Power BI Model.MaxParallelismPerQuery - query parallelization DQ

With the new Model, Power BI already performs many DAX queries in parallel. You can use the MaxParallelismPerQuery setting to raise query parallelization in Power BI even further in order to fully utilize the processing capabilities of your cloud-scale data warehouse and to speed up your DirectQuery queries as much as feasible.

During the preview phase—the default value for the Model.MaxParallelismPerQuery parameter is 1, which effectively deactivates SE query parallelization for a single Dax query.  If you want to enable SE query parallelization for individual DAX and MDX queries, refer to the following code snippet to configure the Model.MaxParallelismPerQuery parameter. This parameter can only be configured through TOM and TMSL. 

using TOM = Microsoft.AnalysisServices.Tabular;
partial class Program
{
    static void Main(string[] args)
    {
        using (var server = new TOM.Server())
        {
            string workspaceUrl = "<URL to workspace>";
            string databaseName = "<database name>";
            // Connect to the dataset.
            server.Connect(workspaceUrl);
            TOM.Database database = server.Databases.FindByName(databaseName);
            if (database == null)
            {
                throw new ApplicationException("Database cannot be found!");
            }
            if (database.CompatibilityLevel < 1569)
            {
                database.CompatibilityLevel = 1569;
                database.Update();
            }
            TOM.Model model = database.Model;
            // The max number of parallel SE query for a single Dax query.
            model.MaxParallelismPerQuery = 10;
            model.SaveChanges();
        }
        Console.WriteLine("\nDone! Press any key to exit.");
        Console.Read();
    }
}

reference: https://powerbi.microsoft.com/en-us/blog/query-parallelization-helps-to-boost-power-bi-dataset-performance-in-directquery-mode/