Examkin

Question

Difficulty: Very hardOptimizing Network Performance and Latency

A financial services provider hosts a latency-critical trading platform in the us-east-1 Region. The architecture consists of a high-throughput compute cluster in VPC-A that queries a key-value database cluster in VPC-B. Both VPCs are connected via an AWS Transit Gateway. The database cluster utilizes a Route 53 Private Hosted Zone (PHZ) for internal endpoint resolution, which is currently associated only with VPC-B. Compute instances in VPC-A must also communicate with the internet for market data feeds, routing all outbound traffic through a single NAT Gateway located in VPC-A's public subnet.

The platform is experiencing performance bottlenecks. Inter-VPC database traffic is capped at 5 Gbps per flow, and DNS resolution failures occur frequently during market open. Furthermore, outbound internet connectivity is vulnerable to Availability Zone outages.

Which two changes should a solutions architect implement to optimize network throughput, reduce latency, and improve availability? (Select TWO.)

  1. Establish a VPC Peering connection directly between VPC-A and VPC-B, configure the VPC route tables to route database traffic over the peering connection, and ensure MTU is configured for Jumbo Frames (9001 bytes) on both sides.Answer
  2. Associate the Route 53 Private Hosted Zone with VPC-A to resolve database endpoints locally, and deploy a NAT Gateway in each Availability Zone where the compute instances are located.Answer
  3. C
    Establish an AWS Direct Connect connection with a Direct Connect Gateway, and configure transitive routing between VPC-A and VPC-B through the Direct Connect Gateway to bypass Transit Gateway latency.
  4. D
    Request an AWS Support ticket to pre-warm the existing NAT Gateway to increase the packet processing limits and eliminate TCP flow bottlenecks during peak traffic.
  5. E
    Associate the Route 53 Private Hosted Zone with the Transit Gateway subnet to allow transitive DNS resolution, and route all egress traffic through the Transit Gateway to a centralized internet gateway in VPC-B.

Answer

Establish a VPC Peering connection directly between VPC-A and VPC-B, configure the VPC route tables to route database traffic over the peering connection, and ensure MTU is configured for Jumbo Frames (9001 bytes) on both sides; and associate the Route 53 Private Hosted Zone with VPC-A to resolve database endpoints locally, and deploy a NAT Gateway in each Availability Zone where the compute instances are located.
To optimize inter-VPC latency and throughput, establishing a VPC Peering connection directly between VPC-A and VPC-B bypasses the Transit Gateway's intermediate hop and allows the use of Jumbo Frames (MTU 9001) for high-performance network paths. To fix DNS resolution, the Route 53 Private Hosted Zone must be associated directly with VPC-A. High availability and localized routing for internet egress are achieved by deploying a NAT Gateway in each Availability Zone.

Step-by-Step Solution

1
Analyze the inter-VPC latency and throughput constraints.
Identified that AWS Transit Gateway introduces an extra processing hop compared to direct VPC Peering. Single-flow TCP bandwidth over Transit Gateway is capped at 10 Gbps, and default paths may not exploit Jumbo Frames.
VPC Peering provides direct, single-hop communication with support for Jumbo Frames (9001 MTU) within the same region, reducing latency and throughput constraints.
2
Address the DNS resolution failures for VPC-A.
Determined that the Route 53 Private Hosted Zone (PHZ) created in the database account must be explicitly associated with VPC-A.
By default, a PHZ is only queryable by resources in the associated VPCs. Cross-account or cross-VPC association is required for VPC-A to resolve endpoints hosted in the PHZ.
3
Resolve the egress network resiliency bottleneck.
Deploying a NAT Gateway in each Availability Zone where the compute resources are hosted, and updating the subnet routing tables.
A single NAT Gateway creates a single point of failure if its Availability Zone goes down. Utilizing an AZ-specific NAT Gateway guarantees localized fault tolerance and high availability.

Key Concept

Optimizing network performance involves bypassing unnecessary network hops using VPC Peering, enabling Jumbo Frames (MTU 9001) for internal high-throughput paths, and resolving DNS and egress paths natively and redundantly within each VPC and Availability Zone.
Rate this question

Topics

Optimizing Network Performance and Latency