A company has an SLA of 15ms for storage latency. Given the above metrics, which of the following is the MOST appropriate change to the environment?

Adding computing and data nodes primarily scales capacity and IOPS for distributed storage, but does not directly guarantee a reduction in latency for individual I/O operations.

Adding more storage shelves increases capacity and potentially IOPS, but does not inherently reduce the latency of individual I/O requests.

Enabling compression reduces the amount of data transferred to and from physical storage, which directly decreases the time required for I/O operations and thus lowers storage latency. The processing overhead of compression is typically less than the latency reduction achieved by moving less data.

While deduplication reduces stored data, its processing overhead, especially during writes, can sometimes increase latency, making it less reliably a primary solution for latency reduction compared to compression.