使用 Perspective + Spark 进行流式横截面数据可视化.pdf

1、2024 Databricks Inc.All rights reservedStreamingStreamingCrossCross-sectionalsectionalVisualizationVisualizationTim Bess/Tim PaineTim Bess/Tim Paine1withwith2024 Databricks Inc.All rights reserved Background and Motivation Perspective What it is,and what it does Connecting Spark Streaming/Perspectiv

2、e Introducing Prospective2Talk OverviewTalk Overview2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved3BackgroundBackgroundandandMotivationMotivation2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved4Background Background EcosystemEcosystemWeb

3、frameworks and data visualization tools abound!Web frameworks and data visualization tools abound!2024 Databricks Inc.All rights reserved5Despite the abundance.Despite the abundance.somethings are still.somethings are still.clunkyclunky2024 Databricks Inc.All rights reserved2024 Databricks Inc.All r

4、ights reservedMany tables require paginationPagination not always convenient or suitableGrids are often insufficient on their ownSorting/Filtering/PivotingOther chart typesStatic data sources/Streaming data sources/bothMany clients snapshot+update?VirtualizationVirtualization“Dumb grids”“Dumb grids”

5、StreamingStreaming6Three Major Pain pointsThree Major Pain points62024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved7Pain Point 1:Pain Point 1:VirtualizationVirtualization2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved8Pain Point 1:Virtualiz

6、ationPain Point 1:Virtualization Virtual tables allow for“infinite scroll”,lazy loading Absence of Virtualization leads to Pagination Pagination is often ill-suited to the task,can be annoying Most tools and frameworks do notdo not support virtualization!2024 Databricks Inc.All rights reserved2024 D