Configure a map image layer

When you configure a map image layer, you can set many properties of its underlying map service, including drawing behavior, supported user operations, and the way server resources are allocated to the service. Configuration is divided into three areas: layers, parameters, and pooling. This workflow explains how to modify the default configuration of a map image layer. To begin the sharing process, set general properties, and share your layer, see Share a map image layer.


If you are configuring a hosted map image layer (a map image layer published from a web feature layer), see the configuration limitations.

Configure layers

You can enhance the functionality of your map image layer by adding layers and capabilities. You may also configure a map image layer when you configure a web map. The information in this topic applies in that situation also.

  1. Open the Share As Web Layer pane, if necessary.

    On the General tab, confirm that Map Image is selected.

  2. Click the Configuration tab. If necessary, click the Configure Layers tab Configure Layers below it.
  3. Under Layer(s), next to Map Image, click Configure Web Layer Properties Configure Web Layer Properties.

    Under Operations, the Map, Data, and Query operations are turned on by default. These operations are described in the following table:


    The Export Map, Generate Renderer, Generate KML, and Query Legends methods can be used. This operation cannot be turned off.


    The Find, Query, and Query Related Records methods can be used.


    The Identify method can be used.

  4. Optionally, uncheck the Data and Query boxes to turn off these operations.
  5. Optionally, under Dynamic Workspaces, uncheck the Allow per request modification of layer order and symbology box.

    Unchecking this box prevents users from changing the appearance or drawing behavior of the web layer. For example, users cannot change symbology, labeling, or the order of component layers.

  6. Under Layer Drawing, choose how the map image layer will be drawn:

    Dynamically from data

    Data is retrieved and drawn each time it is requested. This is the default option.

    Using tiles from a new cache

    Data is drawn from cached images generated at map scales you define.

    Using an existing cache

    Data is drawn from tiles previously generated on the server. To associate an existing cache with the map image layer, the name and server folder of your map image layer must match the name and server folder of the existing cache.

    For more information, see What is map caching? in the ArcGIS Enterprise help.


    The Using an existing cache option searches for a disassociated cache with the same name and server folder as your web layer. A cache is disassociated from a service if the service was deleted or if the cache was copied to the cache directory. If an eligible cache is found on the server, it is used to draw the map image layer. The map name, spatial reference, and extent must match the cache for the layer to draw correctly.

    If you use tiles from a new or existing cache, additional settings may be available in the pane. Continue with the following substeps:

    1. Click the Tiling Scheme drop-down arrow and choose a tiling scheme for your cache.

      The tiling scheme determines the scale levels, tile dimensions, and tile origin of the cache. These properties define tile boundaries and must be consistent for different web layers used in the same map. The available tiling schemes are described in the following table:

      ArcGIS Online/Bing Maps/Google Maps

      The web layer is projected to the WGS 1984 Web Mercator (Auxiliary Sphere) coordinate system. The default tile width and height is 256 by 256 pixels. Scale levels for caching range from 0 to 23.

      WGS 1984 Geographic Coordinate System, Version 2

      The web layer is projected to the WGS 1984 Geographic Coordinate System. The default tile width and height is 256 by 256 pixels. Scale levels for caching range from 0 to 22.

      Existing cached map/image service

      The tiling scheme of an existing hosted tile layer, or an existing ArcGIS Server map or image service is used. The tiling scheme .xml file of the layer or service is downloaded to C:\Users\<user profile>\My Documents\ArcGIS\TilingSchemes\Downloaded. ArcGIS Pro honors the parameters specified in the existing layer or service, with the exception of antialiasing. Antialiasing options set for the application are used.

      Tiling scheme file

      A tiling scheme that you create is used. You can create a tiling scheme .xml file with the Generate Map Server Cache Tiling Scheme geoprocessing tool. Additional tiling scheme files can be imported from C:\Program Files\ArcGIS\Pro\Resources\TilingSchemes. The tile image format defaults to PNG if the tile image format is not defined in the tiling scheme file. ArcGIS Pro honors the parameters specified in the tiling scheme file, with the exception of antialiasing. Antialiasing options set for the application are used.


      This is the default option unless the coordinate system of your map is WGS 1984 Web Mercator (Auxiliary Sphere) or WGS 1984. A tiling scheme is suggested based on the coordinate system of your map or scene. The levels of detail defined in this tiling scheme are valid for a global extent, so any web layers cached using this tiling scheme can be displayed together.

    2. Under Levels of Detail, use the sliders to set the map scales for which tiles are generated.

      A default scale range is provided. Depending on the purpose of your map and the estimated cache size, you may choose not to cache very large or small scales.

    3. For Image format, accept the default or choose a different setting. Use the table below for reference.

      If all sublayers in the web layer have vector data sources, the default setting is PNG. If one or more sublayers has a raster data source, the default setting is MIXED.



      Use PNG to select the correct format (PNG8, PNG24, or PNG32) automatically based on the levels of detail that are being cached.


      Use PNG8 for overlay layers that need a transparent background, such as roads and boundaries. PNG8 creates tiles of very small size on disk with no loss of information.

      Do not use PNG8 if your map contains more than 256 colors. Imagery, hillshades, gradient fills, transparency, and antialiasing may cause your map to use more than 256 colors. Symbols such as highway shields may have antialiasing around the edges that unexpectedly adds colors to your map.


      Use PNG24 for overlay layers, such as roads and boundaries, that have more than 256 colors. (If fewer than 256 colors are used, choose PNG8.)


      Alternatively use PNG32 for overlay layers, such as roads and boundaries, that have more than 256 colors. PNG32 is a good choice for overlay layers that have antialiasing on lines or text. PNG32 creates larger tiles on disk than PNG24.


      Use this format for basemap layers that have large color variation and do not need a transparent background. Raster imagery and very detailed vector basemaps tend to work well with JPEG.

      JPEG is a lossy image format. It attempts to selectively remove data without affecting the appearance of the image. This can cause very small tile sizes on disk, but if your map contains vector line work or labels, it may produce too much noise or blurry area around the lines. If this is the case, you can increase the compression value from the default of 75. A higher value, such as 90, may balance an acceptable quality of line work with the small tile size benefit of the JPEG.


      The MIXED format uses JPEG in the center of the web layer and PNG32 on the edges. Use MIXED to overlay raster layers on other layers.

      When the MIXED format is used, PNG32 tiles are created anywhere that transparency is detected (in other words, anywhere the map background is visible). The rest of the tiles are built using JPEG. This keeps the average file size down while providing a clean overlay of raster layers on top of other layers. If you do not use the MIXED format in this situation, you will see a nontransparent collar around the edge of your image where it overlaps other layers.

    4. For Compression, accept the default compression quality value or specify a different value.

      Compression is available only if the format is JPEG or MIXED. (PNG formats are not compressed.) You can use values between 0 and 100. Higher values result in a larger file sizes with higher quality imagery. Lower values result in smaller file sizes with a lower quality imagery. The default value is 75.

    5. Under Options, choose how to build the cache.

      The storage estimate under Estimate at the bottom of the pane may help you decide how to build the cache.

      Cache automatically on the server

      The cache is built automatically. This is the default setting and is generally best for small caches.

      Cache manually on the server

      You build the cache manually using the Manage Map Server Cache Tiles tool. This option is best for large caches. It is recommended that you build the cache in phases, first caching the largest map scales at areas of interest that are most likely to be requested.


      When you save a service definition, the cache must be built manually after the map image layer is published.

    6. Optionally, click the Cache directory drop-down arrow and change the server cache directory where the cache tiles will be stored.
    7. Optionally, check the Allow clients to export cache tiles box to allow users to download cache tiles for offline use.

      When this option is on, you can specify a value in the Limit export to box to limit the number of tiles that can be downloaded in a single request.

    8. Optionally, check the Create tiles on demand box.


      Not checked

      Tiles are built only through caching, either automatically or manually. This is the default.


      Tiles that are not cached are created on demand (that is, when a client application requests them). This occurs mainly in the following situations:

      • Tiles are cached automatically on the server, and you specified an area of interest for the cache extent. In this case, tiles are cached only for the area of interest. Tiles are created on demand for the rest of the map extent.
      • You cache tiles manually on the server. In this case, tiles are cached for the scales selected in the Manage Map Server Cache Tiles tool. Tiles are created on demand for scales that are not selected.

    9. Under Extent, choose an extent to cache.

      Use map extent

      Tiles are cached for the extent specified in the map properties. The default is the extent of all layers in the map. If you share selected map layers, tiles are cached only for the combined extent of the selected layers.

      Area of Interest

      Tiles are cached for the extent of a feature layer in the map. Use the drop-down list to choose a layer. Alternatively, click the Browse button Browse and browse to a feature class on disk.


      If you cache manually on the server, cache extent options are not available in the Share As Web Layer pane. They are available in the Manage Map Server Cache Tiles geoprocessing tool that you use to build the tiles.

    10. Optionally, under Estimate, click Calculate to estimate the cache size. On the Estimate Cache Size dialog box, click the drop-down arrow and choose Good or Best for the estimation quality. Click the Start estimating cache button Start estimating cache. The number of tiles and storage space is estimated for each cached level of detail and totaled. Click OK.

      The default static estimate is based on a general assumption about average tile size and does not consider the specific properties of your map. When you estimate the cache size, sample tiles of your map are built. This means that the tiling scheme, image format, and compression are taken into account (as is LERC compression for elevation layers). If an area of interest is specified as the caching extent, the area of interest is estimated.

      The best estimate creates more sample tiles and takes longer to calculate than the good estimate. To cancel the estimation process, click Stop estimating cache Stop estimating cache. If you adjust the tiling scheme after estimating the cache size—for example, if you change the levels of detail or the image format—the quality estimate on the sharing pane reverts to static. You need to estimate the cache size again.

  7. At the top of the pane, click the Back button Back to return to the list of layers and capabilities.
  8. Optionally, under Additional Layers, check one or more boxes to create layers in addition to the map image layer.
    • WMS
    • WFS
    • OGC Features

    A WFS layer is available only if Map Image was selected under Reference registered data on the General tab.

  9. Optionally, under Capabilities, check one or more boxes to turn on available capabilities.
    • WCS
    • KML
    • Linear Referencing
    • Network Analysis
    • Topographic Production
    • Validation
    • Version Management

    The list of capabilities depends on your ArcGIS Enterprise version, the types of data included in your map image layer, and whether you are sharing to a server that has server object extensions or server object interceptors.

    After you add a layer or enable a capability, you can configure its properties. See Additional layers and capabilities for information.

Configure parameters

Map image layers have a number of configurable parameters. Parameter settings apply to the map image layer and to any other layers and capabilities you add. Learn more about service parameters.

  1. On the Configuration tab, click the Configure Parameters tab Configure parameters.
  2. Under Properties, specify a value for Maximum records returned by server.

    This setting determines the maximum number of records returned by a user query. A large number may result in slower performance. The default value is 2,000.

  3. Expand the Advanced heading to set the following parameters:
    • Enter a value in the Max number of coded domains box to specify the maximum number of domain codes returned from fields, subtypes, layers, and tables. The default value is 25,000.
    • Specify values in the Max image height and Max image width boxes. These values represent the maximum height and width in pixels of images the map service can export. The default for both values is 4,096.
    • Enter a value in the Max sample size box to specify the maximum number of records that is sampled when a class breaks renderer is computed. The default value is 100,000.
    • Optionally, uncheck the Enable schema locking box to disable schema locking. By default, schema locking is enabled to prevent modifications to datasets that are used by reference in the map image layer.
    • Optionally, check the Disable identify relates box. If the box is not checked (the default), information from related tables is returned when a feature is identified. If the box is checked, information from related tables is not returned.
  4. Under Date Fields, choose a time zone from the Time zone menu, if necessary.

    Specifying the time zone allows date and time values to be converted correctly when users interact with the map image layer. If you choose a time zone other than UTC (coordinated universal time) and you want the time zone to account for daylight saving time, check the Adjust For Daylight Saving box.


    You only need to choose a time zone if your map contains layers with date fields. Choose the time zone in which the dates are stored. If you don't specify a time zone, the dates are assumed to be stored in UTC.

  5. Optionally, under Settings, uncheck the Ensure map is set to allow assignment of unique IDs check box.

    If the box is checked (the default), the map is analyzed to confirm that the Allow assignment of unique numeric IDs for sharing web layers option in Map Properties is enabled. If the box is unchecked, the analyzer is skipped. If the option in Map Properties is not enabled, unique IDs are automatically assigned to layers and tables when the web layer is published.


    Checking the check box does not automatically enable any option in Map Properties. Authoring the map with assigned layer and table IDs ensures they remain static if the contents or layer ordering in your map change. When overwriting the web layer, assigned IDs must match existing service sublayer IDs to maintain references to them in web maps.

    Learn more about assigning layer IDs

  6. If you are publishing data from a cloud data warehouse, additional settings may be available in the pane.

    Under Data Source, choose an option that determines how the service will access the data.

    1. Select Create a materialized view or access data directly if you want the service to access data stored in the data warehouse. When this is selected, two options may be available for each query layer from a cloud data warehouse:

      Materialized view

      Generates a materialized view, a database view containing precomputed query results for the service. This improves query performance because the service will access the materialized view rather than executing the SQL query every time you work with the query layer.

      This option is only available for query layers added from the New Query Layer dialog box. See Create a query layer for data in a cloud data warehouse for more information.

      Access data directly

      References data directly from the database. This is the only available option for the following:

      Depending on the size of the table, you may want to add the query layer again using the New Query Layer dialog box to leverage the materialized view option.

    2. Select Create a snapshot for query layers to generate a snapshot of the data when continuously updated data is not needed. The snapshot is stored locally with the relational data store on the hosting server, resulting in better rendering performance for the service. When an update is needed, the snapshot can be refreshed from the web layer's item page on the portal.

      The table lists each query layer from a data warehouse and whether it supports creating a snapshot. Snapshots are only supported for query layers added from the New Query Layer dialog box. They are not supported for the following:

      • Query layers for tables accessed by adding or dragging them onto the map from a data warehouse connection
      • Bin-enabled feature layers
      • Query layers with parameters

      Depending on the size of your data, you may want to add the query layer again using the New Query Layer dialog box to leverage the snapshot option.


    These options are available only if you are sharing query layers from cloud data warehouses to ArcGIS Enterprise 10.9.1 and later.

Configure pooling

Pooling settings help you manage the responsiveness of your map image layer (through its underlying map service) while conserving memory resources on server machines. To learn more about pooling, see Anticipate and accommodate users and Configure service instance settings in the ArcGIS Enterprise help. See also the online article Introducing shared instances in ArcGIS Server 10.7.

  1. On the Configuration tab, click the Configure Pooling tab Configure Pooling.
  2. Under Instance Type, click Dedicated instance or Shared instance.

    A dedicated instance allocates a server process to handle requests exclusively for the map service exposed by your map image layer. A shared instance allows one server process to handle requests for this map service as well as others. Generally, dedicated instances are better for frequently used services, and shared instances are better for infrequently used services.


    Shared instance pooling can only be set from ArcGIS Pro if your ArcGIS Enterprise version is 10.7.1 or later.

  3. If you choose the dedicated instance type, under Number of instances per machine, set values for the Minimum and Maximum number of instances.

    According to user demand, server processes are allocated up to the maximum number of instances and down to the minimum number on each federated server on which the service runs. If you set the minimum number to zero, processes are released if the map service receives no requests for a certain amount of time.


    This setting is not used if you choose the shared instance type.

Related topics