A company plans to deploy a database server cluster in AWS. The database cluster requires near-real-time replication between nodes. The built-in technology requires
Minimum network latency between nodes
Maximum network throughput
Which solutions meets the performance requirements for the database cluster?
A. Launch EC2 instances that have enhanced networking enabled using ENA. Launch all instances in a spread placement group.
B. Launch network-optimized EC2 instances that have jumbo frames enabled. Launch the instances in same AZ.
C. Launch EC2 instances that have enhanced networking enabled using the ENA. Launch all instances in a cluster placement group and enable jumbo frame support.
D. Launch EC2 instances that have multiple network interfaces. Launch all the instances in a cluster placement group and enable jumbo frame support.
Let me try to answer this:
A. It says "spread placement group". Minimum network latency won’t be achieved using this placement group config.
B. Launching in same AZ is correct to minimize latency however having correct placement group would boost latency improvement.
C. This is correct. ENA can help to improve bandwidth and I/O performance while Jumbo frame can improve throughput. Cluster placement group would help in minimizing latency.
D. Multiple network interfaces won’t help in improving I/O performance.