cinera_handmade.network/miotatsu/riscy/coad/coad013.hmml

81 lines
5.6 KiB
Plaintext

[video member=miotatsu stream_platform=twitch project=book title="1.13 Exercises 12-15" vod_platform=youtube id=Qsz_j4gmzL0 annotator=Miblo]
[0:06][Recap and set the stage for the day]
[1:34][Chapter 1.13, Exercise 1.12 - The pitfall of utilizing a subset of the :performance equations[ref
title="Computer Organization and Design RISC-V Edition"
author="David Patterson & John Hennessy"
publisher="Morgan Kaufmann"
isbn=9780128122754
url=https://www.elsevier.com/books/computer-organization-and-design-risc-v-edition/patterson/978-0-12-812275-4]][:research]
[4:01][Chapter 1.13, Exercise 1.12.1 - Fallaciously equating clock rate with :performance]
[6:45][A few words on pcalc, Unix sockets and TCP]
[7:13][Chapter 1.13, Exercise 1.12.1 continued][:performance]
[17:04][Chapter 1.13, Exercise 1.12.2 Fallaciously equating instruction count to CPU time][:performance]
[27:39][A few thoughts on processor marketing and :performance benchmarks]
[35:37][Chapter 1.13, Exercise 1.12.3 Fallaciously using MIPS to compare two different processors][:blackboard :performance]
[45:40][Chapter 1.13, Exercise 1.12.4 Fallaciously using MFLOPS to compare two different processors[ref
site="RISCY BUSINESS Forums"
page="COAD 1.13 Exercises"
url=https://riscy.handmade.network/forums/t/2546-coad_1.13_exercises]][:blackboard :performance]
[1:02:32][Chapter 1.13, Exercise 1.13 - The pitfall of expecting to improve the overall :performance of a computer by improving only one aspect of it[ref
title="Computer Organization and Design RISC-V Edition"
author="David Patterson & John Hennessy"
publisher="Morgan Kaufmann"
isbn=9780128122754
url=https://www.elsevier.com/books/computer-organization-and-design-risc-v-edition/patterson/978-0-12-812275-4]][:research]
[1:04:20][Chapter 1.13, Exercise 1.13 - The metrics under consideration][:blackboard :performance]
[1:10:34][A few thoughts on homework questions][:rant]
[1:17:13][Amdahl's Law][:blackboard :performance]
[1:22:08][Chapter 1.13, Exercise 1.13.1 - Calculating the total time reduction for a 20% reduction in time for floating point operations][:blackboard :performance]
[1:31:08][@croepha][Hello]
[1:31:19][Chapter 1.10 - Fallacies and Pitfalls, reviewing Amdahl's Law[ref
title="Computer Organization and Design RISC-V Edition"
author="David Patterson & John Hennessy"
publisher="Morgan Kaufmann"
isbn=9780128122754
url=https://www.elsevier.com/books/computer-organization-and-design-risc-v-edition/patterson/978-0-12-812275-4]][:research]
[1:34:47][Chapter 1.13, Exercise 1.13.1 continued][:blackboard :performance]
[1:37:02][Amdahl's Law[ref
site=Wikipedia
page="Amdahl's law"
url=https://en.wikipedia.org/wiki/Amdahl's_law]][:research]
[1:45:06][Chapter 1.13, Exercise 1.13.1 continued][:blackboard :performance]
[1:58:00][Relating our intuitive equation for Amdahl's Law back to that given in the book[ref
site=YouTube
page="Book Club - Day 7: COAD 1.10-1.11, Check Yourself Review"
url=https://www.youtube.com/watch?v=01mtcp0lb4s]][:blackboard]
[2:18:06][Work through the algebra for Amdahl's Law][:blackboard :performance]
[2:30:30][Request for help clarifying Amdahl's Law[ref
site="RISCY BUSINESS Forums"
page="COAD 1.13 Exercises"
url=https://riscy.handmade.network/forums/t/2546-coad_1.13_exercises]]
[2:31:36][Chapter 1.13, Exercise 1.13.1 concluded][:blackboard :performance]
[2:32:17][Chapter 1.13, Exercise 1.13.2 - Calculating the time reduction for integer operations given a total time reduction of 20%][:blackboard :performance]
[2:35:26][Consult the Errata[ref
site=Elsevier
page="Computer Organization and Design, 5th Edition: Errata"
url=http://booksite.elsevier.com/9780124077263/downloads/COD5e_errata_09-2014.pdf]]
[2:39:43][Chapter 1.13, Exercise 1.13.2 continued][:blackboard :performance]
[2:44:10][Chapter 1.13, Exercise 1.13.3 - Trying to reduce the total time by 20% by reducing only the time for branch instructions][:blackboard :performance]
[2:45:34][Chapter 1.13, Exercise 1.14 - Improving the execution time of a given program[ref
title="Computer Organization and Design RISC-V Edition"
author="David Patterson & John Hennessy"
publisher="Morgan Kaufmann"
isbn=9780128122754
url=https://www.elsevier.com/books/computer-organization-and-design-risc-v-edition/patterson/978-0-12-812275-4]][:research :performance]
[2:47:10][Chapter 1.13, Exercise 1.14 - The metrics under consideration][:blackboard :performance]
[2:58:11][Chapter 1.13, Exercise 1.14.1 - Improving the CPI of floating point instructions to make the program run two times faster][:blackboard :performance]
[3:01:07][Chapter 1.13, Exercise 1.14.2 - Improving the CPI of load / store instructions to make the program run two times faster][:blackboard :performance]
[3:07:55][Chapter 1.13, Exercise 1.14.3 - Calculating the execution time improvement given a reduced CPI of integer and floating point instructions by 40%, and load / store and branch instructions by 30%][:blackboard :performance]
[3:16:50][Chapter 1.13, Exercise 1.15 - Parallel processing[ref
title="Computer Organization and Design RISC-V Edition"
author="David Patterson & John Hennessy"
publisher="Morgan Kaufmann"
isbn=9780128122754
url=https://www.elsevier.com/books/computer-organization-and-design-risc-v-edition/patterson/978-0-12-812275-4]][:research :performance]
[3:18:09][Break]
[3:21:20][:afk]
[3:25:59][Return with nourishment]
[3:27:38][Chapter 1.13, Exercise 1.15 - Computing :performance metrics for a program running on multiple parallel processors][:blackboard :performance]
[3:50:37][Reflect on our experience so far, and look forward to Chapter 2]
[/video]