Which of the following steps should the engineer take next to solve the latency issue?

A cloud engineer has provisioned a VM for a high-frequency trading application. After the VM is put into production, users report high latency in trades.

The engineer checks the last six hours of VM metrics and sees the following:

• CPU utilization is between 30% to 60%.

• Networkln is between 50Kbps and 70Kbps.

• NetworkOut is between 3.000Kpbs and 5.000Kbps.

• DiskReadOps is at 30.

• DiskWriteOps is at 70

• Memory utilization is between 50% and 70%.

Which of the following steps should the engineer take next to solve the latency issue?
A . Move to a network-optimized instance type as the network throughput is not enough.
B . Modify the disk IOPS to a higher value as the disk IO is being bottlenecked at 100 IOPS.
C . Increase the memory of the instance as the high-frequency trading application requires more RAM.
D . Increase the instance size to allocate more vCPUs as the CPU utilization is very high.

View Answer

Answer: A

Explanation:

Since the NetworkOut is significantly higher than NetworkIn and considering the nature of a high-frequency trading application, the issue most likely lies with network throughput. Moving to a network-optimized instance type would provide higher network bandwidth, which can reduce latency in trades.

Reference: This solution is derived from the Management and Technical Operations domain of the CompTIA Cloud+ objectives, focusing on performance optimization for cloud services.