How to Connect On-Premises to a Transit Gateway with Single-Digit Latency?

SAA-C03 By islever Editorial Last updated

Short answer

BD Establish an AWS Direct Connect connection with a transit VIF to a Direct Connect gateway, then associate that gateway with the Transit Gateway.

Question

A company needs to design a hybrid network architecture. The company's workloads are currently stored in the AWS Cloud and in on-premises data centers. The workloads require single-digit latencies to communicate. The company uses an AWS Transit Gateway transit gateway to connect multiple VPCs. Which combination of steps will meet these requirements MOST cost-effectively? (Choose two.)
  1. Establish an AWS Site-to-Site VPN connection to each VPC.
  2. Associate an AWS Direct Connect gateway with the transit gateway that is attached to the VPCs. Correct Answer
  3. Establish an AWS Site-to-Site VPN connection to an AWS Direct Connect gateway.
  4. Establish an AWS Direct Connect connection. Create a transit virtual interface (VIF) to a Direct Connect gateway. Correct Answer
  5. Associate AWS Site-to-Site VPN connections with the transit gateway that is attached to the VPCs.

Key Takeaways

The test is whether you map a latency requirement to dedicated private connectivity and know the exact Transit Gateway building blocks, while the trap is choosing VPN because the question says "MOST cost-effectively."

Hybrid workloads that need single-digit latency between on-premises data centers and multiple AWS VPCs must use AWS Direct Connect, not internet-based VPN. The answer (B, D) pairs a transit virtual interface (VIF) to a Direct Connect gateway with associating that gateway to the existing Transit Gateway.

Common mistake: Choosing the VPN options (A or E) to minimize cost, or mixing a VPN with a Direct Connect gateway (C); VPN rides the public internet and cannot deliver consistent single-digit millisecond latency to on-premises workloads.

Expert Analysis

Why the Answer Is Correct

Single-digit latency between on-premises data centers and AWS workloads can only be engineered predictably over a dedicated private circuit, which means AWS Direct Connect — so option D, establishing a Direct Connect connection and creating a transit virtual interface (VIF) attached to a Direct Connect gateway, is mandatory. Option B completes the topology: the Direct Connect gateway must be associated with the Transit Gateway that already attaches the company's VPCs, so one dedicated connection reaches every VPC through the existing hub. Together, B and D give private, high-bandwidth, low-latency transit without building a Direct Connect connection or VIF per VPC. Because the Transit Gateway is reused rather than replaced, the design is also the cheapest way to satisfy the latency requirement.

Why the Other Options Are Wrong

Option A builds a Site-to-Site VPN into each VPC, which multiplies connections and still runs over the public internet with variable latency and encryption overhead — it fails the single-digit latency requirement and does not scale with the Transit Gateway. Option E attaches VPN connections to the Transit Gateway, which is architecturally tidy and cheap but again internet-dependent, so latency is neither low nor predictable. Option C chains a Site-to-Site VPN to a Direct Connect gateway, a topology that does not exist as a supported association for Transit Gateway routing and would not deliver the dedicated-path latency the workloads need. None of these options provide the dedicated private network path that Direct Connect does.

Community Comment Notes

The community consensus is firmly BD, with muhammadahmer36 stating simply "BD is correct" and MSAM16 agreeing. jingen11 initially picked BE and objected that "D is not cheap", which captures the real trap in this question: Direct Connect is the pricier option in isolation, but it is the only one that meets the single-digit latency requirement, so cost must be optimized within that constraint. LeonSauveterre's study memo reinforces the distinction — VPN is encrypted traffic over the public internet with higher latency, while Direct Connect is dedicated private connectivity for large workloads.

Official Reference

Exam Strategy

When a question states an absolute performance floor such as single-digit latency, treat it as a hard constraint and eliminate every internet-based option (VPN) before weighing cost. Then confirm the topology detail the exam wants — transit VIF plus Direct Connect gateway associated to the Transit Gateway — so one connection serves all attached VPCs.

Frequently Asked Questions

Why isn't a Site-to-Site VPN attached to the Transit Gateway cheaper and sufficient here?

VPN traffic crosses the public internet, so latency is variable and cannot be guaranteed at single-digit milliseconds. It is cheaper, but it fails the stated latency requirement.

Why does the Direct Connect gateway need a transit VIF instead of a private VIF?

A transit VIF is what allows the Direct Connect gateway to be associated with a Transit Gateway, giving one dedicated connection reach to every attached VPC.

More SAA-C03 FAQ →

Community Votes

BD
100%

100% of anonymous learners picked answer BD. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 3 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 3

  1. LeonSauveterre 👍 1
    A little memo for me. Hope it'll be helpful for you as well. 1. VPN Connection: For secure, encrypted traffic between on-premises and AWS over the public internet. ==> Relies on public internet (higher latency) 2. Direct Connect: For dedicated private network connectivity between on-premises and AWS (like data centers with large workloads). ==> Higher cost, longer setup time. 3. PrivateLink: For secure private connectivity to AWS services or third-party services (like Salesforce) within AWS. ==> Simple setup but limited to services configured with PrivateLink (not general connectivity). 4. VPC Peering: For direct connectivity between two AWS VPCs (like multi-account or region scenarios). ==> Does not support transitive connections; not suitable for external services like Salesforce.
  2. jingen11 👍 1
    Question asked about cheapest. D is not cheap
  3. MSAM16 👍 1
    BD is correct

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Submitted for moderation before publishing. Keep it helpful and respectful.

Practice All SAA-C03 Questions

Access 120 questions with complete answers and detailed explanations.

View Full SAA-C03 Practice Test →

← Back to SAA-C03 Study Guide