Build one guest, size it deliberately, and measure what a snapshot costs
Task
Create a virtual machine with every setting chosen rather than accepted, then take snapshots and measure the disk they consume -- because the growth is the thing that surprises people, and it is measurable in minutes.
Steps
- Record your sizing decisions BEFORE creating the guest in
lab/vm/sizing.csvwith headersetting,chosen,default,whycovering processors, memory, disk size, disk allocation type, firmware type and network mode. - Create the guest, install a small operating system, and record the host disk consumed by the guest's files in
lab/vm/size-1.txt. - Install the guest additions or tools, then record which of the integration features became available in
lab/vm/additions.csvwith headerfeature,before,after. - Take a snapshot, write at least 1 GB of data inside the guest, and record the host disk consumed again in
lab/vm/size-2.txt. - Take a second snapshot, write more data, record
lab/vm/size-3.txt, then delete both snapshots and record the final size and the time the merge took inlab/vm/merged.md.
Verify
awk -F, 'NR>1 && NF>=4 {n++} END {print n" sizing decision(s)"}' lab/vm/sizing.csv
awk -F, 'NR>1 && NF>=3 {n++} END {print n" integration feature(s)"}' lab/vm/additions.csv
ls lab/vm/size-1.txt lab/vm/size-2.txt lab/vm/size-3.txt | wc -l
grep -c . lab/vm/merged.md
Six sizing decisions each with a rejected default, three or more integration features compared, three size measurements, and a merge note with a duration. If the three sizes are identical, the snapshots were taken on a guest that then wrote nothing, and the growth this lab exists to show did not happen.
Notes
Delete the snapshots. A snapshot chain left in place grows without bound and, on a nearly full host, can reach the state where it cannot be deleted at all -- which is the corner this lab is warning about rather than one to occupy.
This is an independent study companion for CompTIA A+ Core 1 220-1201 and is not produced by or endorsed by CompTIA.