Examkin

Question

Difficulty: MediumOptimizing Network Performance and Latency

A smart grid IoT SaaS provider collects real-time power grid telemetry from millions of sensors. The ingestion tier runs on Amazon EC2 instances in an Ingestion VPC, and the analytics processing tier runs in a separate Processing VPC within the same AWS Region. Currently, the VPCs are connected via an AWS Transit Gateway. During peak demand, the processing tier experiences latency spikes and packet drops when retrieving large payload batches of 90009000 bytes from the ingestion tier. Additionally, the processing instances resolve the ingestion tier's service endpoint using public DNS, which routes the traffic through NAT Gateways. The provider requires a solution that minimizes latency, maximizes throughput, and eliminates unnecessary data transfer costs. Which two actions should the Solutions Architect take to meet these requirements?

  1. Establish a VPC Peering connection between the Ingestion VPC and the Processing VPC, update the route tables to route traffic over the peering connection, and configure the EC2 instances in both VPCs to use a maximum transmission unit (MTU) of 90019001 bytes.Answer
  2. Associate the Route 53 Private Hosted Zone that contains the private DNS records for the ingestion service with the Processing VPC, allowing the processing instances to resolve the service directly to its private IP addresses.Answer
  3. C
    Modify the Transit Gateway route tables to route the inter-VPC traffic through a Direct Connect Gateway connected to an on-premises virtual router to manage transitive routing paths at high speeds.
  4. D
    Request AWS Support to pre-warm the Application Load Balancers in the Ingestion VPC to handle the peak traffic volume, and configure the processing instances to access the ingestion service using the load balancer's public DNS name.
  5. E
    Deploy redundant NAT Gateways across multiple Availability Zones in the Processing VPC, and route all inter-VPC traffic destined for the Ingestion VPC through these NAT Gateways to leverage their 100 Gbps100\text{ Gbps} bandwidth capacity.

Answer

Establishing a VPC Peering connection with an MTU of 90019001 bytes and associating the Route 53 Private Hosted Zone with the Processing VPC.
Establishing a VPC Peering connection within the same region allows network packets to use Jumbo Frames with an MTU of 90019001 bytes, preventing the fragmentation of 90009000-byte payloads that occurs over Transit Gateway attachments (which are capped at 85008500 bytes). Additionally, associating the Route 53 Private Hosted Zone with the Processing VPC ensures that DNS queries for the ingestion service resolve to private IP addresses rather than public IPs, eliminating the need to route traffic through NAT Gateways, which reduces both latency and data processing charges.

Step-by-Step Solution

1
Analyze the MTU constraints of the current network transit path.
AWS Transit Gateway VPC attachments support a maximum MTU of 85008500 bytes. Payloads of 90009000 bytes will be fragmented or dropped, leading to latency and packet loss.
Establishing a direct VPC Peering connection allows for an MTU of 90019001 bytes (Jumbo Frames) within the same region, preventing packet fragmentation.
2
Examine the DNS resolution and path for inter-VPC traffic.
Processing instances are currently resolving the ingestion endpoint using public DNS, which routes traffic via NAT Gateways, increasing latency and data processing costs.
By associating the Route 53 Private Hosted Zone with the Processing VPC, the processing instances resolve the ingestion service directly to its private IPs, keeping traffic within the private network.

Key Concept

Inter-VPC performance optimization using VPC Peering for Jumbo Frames (MTU 90019001) and private DNS resolution with Route 53 Private Hosted Zones.
Estimated Time:2m 0s
Rate this question

Topics

Optimizing Network Performance and Latency