Opening
  1. 8:05-8:15 Opening Remarks
Machine Learning
08:15 - 10:00
Chair: Jakub Kurzak,University of Tennessee, USA

  1. 8:15-9:00 Invited Talk: Just-in-time compilation: Speeding up small linear algebra operations
    Author(s): Sarah Knepper
  2. 9:00-9:30 Use of Code Structural Features for Machine Learning to Predict Effective Optimizations
    Author(s): Yuki Kawarabatake, Mulya Agung, Kazuhiko Komatsu, Ryusuke Egawa and Hiroyuki Takizawa
  3. 9:30-10:00 Effective Machine Learning Based Format Selection and Performance Modeling for SpMV on GPUs
    Author(s): Israt Nisa, Charles Seigel, Aravind Sukumaran Rajam, Abhinav Vishnu and P Sadayappan
Morning Break
10:00 - 10:30

AT Techniques
10:30 - 12:30
Chair: Akihiro Fujii, Kogakuin University, Japan
  1. 10:30-11:00 Tensile: Auto-tuning GEMM GPU Assembly for All Problem Sizes
    Author(s): David Tanner
  2. 11:00-11:30 GreedyTalents: An Energy-Aware Auto-Tuning Method for Many-Core Processor
    Author(s): Tim Platt, Chen Liu and Zhiliu Yang
  3. 11:30-12:00 Auto-tuning for The Era of Relatively High Bandwidth Memory Architectures: A Discussion Based on an FDM Application
    Author(s): Takahiro Katagiri
  4. 12:00-12:30 Threaded Accurate Matrix-Matrix Multiplications with Sparse Matrix-Vector Multiplications
    Author(s): Shuntaro Ichimura, Takahiro Katagiri, Katsuhisa Ozaki, Takeshi Ogita and Toru Nagai
Lunch Break
12:30 - 13:45

Linear Algebra
13:45 - 15:30
Chair: Takahiro Katagiri, Nagoya University, Japan
  1. 13:45-14:30 Invited Talk: High-Performance GEMM on GPUs is Like a 3-Year Old
    Author(s): David Tanner
  2. 14:30-15:00 Algebraic Multigrid Solver using Coarse Grid Aggregation with Independent Aggregation
    Author(s): Naoya Nomura, Akihiro Fujii, Teruo Tanaka, Osni Marques and Kengo Nakajima
  3. 15:00-15:30 A case study on modeling the performance of dense matrix computation: tridiagonalization in the EigenExa eigensolver on the K computer
    Author(s): Takeshi Fukaya, Toshiyuki Imamura and Yusaku Yamamoto
Afternoon Break
15:30 - 16:00

AT Methodologies
16:00 - 17:00
Chair: Hiroyuki Takizawa, Tohoku University, Japan
  1. 16:00-16:30 AutoTuneTMP: Auto-Tuning in C++ With Runtime Template Metaprogramming
    Author(s): David Pfander, Malte Brunn and Dirk Pflüger
  2. 16:30-17:00 Methodology for Adaptive Active Message Coalescing in Task Based Runtime Systems
    Author(s): Bibek Wagle, Samuel Keller, Adrian Serio and Hartmut Kaiser
Closing
  1. 17:20-17:30 Closing talk