Examkin

Soru

Zorluk: ZorHigh-Performing and Elastic Compute Solutions

A financial services firm is deploying a containerized risk-modeling application on Amazon ECS using the Amazon EC2 launch type. The ECS tasks run on the EC2 instances using the host network mode. The compute processes within the containers use the Message Passing Interface (MPI) to perform tightly-coupled parallel calculations across multiple container instances, which requires sub-millisecond node-to-node network latency. Client requests are routed to the application container on port 8443 via a Network Load Balancer (NLB). A separate health-reporting daemon runs on port 8088 on each EC2 instance. Currently, the inter-node MPI communication is slow, resulting in calculation timeouts, and the NLB is marking all target EC2 instances as unhealthy. Which combination of actions should a solutions architect take to resolve these issues?

  1. A
    Launch the EC2 container instances in a spread placement group. In the NLB target group settings, configure the health check port to explicitly query port 8088.
  2. B
    Launch the EC2 container instances in a cluster placement group. Leave the NLB target group health check port set to the default traffic-port.
  3. Launch the EC2 container instances in a cluster placement group. In the NLB target group settings, configure the health check port to explicitly query port 8088.Cevap
  4. D
    Launch the EC2 container instances in a partition placement group. Leave the NLB target group health check port set to the default traffic-port.

Cevap

Launch the EC2 container instances in a cluster placement group. In the NLB target group settings, configure the health check port to explicitly query port 8088.
The correct option is to launch the EC2 container instances in a cluster placement group and configure the health check port to explicitly query port 8088. A cluster placement group packs instances close together within a single Availability Zone, which enables workloads to achieve the low-latency network performance required for Message Passing Interface (MPI) communication. Additionally, since the application receives traffic on port 8443 but the health-reporting daemon runs on port 8088, the target group's health check port must be explicitly configured to port 8088 instead of the default traffic-port.

Adım Adım Çözüm

1
Analyze the inter-node network requirements for the tightly-coupled MPI calculations.
Identify that MPI requires low-latency, high-throughput node-to-node communication.
Choosing the correct placement group is essential to minimize network latency between EC2 container instances.
2
Select the appropriate EC2 placement group type.
Select a cluster placement group to group the instances close together within a single Availability Zone.
A cluster placement group enables low-latency network performance, which is required for tightly-coupled applications like MPI.
3
Analyze the Network Load Balancer target group health check configuration.
Identify that the application traffic is on port 8443, but the health status daemon is on port 8088.
By default, target groups use the traffic-port (8443) for health checks, which leads to failures since the health status service is listening on port 8088.
4
Reconfigure the health check port on the target group.
Explicitly define port 8088 as the health check port.
This allows the NLB to query the correct health daemon, successfully marking healthy instances as healthy.

Anahtar Kavram

Selecting cluster placement groups for low-latency node communication, and configuring custom health check ports in load balancer target groups.
Tahmini Süre:2m 0s
Bu soruyu puanla

Konular

High-Performing and Elastic Compute Solutions

Bu konudan daha fazla soru