Lustre metrics - FSx for Lustre

How to use Amazon FSx for Lustre metrics

The metrics reported by Amazon FSx for Lustre provide information that you can analyze in different ways. The list following shows some common uses for the metrics. These are suggestions to get you started, not a comprehensive list.

How Do I Determine... Relevant Metrics (Dimension | Metric)

My file system's throughput?

SUM(DataReadBytes + DataWriteBytes)/Period (in seconds)

My file system's IOPS?

Total IOPS = SUM(DataReadOperations + DataWriteOperations + MetadataOperations)/Period (in seconds)

My file system's data compression ratio?

SUM(LogicalDiskUsage) / SUM(PhysicalDiskUsage)

If updates to my file system have been synchronized with my S3 bucket?

AutoExport | AgeOfOldestQueuedMessage

If updates to my S3 bucket have been synchronized with my file system?

AutoImport | AgeOfOldestQueuedMessage

Metric math: Data compression ratio

Using metric math, you can query multiple CloudWatch metrics and use math expressions to create new time series based on these metrics. You can visualize the resulting time series in the CloudWatch console and add them to dashboards. For more information on metric math, see Using metric math in the Amazon CloudWatch User Guide.

This metric math expression calculates the data compression ratio of your Amazon FSx for Lustre file system. To calculate this ratio, first get the sum statistic of the total logical disk usage (without compression), which is provided by the LogicalDiskUsage metric. Then divide that by the sum statistic of the total physical disk usage (with compression), provided by the PhysicalDiskUsage metric.

So if your logic is this: sum of LogicalDiskUsage Ă· sum of PhysicalDiskUsage

Then your CloudWatch metric information is the following.

ID Usable metric Statistic Period
m1 LogicalDiskUsage Sum 1 minute
m2 PhysicalDiskUsage Sum 1 minute

Your metric math ID and expression are the following.

ID Expression
e1 m1/m2

e1 is the data compression ratio.