12

In 2PL (two phase locking), what advantage(s) does the rigorous model have over the strict model?

I) There is no advantage over the strict model.

II) In contrast to the strict model, it guarantees that starvation cannot occur.

III) In contrast to the strict model, it guarantees that deadlock cannot occur.

IV) In contrast to the strict model, there is no need to predict data needed in the future.

My note says all of the above are false. I am a bit confused. Can someone clarify for me why all of this is false?

Paul Griffin
  • 2,416
  • 15
  • 19
  • 2
    Rigorous 2PL just means that all locks are released after the transaction ends (as opposed to strict where read-only locks may be released earlier). This doesn't affect deadlocks as those occur in the expanding phase. The point IV might be partially true (depending on the point of view): if you release all the read locks at the end of a transaction then the transaction never even has to consider if it should release a read lock or keep it for later reads in the transaction. – Simo Kivistö Apr 19 '15 at 19:10
  • @SimoKivistö would you please post it as an answer? –  Apr 19 '15 at 21:24
  • Expanded and posted as answer. – Simo Kivistö Apr 20 '15 at 13:55

4 Answers4

17

What is Two-Phase Locking (2PL) Protocol ?

A transaction is two-phase locked if:

before reading x, it sets a read lock on x

before writing x, it sets a write lock on x

it holds each lock until after it executes the corresponding operation

after its first unlock operation, it requests no new locks

Now, what is Strict phase locking ?

Here a transaction must hold all its exclusive locks till it commits/aborts.

But ,whats rigorous 2PL ?

Rigorous two-phase locking is even stricter: here all locks are held till commit/abort. In this protocol transactions can be serialized in the order in which they commit.

Much deeper :

Strict 2PL :

enter image description here

Same as 2PL but Hold all exclusive locks until the transaction has already successfully committed or aborted. –It guarantees cascadeless recoverability

Rigorous 2PL :

enter image description here

Same as Strict 2PL but Hold all locks until the transaction has already successfully committed or aborted. –It is used in dynamic environments where data access patterns are not known before hand.

There is no deadlock. Also, a younger transaction requesting an item held by an older transaction is aborted and restart with the same timestamp, starvation is avoided.

I hope that above clear explanations with diagram must have made you clear about the concept and advantages of rigorous over the other.

Thanks

J. Win.
  • 6,662
  • 7
  • 34
  • 52
Tharif
  • 13,794
  • 9
  • 55
  • 77
6

I - there is an advantage

Look at this lecture note from UCLA:

Rigorous two-phase locking has the advantages of strict 2PL. In addition it has the property that for two conflicting transactions, their commit order is their serializability order. In some systems users might expect this behavior.

These lecture notes have an example (the model in the example is strict - not rigorous):

Consider two transactions conducted at the same site in which a long running transaction T1 which reads x is ordered before a short transaction T2 that writes x. T2 returns first, showing an update version of x long before T1 completes based on the old version.

II and III - does not affect deadlocks/starvation

Rigorous 2PL means that all locks are released after the transaction ends as opposed to strict where read-only locks may be released earlier. This doesn't affect deadlocks or starvation as those occur in the expanding phase (a transaction cannot acquire the needed lock). In a deadlock both processes are always in the expanding phase.

IV - both need to know the needed data for locking in the expanding phase - shrinking phase varies

Strict: I don't know the usual implementation details of strict 2PL but if a read lock is released before a transaction ends there has to be a knowledge (100% sure prediction if you like) that the lock is not needed later in the transaction.
Rigorous: All the read locks are released at the end of a transaction and the transaction never has to evaluate if it should release a read lock or keep it for later reads in the transaction.

Is rigorous or strict more used/preferred?

Which of those two models to use would depend on the situation. Modern DBMS use more complex concurrency handling than simple rigorous or strict 2PL. Having said that judging by the Wikipedia article on two-phase locking the rigorous (SS2PL) model is more widely used:

SS2PL [rigorous] has been the concurrency control protocol of choice for most database systems and utilized since their early days in the 1970s. [...]

2PL in its general form, as well as when combined with Strictness, i.e., Strict 2PL (S2PL), are not known to be utilized in practice. The popular SS2PL does not require marking "end of phase-1" as 2PL and S2PL do, and thus is simpler to implement. Also, unlike the general 2PL, SS2PL provides, as mentioned above, the useful Strictness and Commitment ordering properties. [...]

SS2PL Vs. S2PL: Both provide Serializability and Strictness. Since S2PL is a super class of SS2PL it may, in principle, provide more concurrency. However, no concurrency advantage is typically practically noticed (exactly same locking exists for both, with practically not much earlier lock release for S2PL), and the overhead of dealing with an end-of-phase-1 mechanism in S2PL, separate from transaction-end, is not justified. Also, while SS2PL provides Commitment ordering, S2PL does not. This explains the preference of SS2PL over S2PL. [...]

Simo Kivistö
  • 4,247
  • 3
  • 38
  • 42
  • Added some info to the answer. _Simplified:_ rigorous is preferred. – Simo Kivistö Apr 22 '15 at 20:02
  • my prof. wrote (4) is true. would you please explain why (4) is true? –  Apr 23 '15 at 06:55
  • Well, if _"no need to predict data that need in future"_ means _"no need to predict data that **the current statement** needs later"_ it's because strict _might_ release read_locks before the end of transaction. Therefore it needs to know the possible future use of those rows in the transaction (if they are read later in the same transaction the locks cannot be released). If _"predict data that need in future"_ in the question means something else then you would have to clarify that part of the question. – Simo Kivistö Apr 23 '15 at 07:30
  • need clarification, which one is True? –  Apr 24 '15 at 19:42
  • My problem is in S2PL we dont need to to predict data needed in the future. Isnt it ? –  May 02 '15 at 08:00
1

Transaction T2 in the above example does not follow 2PL and S2PL as a lock request (lock B) is done after the lock release unlock(A) - hence violating the protocol.

user5332534
  • 11
  • 1
1

Rigorous two phase locking is similar to strict two phase locking with two major differences:

  1. In strict two phase locking the shared locks are released in shrinking phase, but in rigorous two phase locking all the shared and exclusive locks are kept until the end of the transaction.
  2. In rigorous two phase locking we do not need to know the access pattern of locks on data items beforehand so it is more appropriate for dynamic environments while in strict two phase locking the access pattern of locks should be specified at the start of transaction.

So the fourth option is the correct one.

HMD
  • 2,202
  • 6
  • 24
  • 37
Alireza Kavianifar
  • 11
  • 2