2020: Proceedings of the 32nd International Symposium on Computer Architecture and High Performance Computing | Proceedings of the International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)

The Proceedings of the 32nd International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD 2020) bring the papers selected and presented in the edition of the event held from September 8 to 11, 2020, in the city of Porto, Portugal. In this edition, the proceedings include 40 full papers. These papers were selected through a peer review process. The SBAC-PAD 2020 program committee was coordinated by Jorge Barbosa (University of Porto), Lucia Drummond (UFF), and Laurent Lefèvre (INRIA, ENS Lyon, University of Lyon), who organized this volume together with Inês Dutra (University of Porto), Jorge Barbosa (University of Porto), and Miguel Areias (University of Porto), the SBAC-PAD 2020 chairs. The papers included in this volume were submitted on 06/28/2020, accepted for publication on 08/06/2020, with the final version submitted on 08/12/2020.

For more information about SBAC-PAD 2020, visit the website of this edition of the event.

Published: 2020-09-08

Full Papers

Exploiting Non-conventional DVFS on GPUs: Application to Deep Learning

Francisco Mendes, Pedro Tomás, Nuno Roma

1-9
- IEEE Xplore
Design Space Exploration of Accelerators and End-to-End DNN Evaluation with TFLITE-SOC

Nicolas Bohm Agostini, Shi Dong, Elmira Karimi, Marti Torrents Lapuerta, José Cano, José L. Abellán, David Kaeli

10-19
- IEEE Xplore
Hardware Multiversioning for Fail-Operational Multithreaded Applications

Rico Amslinger, Christian Piatka, Florian Haas, Sebastian Weis, Theo Ungerer, Sebastian Altmeyer

20-27
- IEEE Xplore
On-chip Parallel Photonic Reservoir Computing using Multiple Delay Lines

Syed Ali Hasnain, Rabi Mahapatra

28-34
- IEEE Xplore
Online Sharing-Aware Thread Mapping in Software Transactional Memory

Douglas Pereira Pasqualin, Matthias Diener, André Rauber Du Bois, Maurício Lima Pilla

35-42
- IEEE Xplore
Optically Connected Memory for Disaggregated Data Centers

Jorge Gonzalez, Alexander Gazman, Maarten Hattink, Mauricio G. Palma, Meisam Bahadori, Ruth Rubio-Noriega, Lois Orosa, Madeleine Glick, Onur Mutlu, Keren Bergman, Rodolfo Azevedo

43-50
- IEEE Xplore
AIR: A Light-Weight Yet High-Performance Dataflow Engine based on Asynchronous Iterative Routing

Vinu E. Venugopal, Martin Theobald, Samira Chaychi, Amal Tawakuli

51-58
- IEEE Xplore
An Optimal Model for Optimizing the Placement and Parallelism of Data Stream Processing Applications on Cloud-Edge Computing

Felipe Rodrigo de Souza, Marcos Dias de Assunçao, Eddy Caron, Alexandre da Silva Veith

59-66
- IEEE Xplore
Evaluating Computation and Data Placements in Edge Infrastructures through a Common Simulator

Anderson Andrei Da Silva, Clément Mommessin, Pierre Neyron, Denis Trystram, Adwait Bauskar, Adrien Lebre, Alexandre Van Kempen, Yanik Ngoko, Yoann Ricordel

67-74
- IEEE Xplore
Optimizing Green Energy Consumption of Fog Computing Architectures

Adrien Gougeon, Benjamin Camus, Anne-Cécile Orgerie

75-82
- IEEE Xplore
Energy-Efficient Time Series Analysis Using Transprecision Computing

Ivan Fernandez, Ricardo Quislant, Eladio Gutierrez, Oscar Plata

83-90
- IEEE Xplore
High Performance and Portable Convolution Operators for Multicore Processors

Pablo San Juan, Adrián Castelló, Manuel F. Dolz, Pedro Alonso-Jordá, Enrique S. Quintana-Ortí

91-98
- IEEE Xplore
High-Performance Low-Memory Lowering: GEMM-based Algorithms for DNN Convolution

Andrew Anderson, Aravind Vasudevan, Cormac Keane, David Gregg

99-106
- IEEE Xplore
Optimized Transactional Data Structure Approach to Concurrency Control for In-Memory Databases

Christina Peterson, Amalee Wilson, Peter Pirkelbauer, Damian Dechev

107-115
- IEEE Xplore
Reliable and Energy-aware Mapping of Streaming Series-parallel Applications onto Hierarchical Platforms

Changjiang Gou, Anne Benoit, Mingsong Chen, Loris Marchal, Tongquan Wei

116-123
- IEEE Xplore
Scalable and Efficient Spatial-Aware Parallelization Strategies for Multimedia Retrieval

Guilherme Andrade, George Teodoro, Renato Ferreira

124-131
- IEEE Xplore
Scheduling Methods to Reduce Response Latency of Function as a Service

Pawel Zuk, Krzysztof Rzadca

132-140
- IEEE Xplore
Selective Protection for Sparse Iterative Solvers to Reduce the Resilience Overhead

Hongyang Sun, Ana Gainaru, Manu Shantharam, Padma Raghavan

141-148
- IEEE Xplore
sputniPIC: An Implicit Particle-in-Cell Code for Multi-GPU Systems

Steven W. D. Chien, Jonas Nylund, Gabriel Bengtsson, Ivy B. Peng, Artur Podobas, Stefano Markidis

149-156
- IEEE Xplore
Using Skip Graphs for Increased NUMA Locality

Samuel Thomas, Roxana Hayne, Jonad Pulaj, Hammurabi Mendes

157-166
- IEEE Xplore
A Fast and Concise Parallel Implementation of the 8x8 2D IDCT using Halide

Martin Johnson, Daniel Playne

167-174
- IEEE Xplore
Controlling Garbage Collection and Request Admission to Improve Performance of FaaS Applications

David Quaresma, Daniel Fireman, Thiago Emmanuel Pereira

175-182
- IEEE Xplore
On the Memory Underutilization: Exploring Disaggregated Memory on HPC Systems

Ivy Peng, Roger Pearce, Maya Gokhale

183-190
- IEEE Xplore
Predicting the Energy Consumption of CUDA Kernels using SimGrid

Dorra Boughzala, Laurent Lefèvre, Anne-Cécile Orgerie

191-198
- IEEE Xplore
TASO: Time and Space Optimization for Memory-Constrained DNN Inference

Yuan Wen, Andrew Anderson, Valentin Radu, Michael F.P. O'Boyle, David Gregg

199-208
- IEEE Xplore
XPySom: High-Performance Self-Organizing Maps

Riccardo Mancini, Antonio Ritacco, Giacomo Lanciano, Tommaso Cucinotta

209-216
- IEEE Xplore
A Robotic Communication Middleware Combining High Performance and High Reliability

Wei Liu, Hao Wu, Ziyue Jiang, Yifan Gong, Jiangming Jin

217-224
- IEEE Xplore
MASA-StarPU: Parallel Sequence Comparison with Multiple Scheduling Policies and Pruning

Rafael A. Lopes, Samuel Thibault, Alba C. M. A. Melo

225-232
- IEEE Xplore
PSU: A Framework for Dynamic Software Updates in Multi-threaded C-Language Programs

Marcus Karpoff, Jose Nelson Amaral, Kai-Ting Amy Wang, Rayson Ho, Brice Dobry

233-240
- IEEE Xplore
Towards Communication Profile, Topology and Node Failure Aware Process Placement

Ioannis Vardas, Manolis Ploumidis, Manolis Marazakis

241-248
- IEEE Xplore
OmpTracing: Easy Profiling of OpenMP Programs

Vitoria Pinho, Hervé Yviquel, Marcio Machado Pereira, Guido Araujo

249-256
- IEEE Xplore
Analyzing the Loop Scheduling Mechanisms on Julia Multithreading

Diana A. Barros, Cristiana Bentes

257-264
- IEEE Xplore
Performance Analysis and Optimization of the Vector-Kronecker Product Multiplication

Alexandre Azevedo, Cristiana Bentes, Maria Clicia Castro, Claude Tadonki

265-272
- IEEE Xplore
JAMPI: A C++ Parallel Programming Interface Allowing the Implementation of Custom and Generic Scheduling Mechanisms

Daniel Di Domenico, Gerson G. H. Cavalheiro

273-280
- IEEE Xplore
Towards Pervasive Containerization of HPC Job Schedulers

Christophe Cérin, Nicolas Greneche, Tarek Menouer

281-288
- IEEE Xplore
Towards Profile-Guided Optimization for Safe and Efficient Parallel Stream Processing in Rust

Stefan Sydow, Mohannad Nabelsee, Sabine Glesner, Paula Herber

289-296
- IEEE Xplore
Re-evaluation of Atomic Operations and Graph Coloring for Unstructured Finite Volume GPU Simulations

Xi Zhang, Xu Sun, Xiaohu Guo, Yunfei Du, Yutong Lu, Yang Liu

297-304
- IEEE Xplore
Extending Heterogeneous Applications to Remote Co-processors with rOpenCL

Rui Alves, José Rufino

305-312
- IEEE Xplore
FFT Optimizations and Performance Assessment Targeted towards Satellite and Airborne Radar Processing

Maron Schlemon, Jamin Naghmouchi

313-320
- IEEE Xplore
A Highly Efficient SGEMM Implementation using DMA on the Intel/Movidius Myriad-2

Suyash Bakshi, Lennart Johnsson

321-328
- IEEE Xplore