6.4Recoverable, cascadeless and strict schedules
The idea
Recoverability is about commits. Recoverable: a reader commits after its writer. Cascadeless: no dirty reads. Strict: no dirty reads or overwrites.
Think of it like this: Copying a classmate's homework: if their answers get cancelled later, yours must not already be marked final.
Why it exists
You classify a schedule that includes commit and abort operations.
How it works
- Recoverable: if Tj reads from Ti then Ci comes before Cj.
- Cascadeless: Tj reads an item written by Ti only after Ti commits.
- Strict: no read or write of an item written by Ti until Ti commits or aborts.
- Strict ⊂ cascadeless ⊂ recoverable.
Worked example: w1(A) r2(A) c2 c1
T2 reads from T1 but commits first: not recoverable, hence also not cascadeless and not strict.
Now play with it
- →Select "Cascadeless but not strict" and spot the overwrite.
- →Reorder the commits to make a schedule recoverable.
| time → | 1 | 2 | 3 | 4 |
|---|---|---|---|---|
| T1 | W(A) | Commit | ||
| T2 | R(A) | Commit |
Strict ⊂ cascadeless ⊂ recoverable. Transactions without a commit or abort in the text are assumed to commit at the end.
Check yourself
w1(A) w2(A) c1 c2 is
Exam notes (the dense version)
Recoverable: if Tj reads from Ti then Ti commits before Tj. Cascadeless (ACA): transactions read only committed data (no dirty reads), so there are no cascading aborts. Strict: no item written by Ti is read or overwritten until Ti commits or aborts. Strict ⊂ cascadeless ⊂ recoverable. Serializability and recoverability are independent properties.
Previous year questions: Transactions & Concurrency Control (15)
PYQs are grouped by topic, so this is the full set for Transactions & Concurrency Control; try the ones that match this subtopic. Open all 15 on the topic page.