 |
CS6290/CS4290 Fall 2001 Project 3: Final Project |
Here are support files for the assignment. Watch the space below for
bugfixes and other notes.
Projects:
Monday 3-4 in CoC 017
---------------------
Johnson: Genetic search of cache designs
Balakrishnan/Singh: Redo Emer97 experiment
Martinez: PIII vs. Athalon benchmarking
Collins/Jayaraman: Data and value prediction
Gray/Wolenetz: CPU2000 characterization
Tuesday 1:30-2:30 in MiRC 102A
------------------------------
Redd: Itanium benchmarking
Hooda/Trivedi: OOO scheduling in simplescalar
Elmore: Predicting multiple branches
Anderson: Dataflow simulator
Singaravelu/Sharma: Branch difference predictor
Wednesday 3-4 in CoC 017
------------------------
Kong/Li/Zhang: IXP
Burger: Parallelization of video texture
Koh/Kang: DSP (survey)
Jeong/Jun: Caching w/page mapping in Shade
Cao: Branch prediction using Shade
Pivoshenko/Ettibaryan: RAID (survey)
Friday 3-4 in CoC 017
---------------------
Huang/Tang: Branch predictors
Liogkas/Somani: Tag bits for generational GC
Viswanath/Agarwala: IXP for SYN-flood problem
Gupta/Shariff: Cache/VM w/remote pages in Shade
Dagon/McNeely: ILP in Java bytecodes
Notes:
- For the Shade system, there are two useful things in the
/net/projects/class/6290/fall01/ directory. The first is the
shade distribution which includes some documentation. The second is
a project from a previous year's 6290 that used Shade to build branch
predictors: see the directory old-project2
- For the SimpleScalar simulator, the source distribution is in
/net/hp82/softarch/SIMULATORS/archive/simplesim-3.0b.tar.gz.
There's also a pre-built compiler for the Sparc in
/net/hp82/softarch/SIMULATORS/nsea/bin/ssbig-na-sstrix-gcc
- For SPEC benchmarks, see
/net/hp82/softarch/BENCHMARKS/spec/cpu2000-tar.gz.
- For the IXP1200, start with
/net/hp82/softarch/CAD/Intel/ixp_cd_v1.3.141/documentation/,
particularly the "HW Ref Manual" and the "Prog Ref Manual", then see
me for the CD with the tools.
Preferred machines:
The Sun "cities" cluster in the CoC 1st floor, e.g.
---------------------------------------------------------------------
| akron | | first-floor cluster |
| boise | Ultra 10, 256MB | (always defer to |
| cheyenne | | anyone at the console) |
| dallas | | |
| ... | | |
---------------------------------------------------------------------
Please observe the following policy when using machines
remotely:
- Defer to the console: Don't use a
public-cluster machine remotely unless there's no one at the console.
- Use nice +19: On any machine, preface any job
you expect to run for more than about 10 seconds with nice +19
to voluntarily lower its priority.
- Limit scripts: Avoid launching unattended jobs
(e.g., scripts of multiple simulation) that you expected to run for
more than about a half an hour.
- No parallel processing: limit yourself to
at most one 6290-related simulation per person active at a time in the
CoC.
Ken Mackenzie,
$Date: 2001/10/22 13:43:31 $