Students that did the "Distributed Systems" Course from the BSc at IST have been exposed to an algorithm named "Quorum Consensus Replication". I am attaching some documentation that has been distributed in that course regarding this algorithm.


It is important to notice that the algorithm is very old (was designed more than 30 years ago) and, at that time, the taxonomy of distributed system was still confusing. In fact, this algorithm solves an important problem but does NOT solve the consensus problem we have been addressing in the last classes. 

You should try to look at the algorithm and answer the following questions:

- Why "Quorum Consensus Replication" does not solve consensus?

- What problem is trying to solve? Did we discuss that problem in the DAD classes? In which class?

Attachments