(a) For this part of the problem, you can assume we are using the immediate update protocol. We saw that…

(a) For this part of the problem, you can assume we are using the immediate update protocol. We saw that UNDOs have to be done right to left i.e. later writes are undone first and then earlier writes. Give an example to show why it should be done this way by doing the following: i. Find a simple example where it does make a difference whether the UNDOs are done left to right or right to left. ii. Show what the results will be if the UNDOs are done right to left. iii. Show what the results will be if the UNDOs are done left to right. (b) For this part of the problem, you can assume we are using the immediate update protocol We saw that REDOs have to be done left to right i.e. earlier writes are redone first and then later writes. Give an example to show why it should be done this way by doing the following: i. Find a simple example where it does make a difference whether the REDOs are done left to right or right to left. ii. Show what the results will be if the REDOs are done left to right. iii. Show what the results will be if the REDOs are done right to left. (c) Explain why is it that if we are using the immediate update protocol we need to keep the old value (BFIM) of an item along with the new value (AFIM), but if we are using deferred update protocol, we need to only keep the new value. 3. (10 points) Suppose that we use the immediate update protocol for the schedule given in the last slide of the week 5 power point notes. (a) Show what will be the steps taken at the time of the recovery process i.e. specify, in the correct order, what are all the operations (undo and/or redo) which are carried out by showing what changes are made. No additional explanation is needed. (b) What are the final values of A and B; no explanation needed