1 / 5

Dan Connors, Univ. of Colorado Tom Conte, North Carolina State Univ. Phil Emma, IBM Research

Dan Connors, Univ. of Colorado Tom Conte, North Carolina State Univ. Phil Emma, IBM Research Konrad Lai, MRL/MTL Intel Scott McFarling, Microsoft Chuck Moore, AMD Yale N. Patt, Univ. of Texas at Austin Jim Smith, Univ. of Wisconsin Jared Stark, MRL/MTL Intel

daleedwards
Download Presentation

Dan Connors, Univ. of Colorado Tom Conte, North Carolina State Univ. Phil Emma, IBM Research

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Dan Connors, Univ. of Colorado Tom Conte, North Carolina State Univ. Phil Emma, IBM Research Konrad Lai, MRL/MTL Intel Scott McFarling, Microsoft Chuck Moore, AMD Yale N. Patt, Univ. of Texas at Austin Jim Smith, Univ. of Wisconsin Jared Stark, MRL/MTL Intel Mateo Valero, Univ. Politecnica Catalunya Chris Wilkerson, MRL/MTL Intel The Journal of Instruction Level ParallelismChampionship Branch Predictionwebsite: http://www.jilp.org/cbp

  2. Introduction: Competition Overview and Workshop Agenda Jared Stark, MRL/MTL Intel

  3. Overview • Objective: Compare branch predictors in a common framework. • Framework • Traces • 30M inst long, includes user + system activity • 2 trace sets: distributed + undistributed • 20 traces / set: 5 INT, 5 FP, 5 MEDIA, 5 SERVER • Driver • Static Info: PC, type (BR, ALU, ...), reg src/dst specifiers • Dynamic Info: results, LD/ST addresses, branch outcomes

  4. Overview (cont.) • Budget: (64K + 256) bits; no attempt to assess other costs (power, latency, design effort, ...) • Performance: TK/NT accuracy only; accuracy in mispredicts per 1000 inst • Competition has 2 rounds: • Initial: 6 predictors w/ lowest avg mispredict rate on distributed trace set selected as finalists • Final: Finalists ranked and champion declared using the undistributed trace set

  5. Agenda 1:00 Introduction: Competition Overview and Workshop Agenda, Jared Stark, MRL/MTL Intel 1:10 Perceptrons for Dummies, Daniel A. Jiménez, Rutgers University 1:30 Idealized Piecewise Linear Branch Prediction, Daniel A. Jiménez, Rutgers University 1:50 A PPM-like, Tag-based Predictor, Pierre Michaud, IRISA/INRIA 2:10 Adaptive Information Processing: An Effective Way to Improve Perceptron Predictors Hongliang Gao and Huiyang Zhou, University of Central Florida 2:30 Break 3:00 A 2bcgskew Predictor Fused by a Redundant History Skewed Perceptron Predictor Veerle Desmet, Hans Vandierendonck, and Koen De Bosschere, Ghent University 3:20 The O-GEHL Branch Predictor, André Seznec, IRISA/INRIA 3:40 The Frankenpredictor, Gabriel Loh, Georgia Institute of Technology 4:00 Branch Prediction Caveats and Second-Order Effects, Phil Emma, IBM Research 4:20 Conclusion: Ranking of the Finalists, Anointing of the Champion, and “What Next?” Chris Wilkerson,MRL/MTL Intel 4:30 Adjourn

More Related