Power BI embedded feature roundup—new features now available
Дата обновления: 24 мая 2018 г.
The following new features are now available in Power BI Embedded:
Custom report Tooltips and Q&A Explorer
· Q&A Explorer—The natural-language querying feature is now available for your users who consume and interact with reports. In Power BI Embedded, you can also embed Q&A in your application separately from reports or dashboards.
New Azure Resource Metrics
In a recent blog post we announced the integration of Azure Monitor Metrics to be used on Power BI Embedded resources. This month, we're releasing three new metrics that, in addition to existing metrics, will give a good view of the status and performance of your resource:
· Memory—Shows the usage of memory in your resource. The units are in gigabytes (GB) so you can track exactly how much of your RAM is being used every minute.
· Memory thrashing—Shows the percentage of memory being thrashed from the RAM of your resource, relative to the total size of RAM in the resource. Note that this metric is relevant only for Import mode datasets, since they're hosting data in-memory. Data sets using DirectQuery or Live connection to Analysis Service are not monitored by this metric.
· QPU High Utilization—Tracks the load on the query processing units on your resource. Every minute, the metric checks if your resource’s QPU usage has exceeded 80 percent utilization of the resource so you can check for a potential case of performance degradation.
Setting up Alerts on your Azure resource
Power BI Embedded in Azure, which allows you to use monitoring metrics to track your resource performance and load, has been integrated with Azure Alerts. Alerts offers a method of monitoring in Azure that allows you to configure conditions over data and to be notified when the conditions match the latest monitoring data. When you have an active Power BI Embedded resource, you can use the Metrics that monitors your resource and define rules over those metrics.
Read more at our Power BI developer blog.