Program
Proceedings are available online as part of Springer LNCS book series.
Tuesday (10/22/2019)
Venue: TSRB Room 118
LIGHT BREAKFAST, 8 am – 9 am
Opening : Welcome and overview of LCPC 2019: 8:45 am to 9:00 am
Session 1: Cryptographic Computations need Compilers, Keynote by Madan Musuvathi, Microsoft Research (Session Chair: Vivek Sarkar), 9:00 am – 10am, Slides
BREAK, 10am – 10:30am
Session 2 (Session Chair: Chen Ding): 10:30 am to 12 noon
- HPVM: Performance, Portability, Programmability and Approximation for Heterogeneous Parallel Systems, Invited Talk by Vikram Adve, UIUC
- Is a Post Moore’s Law Compiler Renaissance needed? Panel moderated by Rudi Eigenmann, University Of Delaware.
Tentative Panel: Albert Cohen (Google), Mary Hall (University of Utah), P. (Saday) Sadayappan (University of Utah), David Padua (University of Illinois at Urbana‐Champaign)
LUNCH, 12noon – 1:30pm at Georgia Tech Hotel Dining
Session 3: Compiler Optimizations for Tensors/CNNs (Session Chair : P. (Saday) Sadayappan): 1:30pm – 3pm
- Optimization in the Sparse Tensor Algebra Compiler, Invited Talk by Fredrik Kjolstad, MIT
- SWIRL++: Evaluating Performance Models to Guide Code Transformation in Convolutional Neural Networks, Tharindu Rusira (University of Utah), Anand Venkat (Intel Labs), Raj Barik (Uber Technologies Inc.), and Mary Hall (University of Utah)
- CubeGen: Code Generation for Accelerated GEMM-based Convolution with Tiling, Amarin Phaosawasdi (University of Illinois at Urbana-Champaign), Christopher Rodrigues (Futurewei Technologies), Long Chen (Futurewei Technologies), and Peng Wu (Futurewei Technologies)
BREAK, 3pm – 3:30pm
Session 4: Data movement/Caches/Consistency (Session Chair : Lawrence Rauchwerger): 3:30pm – 5pm
- Optimizing Data Movement and Achieving Performance Portability with Brick Data Layouts, Invited Talk by Mary Hall, University of Utah
- CLAM: Compiler Leasing of Accelerator Memory, Chen Ding (University of Rochester), Dong Chen (National University of Defense Technology, China), and Dorin Patru (Rochester Institute of Technology)
- Consistency types for distributed programming languages, Xin Zhao (KTH Royal Institute of Technology, Sweden), and Philipp Haller (KTH Royal Institute of Technology, Sweden)
Reception from 5:45 pm to 7:15 pm at Georgia Tech Hotel, 2nd Floor, Salon B Foyer
Dinner from 7:15 pm to 9:00 pm at Georgia Tech Hotel, 2nd Floor, Salon B
Wednesday (10/23/2019)
Venue: TSRB Room 118
LIGHT BREAKFAST, 8 am – 9 am
Session 5: Abstractions/DSLs (Session Chair: Albert Cohen): 9:00 am to 10:30 am
- DPC++: A Direct Programming Model for Diverse Architectures, Invited talk by Xinmin Tian, Intel
- Abstractions for Polyhedral Topology-Aware Tasking, Martin Kong (University of Oklahoma)
- A Structured Grid Solver with Polyhedral+Dataflow Representation, Eddie C. Davis (Boise State University), Catherine RM. Olschanowsky (Boise State University), and Brian Van Straalen (Lawrence Berkeley National Laboratory)
BREAK, 10:30 am – 11 am
Session 6: Invited Talks (Session Chair: Fredrik Kjolstad): 11 am to 12 pm
- Model-driven multi-level tiling for tensor contractions, Invited talk by P. (Saday) Sadayappan, University of Utah
- A framework for compositional transformations of recursions and loops, Milind Kulkarni, Purdue
LUNCH, 12 pm to 1:30 pm at Georgia Tech Hotel Dining
Session 7: Resource-Efficient Quantum Computing by Breaking Abstractions, Keynote by Fred Chong, University of Chicago (Session Chair: David Padua), 1:30 pm to 2:30 pm
Georgia Aquarium Excursion, depart at 3:15 pm from Georgia Tech Hotel Lobby
Thursday (10/24/2019)
Venue: GLC Room 233
LIGHT BREAKFAST, 8 am – 9 am
Session 8: Invited talks (Session Chair: Rudi Eigenmann): 9:00 am to 10:00 am
- A Lightweight Polyhedral Abstraction for MLIR, Invited talk by Albert Cohen, Google
- Filling in the gaps between applications and OpenMP for in-node programming on exascale systems, Invited talk by Oscar Hernandez, Oak Ridge National Laboratory
BREAK, 10:00 am – 10:30 am
Session 9: Vectorization and GPUs (Session Chair: Martin Kong): 10:30am – 12noon
- Common Subexpression Convergence: A New Code Optimization for SIMT processors, Sana Damani (Georgia Tech), and Vivek Sarkar (Georgia Tech)
- PostSLP: Cross-Region Vectorization of Fully or Partially Vectorized Code, Vasileios Porpodas (Intel Corporation), and Pushkar Ratnalikar (Intel Corporation)
- FLARE: Flexibly Sharing Commodity GPUs to Enforce QoS and Improve Utilization, Wei Han (Colorado School of Mines), Daniel Mawhirter (Colorado School of Mines), Bo Wu (Colorado School of Mines), Lin Ma (Huawei US R&D Center), and Chen Tian (Huawei US R&D Center)
LUNCH, 12:00 pm – 1:15 pm at GLC Dining
Session 10: Feedback/JIT Optimization (Session Chair: Michael Phillipsen): 1:15 pm to 2:15 pm
- Using performance event profiles to deduce an execution model of MATLAB with Just-In-Time compilation, Patryk Kiepas (MINES ParisTech / PSL University), Corinne Ancourt (MINES ParisTech / PSL University), Claude Tadonki (MINES ParisTech / PSL University), and Jarosław Koźlak(AGH University of Science and Technology, Kraków)
- Performance of Static and Dynamic Task Scheduling for Real-Time Control System on Embedded Multicore Processor, Yoshitake Oki (Waseda University), Hiroki Mikami (Waseda University), Hikaru Nishida (Waseda University), Dan Umeda (Waseda University), Keiji Kimura (Waseda University), and Hironori Kasahara (Waseda University)
LCPC 2019 Concludes : 2:15 pm
Informal meet-ups/networking: 2:15 pm to 3:00 pm (Snacks and drinks served)
Papers
This is a password-protected zip file. Please download the file, add .zip extension to it, and check with workshop organizers onsite for the password.