EPIC-5

in conjunction with the IEEE/ACM

International Symposium on Code Generation and Optimization (CGO'06)

Sunday, March 26, 2006

Manhattan, New York, NY


IMPORTANT FINAL SUBMISSION DEADLINE: January 27, 2006

The Explicitly Parallel Instruction Computing (EPIC) style of architecture was developed to enable new levels of instruction-level parallelism not achieved with traditional architectures. By allowing the compiler to express program parallelism and other relevant information directly to the processor, EPIC architectures can overcome hardware complexity issues that limit performance in traditional microprocessors. The major challenge in realizing the full potential of EPIC architectures is developing compiler and run time optimization technologies that effectively deploy explicitly defined hardware mechanisms, and deliver performance for both commercial and scientific applications. This workshop will focus on promising research concepts that enable the EPIC architecture model.

 

TOPICS OF INTEREST

Topics of interest include, but are not limited to:

Compiler Optimizations
    - Predicated execution
    - Compiler-directed control- and data speculation
    - Compiler controlled memory prefetching and memory hierarchy management
    - Validation of compiler optimizations

Feedback Directed Optimizations
    - Profile-driven compiler optimizations
    - EPIC performance monitoring unit feedback for dynamic compilation


Micro-architecture
    - Novel architectures and micro-architectures
    - In order versus out of order execution implementations
    - Multi-threaded, and multi-core EPIC architectures

New Uses
    - Power and energy aware computing techniques for EPIC machines
    - Application of EPIC architectures to special purpose applications

Performance analysis of EPIC architectures
    - Commercial and scientific workload studies for EPIC models
    - Effects of architectural features on workload behavior
    - Experimental evaluation of Itanium microprocessors
    - Performance comparisons with other architectures
    - Tools for analysis, instrumentation, and architecture experiments
 


Preliminary Program
Sunday March 26, 8:00AM-12:15PM
Manhattan, New York, NY

Session 1

8:00-8:45AM

Keynote: Trends in Microprocessor Design and Organization –What are the compiler issues?

Jerry Huck, Hewlett-Packard Company

ABSTRACT

SLIDES

8:45-9:15AM

Heterogeneous Multi-Module Data Cache for VLIW Processors
Enric Gibert (1,2), Jaume Abella (1,2), Jesús Sánchez (2), Xavier Vera (2) and Antonio González (1,2)

1) Computer Architecture Department, Universitat Politčcnica de Catalunya (UPC), 2) Intel Barcelona Research Center (IBRC) – Intel Labs

ABSTRACT

SLIDES

9:15-9:45AM

Combining predication and software speculation with dynamic speculation
Ron Barnes, George Mason University

ABSTRACT

SLIDES

9:45 – 10:00AM: Break

Session 2

10:00-10:45AM

Keynote: Porting High-Performance Applications to Itanium® 2
Tommy Hoffner/Patrik Sandahl, Ericsson

ABSTRACT

SLIDES

10:45-11:15AM

Fusing Instructions to Reduce Resource Usage in If-Converted Region

Matthew Bridges, Princeton University

Howard Chen, Gerolf Hoflehner, Daniel Lavery, Intel Corporation

ABSTRACT

SLIDES

11:15-11:45AM

Efficent JavaTM Synchronization on the Itanium® 2 Processor

Ali-Reza Adl-Tabatabai, Brian R. Murphy, Tatiana Shpeisman, Intel Corporation

ABSTRACT

SLIDES

11:45-12:15PM

Profile-Directed Predicated Partial Dead Code Elimination

Shane Ryoo, Sain-Zee Ueng, and Wen-mei W. Hwu, University of Illinois

ABSTRACT

SLIDES


IMPORTANT DATES

 
Submission Deadline:              January 27, 2006
Acceptances Mailed:              February 20, 2006
Final Version Due:                  March 11, 2006
Workshop Date:                     March 26, 2006 (half day workshop)

SUBMISSION GUIDELINES


Full papers of up to 20 pages or extended abstracts of approximately 8 pages can be submitted. Clearly describe the nature of the work, its significance and the current status of the research. Include a title page containing the title of the paper, list of authors and their affiliations, addresses, telephone and fax numbers, email addresses and the name of the corresponding author.

 

PROGRAM CO-CHAIRS

Kalyan Muthukumar             kalyan.muthukumar@intel.com             Intel

Rick Hank                            rick.hank@hp.com                              Hewlett-Packard


Page maintained by Kalyan Muthukumar