The Multidimensional analysis view enables you to analyze web requests of your services with fine-tuned filtering, so you can focus your analysis on the dimensions that matter most. This view is easily configurable and serves as a convenient entry point for in-depth analysis of your services.
With multidimensional analysis, you can investigate custom views: Top web requests, Database statements, and Exception analysis.
To access the Multidimensional analysis page
- In the Dynatrace menu, go to Multidimensional analysis.
- Choose one of the analysis options that you want to investigate, or Create analysis view.
Multidimensional analysis uses trace and request data as its data source, which includes information on traces, code-level insights, and errors and on requests. For timeframes shorter than two hours, high-resolution data is used. For longer timeframes, sampling is applied: only a fraction of trace and request data is used.
- To view the information about the data used for analysis and suggestions on how to improve accuracy, hover over Refine.
- To include more data to the analysis, select Refine.
Unlike the Data explorer, multidimensional analysis uses trace and request data, not metric data, so values on multidimensional analysis charts might differ from values on custom charts.
The Configure view pane offers multiple filtering capabilities. The view updates automatically as you change the parameters.
The metric to be analyzed.
How the metric values are aggregated. Available aggregations depend on the selected metric.
How dimensions specified in Split by dimension are treated.
Split by dimension
A list of dimensions by which the requests are split.
You can specify several dimensions. Place your cursor in the input field to see the available options. The requests are split by dimension in the specified order.
Filter the requests to be included to the view. Place your cursor in the input field to see the available options.
Criteria of the same type are grouped by OR logic. Criteria of different types are grouped by AND logic.
The chart shows the top 15 dimensions (all other dimensions are aggregated into a single dimension) and the table beneath contains up to 85 more dimensions, bringing the total number of chartable dimensions to 100. The view instantly adapts to the changes you make in the Configure view pane.
The Actions column of the table contains the filter button , which filters the view to the specified dimension, and the menu button , which provides quick access to further analysis options and the Exception details view.
For timeframes containing data older than 10 days, you can turn on Show data retention to better understand which data is available for which period directly from the chart.
Once the view is configured, you can save it for quick access in the future. Just select Save view and provide a name.
Dynatrace provides several views out of the box:
Calculated service metric
Only new data is written to the calculated metric. Retrospective data is not added.
To create a calculated service metric from a multidimensional analysis view
In the Dynatrace menu, select Multidimensional analysis.
Select Create analysis view.
optional Select the management zone. The new metric will be restricted to data from this zone.
Configure the view. Find the description of available options in the Configure view section above.
In the View pane, select Create metric.
Specify the name for the metric. The name is added to the metric key automatically.
optional If needed, customize the metric key. Note: once a metric is created, you can't change its key.
optional If you want to fine-tune your metric, select Advanced options to configure additional parameters of the metric. For details, see Calculated metrics for services.
Select Create metric.