Exam DP-900 topic 1 question 9 discussion

It's A.

The correct answer is "across compute nodes". Distributions map to Compute nodes for processing. As you pay for more compute resources, distributions are remapped to available Compute nodes. Reference: https://docs.microsoft.com/en-us/azure/synapse-analytics/sql-data-warehouse/massively-parallel-processing-mpp-architecture [see the architecture diagram]

a. distributes processing across compute nodes, Azure Synapse Analytics uses a Massively Parallel Processing (MPP) architecture, which distributes data and processing tasks across multiple compute nodes to enhance performance and scalability.

option A

MPP systems are designed to be distributed parallel processing solutions, meaning they are not only able to scale up by adding more compute resources but can also scale out by adding more nodes to the system.

https://docs.microsoft.com/en-us/shows/learn-azure/understand-massively-parallel-processing-concepts

Correct Answer: A

Correct Answer

Vote for A

It is A. https://docs.microsoft.com/en-us/learn/modules/explore-data-storage-processing-azure/2-describe-azure

The control node gets the process and distributes it across compute node. Therefore the answer is A.

It's A. Dedicated SQL pool (formerly SQL DW) uses a node-based architecture. Applications connect and issue T-SQL commands to a Control node. The Control node hosts the distributed query engine, which optimizes queries for parallel processing, and then passes operations to Compute nodes to do their work in parallel.

All of the source about MPP say that A is correct.

which one is the correct answer? A or C ?

I think "A" is a correct answer - not redirecting connections ("C") but distributing processing

Correct MPP across nodes