Skip to main content
Department of Information Technology

AVDARK reading instructions for 4:th edition of the book

Sections marked RT are "read through" sections, i.e., understand the overall message but do not concentrate on any details. You should be able to answer simple basic questions about these topics.

All the slides shown in class are to be considered as course material, unless stated otherwise. Note that htis includes the sldies by the guest lecturers as well. All slides are available under Slides in the menue.

Caches and Memory

All of Appendix C
(not Miss Penalty for Out-of-Order Execution on pg C19-C21)
(not A Segmented Memory examples pg C49-C52)

5.1 All
5.2 The 7th, 9th, 10th, 11th optimization and fig 5.11 which are included. The rest is RT.
5.3 All
5.4 Not included in course, but is very interesting!
5.5 Not included
5.6 All (Performance of the Opteron.. pg331-335 is RT)
The rest of Chapter 5 is not included

Slides by Virtutech

Multiprocessors

The paper by Adve on Memory Models is included.
The Scott paper (except the issues that deal with timeout) is included.
The paper by Hagersten on WildFire is RT.
4.1 All
4.2 All (not from Performance Measurement on Commercial Workload on pg 231 to the end of 4.2.)
4.3 Only introduction (nothing from the Workload sections)
4.4 All
4.5 All
4.6 All
4.7 Not included
4.8 Is RT (very interesting! More about this issue on Futures!)
4.9 Is RT
4.10 Is RT

CPU and Pipelining

Appendix A (A.3, A.7-9 is RT) (A.6 not included)
Appendix B is RT
Ch 2 (2.5, 2.10 is RT)

Network processors

Slides by Jakob

Programming MPs

Slides by Sverker

Future

Chapter 1
The Agarwal paper (the trends but not the formulas).
3.5

GPUs

Slides by David

Updated  2010-12-13 09:58:42 by Erik Hagersten.