Sự kiện
Giải vô địch thế giới Power BI DataViz
16 giờ 14 thg 2 - 16 giờ 31 thg 3
Với 4 cơ hội tham gia, bạn có thể giành được gói hội nghị và lọt vào Chung kết TRỰC TIẾP ở Las Vegas
Tìm hiểu thêmTrình duyệt này không còn được hỗ trợ nữa.
Hãy nâng cấp lên Microsoft Edge để tận dụng các tính năng mới nhất, bản cập nhật bảo mật và hỗ trợ kỹ thuật.
APPLIES TO:
Power BI service for business users
Power BI service for designers & developers
Power BI Desktop
Requires Pro or Premium license
Visualization rendering in Power BI must be quick and accurate, which requires configuration of underlying algorithms for each visual type. Visuals in Power BI must be flexible enough to handle different sizes of semantic models. Some semantic models have only a handful of data points, while other semantic models have petabytes of data points. This article explains the strategies used by Power BI to render visualizations.
Each visual employs one or more data reduction strategies to handle the potentially large volumes of data being analyzed. Even a simple table employs a strategy to avoid loading the entire semantic model to the client. The reduction strategy varies by visual type. Each visual selects from the supported data reduction strategies as part of generating the data request sent to the server.
Each visual controls the parameters on those strategies to influence the overall amount of data.
For each strategy, there are defaults based on the shape and type of visualized data. To provide the right user experience, you can override defaults in the Power BI Formatting pane.
Certain models can provide statistics about the number of values for certain columns. When such information is present, we use that information to provide better balancing across multiple hierarchies if a visual doesn't explicitly override the count of values for a strategy.
For more information, see What's new in SQL Server Analysis Services.
In addition to the previously mentioned strategies, visuals with two hierarchies of grouping columns (axis and legend, or category and series) use one other strategy called dynamic limits. Dynamic limits are designed to better balance data points.
Dynamic limits provide a better selection of points for sparse data than static limits. For example, you can configure a visual to select 100 categories and 10 series with a total of 1,000 points. But the actual data has 50 categories and 20 series. At query runtime, dynamic limits select all 20 series to fill up the 1,000 points requested.
This also means that dynamic limits are responsive if only one of the grouping columns (just category, or just series) is defined. In this case, the number of items possible to return can be up to the limit of maximum points the visual specifies it can handle.
Dynamic limits are automatically applied when the server is capable:
Find details for each visual type in the following sections.
For more information about area chart visuals, see How the new line sampling algorithm works.
A combo chart uses the same strategies as a column chart. Notice that the line in the combo chart doesn't use the high-density algorithm that the line chart uses.
Power BI visuals can get up to 30,000 data points, but it's up to the visual authors to indicate which strategies to use. The default limit is 1,000, but the visual creator can change that up to a maximum of 30,000.
The filled map can use statistics or dynamic limits. Power BI tries to use reduction in the following order: dynamic limits, statistics, and configuration.
Trend axis limits:
For more information about line chart visuals, see How the new line sampling algorithm works.
For more information, see High-density line sampling in Power BI.
Depending on the configuration, a map can have:
For more information, see High-density sampling in Power BI scatter charts.
PowerApps visuals can get up to 30,000 data points, but it's up to the visual authors to indicate which strategies to use. The default limit is 1,000, but the visual creator can change that up to a maximum of 30,000.
No reduction strategy.
For more information, see High-density sampling in Power BI scatter charts.
R & Python visuals are limited to 150,000 rows. If you select more than 150,000 rows, the visual only uses the top 150,000 rows.
The shape map can use statistics or dynamic limits.
Sự kiện
Giải vô địch thế giới Power BI DataViz
16 giờ 14 thg 2 - 16 giờ 31 thg 3
Với 4 cơ hội tham gia, bạn có thể giành được gói hội nghị và lọt vào Chung kết TRỰC TIẾP ở Las Vegas
Tìm hiểu thêmĐào tạo
Lộ trình học tập
Use advance techniques in canvas apps to perform custom updates and optimization - Training
Use advance techniques in canvas apps to perform custom updates and optimization
Chứng chỉ
Được Microsoft Chứng nhận: Liên kết Phân tích Dữ liệu Power BI - Certifications
Thể hiện các phương pháp và thực tiễn tốt nhất phù hợp với yêu cầu kỹ thuật và kinh doanh để lập mô hình, trực quan hóa và phân tích dữ liệu với Microsoft Power BI.
Tài liệu
Change how visuals interact in a report - Power BI
Learn how to set Visual interactions in a Microsoft Power BI service report and a Power BI Desktop report.
Create small multiples in Power BI - Power BI
Small multiples, or trellising, split a visual into multiple versions of itself, presented side by side, with its data partitioned across these versions by a chosen dimension.
Troubleshoot visualizations in Power BI - Power BI
Troubleshoot visualizations in Power BI, including common problems and workarounds to overcome known obstacles or potential problems.