Speed up single-key demand forecasts with request response mode
Short answer
D Enable Use request response mode in the demand forecasting parameters so single-allocation-key forecasts are processed synchronously instead of through the slower file-based workflow.
Question
The process takes too long to complete.
You need to ensure better performance of the service.
What should you configure?
Key Takeaways
This tests whether you know that demand forecasting throughput for small, incremental requests is improved by the Azure Machine Learning request response mode, not by any storage account setting.
For MB-330 demand forecasting performance, the fix for generating a baseline forecast one allocation key at a time is enabling Use request response mode in the demand forecasting parameters, making option D correct.
Common mistake: Option C, a custom storage account, is the most plausible wrong answer because it suggests an Azure-side performance optimisation, but storage configuration only controls where files are stored, not how fast requests are processed.
Expert Analysis
The attractive wrong answer: option C
A custom storage account looks like the natural performance fix here. The demand forecasting service moves large volumes of data through Azure Blob storage — input files, intermediate results, and output forecasts — so it is tempting to reason that pointing the service at a dedicated or higher-tier storage account will speed things up. Option C also echoes a familiar Azure pattern where "custom" configuration signals an optimisation.
It fails on one specific point: storage account choice affects where files live, not how requests reach the Azure Machine Learning service. The question states the bottleneck is the per-allocation-key workflow — the service is being driven one key at a time through the slower file-based exchange, and no storage endpoint setting changes that round-trip. Options A (storage account name) and B (storage account key) are purely credentials/identifiers for the existing storage account and have no bearing on throughput at all.
The correct answer: option D
When forecasts are generated for a small set of items — one item or one item allocation key at a time — enabling the Use request response mode checkbox (Master planning > Setup > Demand forecasting > Demand forecasting parameters > Azure Machine Learning tab) lets the forecasting service handle requests synchronously instead of waiting on the file/storage-based workflow, which is exactly the performance lever the question asks for.
Exam takeaway
In MB-330, whenever a demand forecasting or master planning question ties slowness to small, incremental request volumes, look for the request/response toggle in the demand forecasting parameters rather than any Azure storage configuration value.
Official Reference
This link opens the official vendor documentation in a new tab.
Exam Strategy
Read for the workload shape: "one allocation key at a time" signals small, repeated requests rather than bulk data movement, which rules out every storage-related option immediately. Then map the symptom to its setup path — Master planning > Setup > Demand forecasting > Demand forecasting parameters > Azure Machine Learning — and pick the toggle that changes the request transport, not the data location. On MB-330, storage account name/key/custom options are usually distractors for throughput questions.
Frequently Asked Questions
Where do I enable request response mode for demand forecasting?
When does request response mode improve forecasting performance?
Does changing the storage account speed up demand forecasting?
Community Votes
100% of anonymous learners picked answer D. Votes are pick records left by other test-takers — they are not the verified answer.
Discussion 4 contributions
Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.
From public community threads 4
-
Vnes80 👍 3Correct https://learn.microsoft.com/en-us/dynamics365/supply-chain/master-planning/generate-statistical-baseline-forecast
-
Gemami 👍 1 Answered DCorrect answer: D. Request response mode Why: When demand forecasting generates a statistical baseline forecast one allocation key at a time, enabling/configuring request-response mode improves performance by allowing the forecasting service to process requests directly instead of relying on the slower file/storage-based workflow. Why not: - A. Storage account name — just identifies the Azure storage account. - B. Storage account key — authentication for storage, not a performance setting. - C. Custom storage account — changes where forecasting data is stored, not the processing mode used to improve this scenario.
-
globeearth 👍 1 Answered DIf the forecast isn't generated for a large set of items, but, for example, for one item or one item allocation key at a time, then in order to get better performance, you can select the Use request response mode checkbox on the Master planning - Setup - Demand forecasting - Demand forecasting parameters - Azure Machine Learning tab.
-
globeearth 👍 1sorry it is D
Reader comments & corrections
No reader corrections yet — spotted an error or have a note? Share it below.