The snapshot takes a copy, or "snapshot", of all the information used by the planning process so that the planning process can operate on a picture of the data frozen at a point in time, and report against that "snapshot" at any future time.
You can choose to lock tables during the snapshot process to guarantee consistent data.

Locking tables controls the degree of read consistency. With unlocked tables, you can still get a high level of data consistency. However, if you feel you must lock tables, tables lock for only a brief period, reducing inconvenience to users.

Example
Consider a situation where you have a subinventory with a quantity of 100 units, and a discrete job for 25 units. Between the time the snapshot reads the inventory information and the time the snapshot reads the discrete job information, 10 units are completed from the job. If you ran the snapshot and choose to lock tables, you would guarantee that the snapshot would read either 100 units in the subinventory and 25 units outstanding on the discrete job, or 110 units in the subinventory and 15 units outstanding on the discrete job.

If you ran the snapshot and selected not to lock tables, there is a chance the Snapshot would read 100 units in the subinventory and 15 units outstanding on the discrete job. The chances of inconsistencies caused by launching the snapshot without locked tables is remote. The results you get from locked tables are better than those you get from the unlocked tables. However, you can still choose to run the Planning Engine with unlocked tables.