1. (15%) True or False questions (Please justify your answer.) (a) A dynamic branch predictor is…
1. (15%) True or False questions (Please justify your answer.)
(a) A dynamic branch predictor is always better than a static one.
(b) In the MIPS pipeline, both reading and writing the register file can be done during half a clock cycle period. In particular reading is done in the first half while writing is done in the second half
(c) Dynamic (out-of-order) superscalar processors do not guarantee that instructions are written back in strict program order.
(d) Floating point data points are uniformly represented on the real line.
(e) A 2-way superscalar processor has a peak IPC of 0.5
(a) (6%) Other than finding a good parallel algorithm, please list two most significant impediments to obtain a speedup of Non a N-core processor.
(b) (11%) Suppose you have a 6-core chip that you want to program to solve a given problem with the following assumptions:
(i) You can use as few as 1 core or as many as 6 cores,
(ii) the problem requires 450,000 iterations of a main loop to complete;
(iii) each loop iteration requires 100 clock cycles;
(iv) any startup overhead can be ignored;
(v) if more than 1 core is used to solve the problem, communication overhead must be added to the total execution time;
(vi) an additional core would give an additional 10 cycles per iteration How many cores should be used to solve this problem?
(c) (6%) Assume that you have 10 cores that you can use to solve a problem whose 98% of the code is parallelizable. Can you get a speedup of 7? If so, how many cores are needed?
3. (12%) Consider a 100 MB/s magnetic hard drive having 4 double-sided platters (disks). The platters rotate at 7200 rpm. Each platter’s surface is divided into 16000 tracks and each track is divided into 200 sectors. The storage capacity of each sector is 4096 bytes. The actual average seek time is 4ms. Assume the controller overhead is zero and that the disk is initially idle.
(a) (4%) What is the total storage capacity of this hard drive?
(b) (4%) How much time does it take to read a 4 MB file with randomly scattered sectors?
(c) (4%) What is the speed-up achieved when reading the same 4 MB file after defragmentation?