Simon Fraser University CMPT 295 Assignment 9.
1. [5 marks] Results (data sets and graph) from our Lab 6 on Microbenchmarking
Data Sheet
Data set: Version 1 - Branch
N t1 t2 t3 t4 t5 μ
100 1216 1218 1249 1250 1267 1240
150 1218 1224 1228 1408 1425 1300.6
200 1397 1436 1451 1577 1604 1493
250 1380 1422 1437 1503 1541 1456.6
Data set: Version 2 – Conditional move
N t1 t2 t3 t4 t5 μ
...[Show More]
Simon Fraser University CMPT 295 Assignment 9.
1. [5 marks] Results (data sets and graph) from our Lab 6 on Microbenchmarking
Data Sheet
Data set: Version 1 - Branch
N t1 t2 t3 t4 t5 μ
100 1216 1218 1249 1250 1267 1240
150 1218 1224 1228 1408 1425 1300.6
200 1397 1436 1451 1577 1604 1493
250 1380 1422 1437 1503 1541 1456.6
Data set: Version 2 – Conditional move
N t1 t2 t3 t4 t5 μ
100 1127 1273 1319 1479 1756 1390.8
150 1197 1302 1410 1576 1604 1417.8
CMPT 295 – Spring 2022
200 1245 1250 1251 1266 1283 1259
250 1290 1343 1354 1382 1485 1370.8
80 100 120 140 160 180 200 220 240 260
0
200
400
600
800
1000
1200
1400
1600
f(x) = - 0.44 x + 1436.18
f(x) = 1.68 x + 1077.78
Average Clock Cycle
Branch Linear (Branch)
Conditional move Linear (Conditional move)
μ
2. [8 marks] - Performance analysis of Pipelined Instruction Execution microprocessor (uniform and non-uniform stage lengths)
Part 1
- minimum clock cycle time: 600/8 ps + 30 ps = 75 ps + 30 ps = 105 ps
- latency: (600/8 ps + 30 ps) 8 = (105 ps) 8 = 840 ps, 1 intruction takes 840 ps to execute, to go through all 8 stages of its execution
[Show Less]