A company recently migrated a data warehouse to AWS. The company has an AWS Direct Connect connection to AWS. Company users query the data warehouse by using a visualization tool. The average size…

Question

Options

• AOn-premises tool via internet: Incurs high costs due to large data transfers over the internet.
• BAWS Region tool via internet: Does not utilize Direct Connect, leading to potential latency and
• COn-premises tool via Direct Connect: Adds latency for querying and visualization.
• DAWS Region tool via Direct Connect: Reduces latency and leverages Direct Connect for

• A
  4% (2)
• B
  13% (6)
• C
  9% (4)
• D
  73% (33)

Explanation

Option D is correct because placing the visualization tool in the AWS Region means only the 500 KB rendered visualization needs to travel back to the company - not the raw 50 MB query result. This 100x reduction in transferred data is the primary cost optimization, and routing that small payload over Direct Connect (rather than the internet) keeps egress costs lower still.

Why the distractors fail:

• A (on-premises via internet): Worst of both worlds - 50 MB crosses the internet per query, incurring maximum data transfer costs.
• B (AWS Region via internet): Correctly co-locates the tool with the data, but internet egress is more expensive than Direct Connect for the return trip.
• C (on-premises via Direct Connect): Direct Connect is cheaper than the internet, but you're still moving 50 MB per query instead of 500 KB, so the volume problem remains.

Memory tip: Think "process where the data lives." When raw data is large but the output is small (50 MB → 500 KB here), always move the compute/visualization to the data, not the data to the compute. Then ask which network path is cheapest for the small result that does need to travel - Direct Connect beats public internet for predictable, high-volume transfers.

No community discussion yet for this question.