Optimized modes for tile visualization on lower zoom level

On lower zoom levels, to reduce the payload, tiled data can be optimized in size. This will result in better performace for rendering clients. This affects the amount of features, their attributes and coordinates.

All described parameters and values are related to tile requests.

mode=viz

The strategies to reduce datasize in "viz" mode are as follows.

  • reduce payload of features
    default, every feature will contain an empty feature.properties value. (e.g. properties:{} )
    To get specific properties attributes, they must be explicit specified with tile parameter "selection=". (e.g. "http://....?mode=viz&selection=p.firstAttrib,p.secondAttrib,...p.lastAttrib ) For parameter "selection=" the special values "*", for all attributes, and "none" can be used.

  • reduce coordinates of geometry
    depending on requested tile level the coordinates of the geometries (except for point objects) are reduced, to at most one coordinate per pixel (512 pixelsize per tile)

  • reduce amount of features per tile
    default, a sampling threshold of at max 30k objects for the requested tiles is used
    The threshold can be controled with the additional parameter vizSampling (s.b.)

vizSampling={value}

In order to reduce the amount of data returned, only an evenly distributed sample of data can be requested. Parameter vizSampling sets the wanted threshold for the requested tiles. E.g. setting "vizSampling=med" when requesting a specific tile, will calculate the sampling ratio such that, the "heaviest" tile in an observed region of the dataset will contain at most 30k objects.

value threshold
low 70k
med 30k (default)
high 15k
off turn off vizSampling

Request

Try in Swagger

GET /spaces/{spaceId}/tile/{type}/{tileId}?mode=viz&vizSampling=med&selection=*

results matching ""

    No results matching ""