A Cloud Dataflow pipeline will receive data from 50,000 installations. To allow Dataflow to scale compute as needed, which pipeline configuration should you change?

To allow Cloud Dataflow to scale compute as needed for 50,000 installations, you should change the maximum number of workers. This parameter directly controls the upper limit of instances Dataflow can provision to process your data, ensuring it can handle increased loads by adding more parallel processing power. The zone specifies the geographic region for your pipeline and doesn't directly impact scaling capabilities. The number of workers is often an initial setting, but the maximum number dictates how much it can scale up. The disk size per worker affects storage capacity and I/O performance for individual workers, not the overall compute scaling of the pipeline.