Question

What is the primary operational difference between Two-Phase Commit (2PC) and Three-Phase Commit (3PC) when a coordinator fails after sending a &#x27;prepare&#x27; message?

Accepted Answer

In a Two-Phase Commit protocol, if the coordinator fails after sending a prepare message, the system enters a state of uncertainty where participants are blocked. Because the participants do not know if the coordinator decided to commit or abort, they must wait for the coordinator to recover to avoid data inconsistency, as there is no way for the participants to safely reach a consensus on their own. In contrast, the Three-Phase Commit protocol introduces an intermediate state called pre-commit between the prepare and commit phases. This extra phase serves as a buffer that decouples the voting process from the final decision. If the coordinator fails after sending a prepare message in Three-Phase Commit, the participants can communicate with each other. If no participant has received a pre-commit message, they know that no node has committed, allowing them to safely elect a new coordinator and abort the transaction. This ability to reach a decision without waiting for the original coordinator is the primary operational difference that prevents the blocking problem found in Two-Phase Commit.

Home → All Courses → Engineering and Technology Courses → Distributed Systems for Internet-Scale Services → Flashcard

What is the primary operational difference between Two-Phase Commit (2PC) and Three-Phase Commit (3PC) when a coordinator fails after sending a 'prepare' message?