Explain load average versus CPU utilization and why the distinction matters.

Load average is the exponentially-weighted count of processes that are runnable or in uninterruptible sleep (often disk or I/O wait) over 1, 5, and 15 minutes, whereas CPU utilization is the percentage of time cores are actually executing.

You compare load to core count: a load of 4 on 4 cores is saturated, on 8 cores it's fine. The key insight is that high load with low CPU usually means I/O wait or lock contention, not compute, so you'd reach for iostat and thread dumps rather than profiling CPU; conversely high CPU with modest load points at a few hot threads.