ECS 201A Final: ECS201A Final Exam Fall 2013
9 Jan 2019
School
Department
Course
Professor
Document Summary
Which is more sensitive to where data is placed, an uma machine or a. Why not just fabricate one huge die per wafer? (2) prefetching is one technique for reducing the miss rate. List 2 others. (2) lowering the associativity is one technique for reducing the hit time. If so, explain your answer. (3) writes are inherently slower than reads. What two principles make it work? (two types of the same principle, actually). How does it work? (4) some instruction sets are harder to write virtual machine monitors for than others. 4 - (4) in class we mentioned vliw and superscalar as two ways to circumvent the flynn limit of 1. Give 1 advantage and 1 disadvantage to using a mimd processor. (4) supporting precise interrupts in machines that allow out of order completion is a challenge. List 3 existing dynamic branch prediction strategies in order of (average) decreasing effectiveness. (4) compare and contrast superscalar and vliw.
