| |
|
|
EECS290a, Fall
2002
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Professor: |
Kurt Keutzer
|
| When : |
Wednesdays, 2:00 - 4:00 pm (credited
with 2h) |
| Where : |
Cory
540AB |
| Contact : |
Matthias Gries , Cory
545W, Tel. 642-0412 Chidamber Kulkarni , Cory 545S, Tel. 642-2596 |
|
One
of the major challenges facing embedded system designers today
is the efficient mapping from a domain-specific application to an
equally domain-specific integrated circuit. This seminar explores
the premise that the key to this mapping is the development of
a model of the concurrent
system architecture of the application, a model
of the concurrent system
architecture of the target device, then translating
between the two models.
What is a concurrent system architecture? It is a generalization of the concepts of software architecture and design patterns to consider highly concurrent hardware-software systems. In the seminar we will develop concurrent system architectural models of applications, such as packet forwarding. We will also develop concurrent system architectures of integrated circuits that are targeted for specific applications, such as the Intel IXP1200. We will then investigate techniques for translating one model into another. Prerequisits: Students should have a firm grounding in one of: computer architecture, software architecture, or system applications. |
|
|
||
|---|---|---|
| Date |
Topic |
Slides (access restricted to CAD group) |
| 08/28 |
Introduction: programming platforms
and models, application - architecture mapping: Kurt Keutzer |
csa-intro.pdf
|
| 09/04 |
Models of computation, application - architecture mapping:
Stephen Neuendorffer |
moc-intro.pdf |
| 09/11 |
Click model of computation, Patterns
of Implementation: Andrew Mihal |
moc-click.pdf |
| 09/18 |
MPEG 2 decoder case study, Philips
Nexperia: Kees Vissers |
mpeg_Nexpera.pdf |
| 09/25 |
Layers of abstraction for applications
and architectures: Kurt Keutzer |
Kurt's
talk @ chess |
| 10/02 |
Programming NPs: Chidamber Kulkarni Architecture abstraction of the Intel IXP: Niraj Shah |
ixp_cport.pdf,
ixp_cport_grey.pdf abstraction_ixp.pdf |
| 10/09 |
Programming models for Voice over
Packet (VoP) gateways: Broadcom Calisto platform |
Calisto
@ HotChips14 |
| 10/16 |
Voice over IP (VoIP) overview:
Matthias Gries |
VoIP.pdf |
| 10/23 |
The Embedded Machine: Predictable, Portable Real-Time Code: Christoph Kirsch | abstract, e-machine.pdf |
| 10/30 |
Single-chip Gigabit Mixed-version IP Router on Virtex-II Pro: Gordon Brebner | abstract, xilinx_talk.pdf |
| 11/06 |
UCB Score: Stream Computations Organized for Reconfigurable Execution: Eylon Caspi | soc2002_score.pdf |
| 11/13 |
MIT Raw Architecture Workstation (RAW): Scott Weber, CMU iWarp Multiprocessor: Yujia Jin |
raw.pdf,
iWarp.pdf |
| 11/20 |
Programmable System-on-a-Chip (PSoC)
- Cypress semiconductor: Douglas Densmore |
psoc.pdf |
| 11/27 |
[holiday] |
|
| 12/04 |
Tensilica Xtensa processor: Andrew Mihal |
xtensa.pdf |
| 12/11 |
Summary: Kaushik Ravindran |
CSA_wrapup.pdf |
|
|
|
|
|
|
|
|
| CAD Group Home
|