Make a machine slow four different ways and produce the recommendation each needs

short · 40 min · Objective 3.1

Task

Create four distinct software causes of slowness on the guest and write the measurement and recommendation for each. The recommendations are the deliverable, and each one must state what it will not fix -- which is the sentence that distinguishes a diagnosis from a sales pitch.

Steps

  1. Fill the guest's system volume to under five per cent free, measure, and record in lab/slow/cases.csv with header cause,symptom,measurement,value,recommendation,will_not_fix.
  2. Add eight startup items, restart, time the interval from sign-in to a usable desktop, and add the row.
  3. Start a background job that saturates the disk, reproduce ordinary work, and add the row including queue length and response time.
  4. Reduce memory until hard faults are sustained, and add the row including the hard fault rate.
  5. Write lab/slow/hardware.md giving the measurements that would justify a hardware recommendation, and the ones that would not -- specifically why a high memory percentage alone does not.

Verify

awk -F, 'NR>1 && NF>=6 {n++} END {print n" case(s)"}' lab/slow/cases.csv
awk -F, 'NR>1 && $6!="" {n++} END {print n" recommendation(s) with a negative clause"}' lab/slow/cases.csv
awk -F, 'NR>1 {print $3}' lab/slow/cases.csv | sort -u | wc -l
grep -Eic 'hard fault' lab/slow/cases.csv lab/slow/hardware.md
grep -Eic 'percentage|cache|committed' lab/slow/hardware.md

Four cases with four distinct measurements, every recommendation carrying a negative clause, and hard faults recorded. A recommendation with an empty will-not-fix column has not been thought through.

Notes

The memory case is the one that produces the wrong recommendation most often in real life, because the disk is pinned and the obvious conclusion is a new disk. Hard faults are what distinguish them.

This is an independent study companion for CompTIA A+ Core 2 220-1202 and is not produced by or endorsed by CompTIA.