Azure DCR Rules

Question

If we write a DCR to ingest VMInsights Metrics into Log Analytics workspace as shown below:

"dataSources": { "performanceCounters": [ { "streams": [ "Microsoft-InsightsMetrics" ], "samplingFrequencyInSeconds": 60, "counterSpecifiers": [ "\VmInsights\DetailedMetrics" ], "name": "VMInsightsPerfCounters" }

Then we write another DCR to ingest Perf Data for VM's into the Log Analytics workspace which captures below :

"dataSources": { "performanceCounters": [ { "streams": [ "Microsoft-Perf" ], "samplingFrequencyInSeconds": 60, "counterSpecifiers": [ "\LogicalDisk(*)\Avg. Disk sec/Transfer" ], "name": "perfCounterDataSource60" },

Isn't this duplication of data for the same VM in that environment ? My understanding is both VMInsights Metrics and Perf have some common counters such as processor, memory and logical disks ?

Answer 1

Hi Gayatri Jagadev Ray

Welcome to the Microsoft Q&A Platform! Thank you for asking your question here.

Yes, you are correct in observing that there can be an overlap between the data ingested by VMInsights Metrics and traditional performance counters when configured in Data Collection Rules (DCRs). Both VMInsights and standard performance counters can collect similar metrics related to processor, memory, and logical disks.

VMInsights offers an overview of your virtual machine's performance and overall health, highlighting dependencies and resource usage. It combines and processes data, often including calculated metrics that provide a broader understanding of performance trends.

Performance counters focus on more granular data, offering specific metrics like CPU, memory, and disk usage in detail.

These counters are useful for in-depth monitoring and troubleshooting over time.

When both VMInsights and traditional performance counters are configured on the same virtual machines, there's potential for data duplication. This overlap can result in:

Increased data ingestion costs, as you're collecting similar data from both sources.

Redundant metrics stored in your Log Analytics workspace, which may lead to confusion when analyzing the data.

Data overload, making it more difficult to pinpoint the insights you're looking for.

Use VMInsights for higher-level monitoring and remove or reduce the collection of similar performance counter data.

Alternatively, focus on specific performance counters where fine-grained analysis is needed and use VMInsights for broader insights.

By minimizing the duplication of data, you can optimize data ingestion costs. Removing redundant data collection will lead to both cost savings and more focused insights

reference
https://learn.microsoft.com/en-us/azure/azure-monitor/vm/vminsights-performance
https://learn.microsoft.com/en-us/azure/azure-monitor/vm/monitor-virtual-machine
https://learn.microsoft.com/en-us/azure/azure-monitor/vm/vminsights-overview?tabs=powershell

If you have any further queries, do let us know.

---

If the answer is helpful, please and "Upvote it".