HPCA-7 Advance Program



Saturday, 1/20/2001

  Workshops (times to be announced, each is a full day).

    1. "Interact-V: Workshop on Interaction between Compilers and Computer Architectures"
        by Gyungho Lee and Pen-Chung Yew


Sunday, 1/21/2001

  Workshops (times to be announced, each is a full day).

    2. "Fourth Workshop on Computer Architecture Evaluation using Commercial Workloads"
        by Russell Clapp, Kimberly Keeton, Ashwini Nanda, and Josep Torrellas

    (Our apologies, this workshop has been cancelled)
    3. "Microprocessors for Networks and Communications"
        by Ilan Spillinger and Mario Nemirovsky


  Tutorials (times to be announced)

  Morning

    VIA and InfiniBand Communication Architecture
    Prof. Dhabaleswar K. Panda, The Ohio State University
    Email contacts: panda@cis.ohio-state.edu

  Afternoon

    Power-Performance Modeling, Analyis and Validation
    Margaret Martonosi, David Brooks (Princeton University) and Pradip Bose (IBM T. J. Watson Research Center)
    Email contacts: mrm@ee.princeton.edu, dbrooks@ee.princeton.edu, pbose@us.ibm.com


Monday, 1/22/2001

Conference Program

8:30 – 9:00Breakfast
 
9:00 –  9:15Welcome
 
9:15 – 10:15Keynote Address (Ted Selker)
 
10:15 – 10:45Break
 
10:45 – 12:45Session 1: Microarchitecture – I (Chair: Shubu Mukherjee)
 
 "Stack Value File: Custom Microarchitecture for the Stack"
 Hsien-Hsin Lee, Smelyanskiy Mikhail, Chris Newburn, and Gary Tyson, University of Michigan and Intel Corp
 
 "Register Renaming and Scheduling for Dynamic Execution of Predicated Code"
 Perry Wang, Hong Wang, Ralph Kling, Kalpana Ramakrishnan, John Shen, and Wen-Hann Wang, Intel
 
 "Data-flow Prescheduling for Large Instruction Windows in Out-of-Order Processors"
 Pierre Michaud and Andrι Seznec, IRISA/INRIA
 
 "Speculative Data-Driven Multithreading"
 Amir Roth and Gurindar Sohi, University of Wisconsin at Madison
 
12:45 – 2:00Lunch (Sala Mayor de Rectoria)
 
2:00 – 3:30Session 2: Memory Architectures (Chair: John Carter)
 
 "Towards Virtually-Addressed Memory Hierarchies"
 Xiaogang Qiu and Michel Dubois, Sun Microsystems and University of Southern California
 
 "Reevaluating Superpage Promotion with Hardware Support"
 Zhen Fang, Lixin Zhang, John Carter, Wilson Hsieh, and Sally McKee, University of Utah
 
 "Performance of Hardware Compressed Main Memory"
 Bulent Abali, Hubertus Franke, Xiaowei Shen, Dan E. Poff, and T. Basil Smith, IBM T.J. Watson Research Center
 
3:30 – 4:00Coffee Break
 
4:00 – 5:30Session 3: Multiprocessor Systems (Chair: Steve Lumetta)
 
 "JETTY: Filtering Snoops for Reduced Power Consumption in SMP Servers"
 Andreas Moshovos, Gokhan Memik, Babak Falsafi, and Alok Choudhary, University of Toronto and Carnegie Mellon University
 
 "A New Scalable Directory Architecture for Large-Scale Multiprocessors"
 Manuel Acacio, Jose Gonzalez, Jose Garcia, Jose Duato, Universities of Murcia and Valencia
 
 "Self-Tuned Congestion Control for Multiprocessor Networks"
 Mithuna Thottethodi, Alvin Lebeck, and Shubhendu Mukherjee, Duke University and Compaq Computer Corporation
 
5:30 – 5:45Break
 
5:45 – 6:45Session 4: Code Generation Techniques (Chair: Mateo Valero)
 
 "Automatically Mapping Code in an Intelligent Memory Architecture"
 Jaejin Lee, Yan Solihin, and Josep Torrellas, Michigan State University, Los Alamos National Laboratory, and University of Illinois at Urbana-Champaign
 
 "CARS: A New Code Generation Framework for Clustered ILP Processors"
 Krishnan Kailas, Kemal Ebcioglu, and Ashok Agrawala, University of Maryland at College Park and IBM T.J.Watson Research Center
 
7:00 – 8:00IEEE TCCA Business Meeting (ITESM)
 
8:30 – 10:00Conference Reception (Museo Historia de Mexico)



Tuesday, 1/23/2001

Conference Program

9:00 – 9:30Breakfast
 
9:30 – 10:30Invited Lecture
 "Performance SPECulations - Benchmarks, Friend or Foe"
 Kaivalya Dixit
 
10:30 – 11:00Coffee Break
 
11:00 – 12:30Session 5: Energy and Thermal Management (Chair: Yale Patt)
 
 "An Integrated Circuit/Architecture Approach to Reducing Leakage in Deep-Submicron High-Performance I-Caches"
 Se-Hyun Yang, Michael Powell, Babak Falsafi, Kaushik Roy, and T. Vijaykumar, Purdue University and Carnegie Mellon University
 
 "DRAM Energy Management Using Software and Hardware Directed Power Mode Control"
 Victor Delaluz, Mahmut Kandemir, Vijaykrishnan N., Anand Sivasubramaniam, and Mary Jane Irwin, Penn State University
 
 "Dynamic Thermal Management for High-Performance Microprocessors"
 David Brooks and Margaret Martonosi, Princeton University
 
12:30 – 2:00Lunch (Sala Mayor de Rectoria)
 
2:00 – 3:30Session 6: Prediction Techniques (Chair: Lawrence Rauchwerger)
 
 "Dynamic Prediction of the Critical Performance Path"
 Eric Tune, Dongning Liang, Dean Tullsen, and Brad Calder, UCSD
 
 "Dynamic Branch Prediction with Perceptrons"
 Daniel Jimenez and Calvin Lin, University of Texas at Austin
 
 "Differential FCM: Increasing Value Prediction Accuracy by Improving Table Usage Efficiency"
 Bart Goeman, Hans Van Dierendonck, and Koen DeBosschere, Ghent University
 
3:30 – 4:00Coffee Break
 
4:00 – 5:30Session 7: Application-specific Designs (Chair: Steve Lumetta)
 
 "DLP + TLP Processors for the Next Generation of Media Workloads"
 Jesus Corbal, Roger Espasa, and Mateo Valero, Universitat Politecnica de Catalunya
 
 "An Architectural Evaluation of Java TPC-W"
 Harold Cain, Ravi Rajwar, Morris Marden, and Mikko Lipasti, University of Wisconsin
 
 "A Programmable Co-processor for Profiling"
 Craig Zilles and Gurindar Sohi, University of Wisconsin at Madison
 
5:30-5:45Break
 
5:45 – 7:30Second Annual HPCA Work In Progress Session (Chair: Jose Duato)
 
8:00 – 10:00Banquet (ITESM)



Wednesday, 1/24/2001

Conference Program

8:30 – 9:00Breakfast
 
9:00 – 10:00Session 8: Performance Modeling and Analysis (Chair: Sally McKee)
 
 "A Delay Model for Pipelined Routers"
 Li-Shiuan Peh and William Dally, Stanford University
 
 "Quantifying the Impact of Architectural Scaling on Communication"
 Taliver Heath, Samian Kaur, Richard Martin, and Thu Nguyen, Rutgers University
 
10:00 – 10:30Coffee Break
 
10:30 – 12:00Session 9: Latency Tolerance Techniques (Chair: Josep Torrellas)
 
 "Call Graph Prefetching for Database Applications"
 Murali Annavaram, Jignesh Patel, and Edward Davidson, University of Michigan
 
 "Branch History Guided Instruction Prefetching"
 Viji Srinivasan, Edward Davidson, Gary Tyson, Mark Charney, and Thomas Puzak, University of Michigan and IBM T.J. Watson Research Center
 
 "Reducing DRAM Latencies with an Integrated Memory Hierarchy Design"
 Wei-fen Lin, Steven Reinhardt, and Doug Burger Universities of Michigan and Texas at Austin
 
12:00 – 12:30Closing Remarks and Awards
 
Questions? Contact heinrich@csl.cornell.edu
Cornell CSL Logo