The on-premises data gateway has settings that control resource usage on the machine where the gateway is installed. By default, gateway releases starting in June 2019 (3000.6.204) automatically scale these settings, using more or less resources depending on CPU capacity:
Setting
Description
MashupDefaultPoolContainerMaxCount
Maximum container count for Power BI refresh, Azure Analysis Services, and others.
MashupDefaultPoolContainerMaxWorkingSetInMB
Maximum working set size for Power BI refresh, Azure Analysis Services, and others.
MashupDQPoolContainerMaxCount
Maximum container count for Power BI Direct Query.
MashupTestConnectionPoolContainerMaxInstanceCount
Maximum container count for test connections.
MashupAzureConnectorsCachingPoolContainerMaxCount
Maximum container count for LogicApps, Power Apps, and Power Automate.
PowerQueryOnlineCachingPoolContainerMaxCount
Maximum container count for Power Query Online.
Use the MashupDefaultPoolContainerMaxWorkingSetInMB setting to change the default pool. Changing the memory set is only possible for the default pool. For the other pools, it isn't possible due to performance and stability reasons. The MashupDQPoolContainerMaxWorkingSetInMB settings can't be changed in the config.
Most queries use mashup containers to execute. So the number of mashup containers determines the number of queries that can be executed in parallel. A working set defines the memory allocated to each container. These settings are available in \Program Files\On-premises data gateway\Microsoft.PowerBI.DataMovement.Pipeline.GatewayCore.dll.config.
If you've changed any of these settings manually, for these settings to take effect you must also disable automatic scaling by setting MashupDisableContainerAutoConfig to True in \Program Files\On-premises data gateway\Microsoft.PowerBI.DataMovement.Pipeline.GatewayCore.dll.config. If MashupDisableContainerAutoConfig is set to False, automatic scaling is always enabled.
With Microsoft Power BI, you can use a single semantic model to build many reports. Reduce your administrative overhead even more with scheduled semantic model refreshes and resolving connectivity errors.
Discusses how to fix blocked outbound connections, how to configure certain ports for the on-premises data gateway to create an outbound connection to Azure Relay, how to force the gateway to communicate with Azure Relay by using HTTPS instead of direct TCP, and how to ensure your gateway machine is using TLS 1.2 to communicate with the Microsoft Power BI service.