Suppose you have a large relation r(A, B, C) and a materialized view v = A sum

Suppose you have a large relation r(A, B, C) and a materialized view v = A_Уsum(B)(r). View maintenance can be performed as part of each transaction that updates r, on a parallel/distributed storage system that supports transactions across multiple nodes. Suppose the system uses two-phase commit along with a consensus protocol such as Paxos, across geographically distributed data centers.

a. Explain why it is not a good idea to perform view maintenance as part of the update transaction, if some values of attribute A are “hot” at certain points in time, that is, many updates pertain to those values of A.

b. Explain how operation locking (if supported) could solve this problem.

c. Explain the tradeoffs of using asynchronous view maintenance in this context.