A customer has identified a performance bottleneck with Online Transaction Processing (OLTP) production host LUNs. The performance improvement plan includes a change to the Hitachi Dynamic Link…

Question

Options

• ALeast I/Os
• BLeast blocks
• CExtended Round Robin
• DExtended Least blocks

• A
  71% (39)
• B
  16% (9)
• C
  9% (5)
• D
  4% (2)

Explanation

Least I/Os is the correct choice for OLTP workloads because OLTP generates many small, random I/O operations, and this algorithm dynamically routes each new request to the path with the fewest outstanding I/Os - minimizing queue depth and latency where fast transaction response time is critical.

Least Blocks (B) is wrong because it distributes load based on the volume of data (blocks) in flight, making it better suited for large sequential workloads like data warehousing, not small OLTP transactions. Extended Round Robin (C) is wrong because it still rotates paths in sequence (with weighting), ignoring real-time path congestion - a poor fit when OLTP spikes can saturate individual paths unevenly. Extended Least Blocks (D) fails for the same reason as Least Blocks - it optimizes for throughput over transaction count, mismatching OLTP's I/O profile.

Memory tip: Match the metric to the workload - OLTP cares about Operations (Least I/Os), bulk/batch jobs care about data volume (Least Blocks). If you remember "OLTP = I/Os," you can eliminate the Blocks options and the rotation-based options immediately.

No community discussion yet for this question.