Edson Borin

Hospital de Clínicas da Unicamp

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Otávio O. Napoli

Universidade Estadual de Campinas (UNICAMP)

Vanderson Martins do Rosário

Cadence Design Systems (United States)

Maurício Breternitz

Iscte – Instituto Universitário de Lisboa

Flávia Pisani

Universidade Estadual de Campinas (UNICAMP)

Guido Araújo

Universidade Federal de Mato Grosso do Sul

Rafael Auler

Meta (United States)

Martin Tygel

Universidade Estadual de Campinas (UNICAMP)

Youfeng Wu

National Chung Cheng University

Tiago A. Coimbra

Universidade Estadual de Campinas (UNICAMP)

Diego F. Aranha

Aarhus University

Cooperative Institutions

Universidade Estadual de Campinas (UNICAMP)

110

Hospital de Clínicas da Unicamp

Universidade Federal do Rio Grande do Sul

Universidade Federal de Minas Gerais

Universidade de São Paulo

Intel (United States)

Universidade Estadual Paulista (Unesp)

Universidade Estadual de Maringá

Universidade do Estado do Rio de Janeiro

Institut national de recherche en informatique et en automatique

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Optimizing the Execution Costs of High-Performance Geophysics Software on the Cloud

81st EAGE Conference and Exhibition 2019 (2019)

Nicholas T. Okita Tiago A. Coimbra Charles Rodamilans Martin Tygel Edson Borin

Summary The cloud computing paradigm is enabling the execution of high performance computing software and is a potential alternative to traditional clusters. Not only that, but public providers, such as Microsoft and Amazon, offer a wide variety of virtual machine specifications and prices. Correctly choosing the proper hardware specification for the software can reduce execution times and money. In light of this, we developed an algorithm that automatically selects the virtual machine configurations with the best cost per performance ratio in the Amazon Web Service cloud platform for a running software, using its performance in differently priced configurations as input to the selection. Our results showed that the algorithm is capable of selecting the instances that offers the best cost per performance within minutes, approaching the optimal cost as the dataset size increases.

10.3997/2214-4609.201900770

Cite

Citations (1)

GPU上でのセグメント化ソーティング戦略の評価【Powered by NICT】

IEEE Conference Proceedings (2016)

Schmid Rafael Flávia Pisani Edson Borin Caceres Edson

Source

Cite

Citations (0)

A Database for Reproducible Computational Research

Gabriel F.T. Gomes Edson Borin

The ability to reproduce the experiments of a scientific research is one of the fundamental principles of the scientific method, as failing to reproduce it and obtain results equal (or very similar) to the original might imply that the later were wrong. In compiler and computer architecture research, reproducibility requires a researcher to be able to compile and execute software on (possibly many) computing systems in a repeatable manner. Though the source code for the studied applications, operating systems and compiler tool chains are often available, it is sometimes still impossible to exactly reproduce one experiment due to the lack of information about versioning of the code, build configuration and environmental setup. In this work we propose a database for gathering experimental setup information as well as related experimental results to enforce reproducibility and to enable results sharing among the researchers.

Code (set theory)

10.1109/wscad-ssc.2012.27

Cite

Citations (0)

Dynamic binary control-flow errors detection

ACM SIGARCH Computer Architecture News (2005)

Edson Borin Cheng Wang Youfeng Wu Guido Araújo

Shrinking microprocessor feature size will increase the soft-error rates to unacceptable levels in the near future. While reliable systems typically employ hardware techniques to address soft-errors, software techniques can provide a less expensive and more flexible alternative. This paper presents a control-flow error classification and proposes new software based control-flow error detection techniques. The new techniques are better than the previous ones in the sense that they detect errors in all the branch-error categories. We also compare the performance of our new techniques with that of the previous ones using our dynamic binary translator.

Microprocessor

Control flow

Soft error

Feature (linguistics)

10.1145/1127577.1127582

Cite

Citations (11)

Characterization of DBT overhead

Edson Borin Youfeng Wu

In recent years, dynamic binary translation has emerged as an important tool with many real world applications. Besides supporting legacy binary code and ISA virtualization, it enables innovative co-designed microarchitectures and allows transparent binary instrumentation. The dynamic nature of the translation usually incurs extra execution overhead and many research works had proposed software and hardware solutions to minimize the overhead. In this paper, we analyze our dynamic binary translator performance and depict the main sources of overhead in details. We classify the translation operations and associated overhead into five major categories, and quantify their contribution to the overall overhead. Based on the analysis and detailed evaluation, we identify and point out the most promising solutions to address the overhead problem. We believe this study is an important first step toward the grand goal of zero-overhead dynamic binary translation.

Binary translation

Code (set theory)

10.1109/iiswc.2009.5306785

Cite

Citations (32)

Aceleração de métodos de empilhamento de dados sísmicos na nuvem com CUDA, OpenCL e SPITS

Gustavo Ciotto Pinton Edson Borin

Técnicas de imageamento do subsolo marítimo vêm sendo fundamentais para diversas aplicações da indústria, notadamente para a exploração de petróleo e gás. Neste artigo, nós avaliamos uma implementação para a procura de parâmetros do meio marítimo maximizando a medida de coerência para três diferentes modelos de tempo de trânsito 2D (incluindo o modelo OCT) que automaticamente realiza a distribuição de tarefas idempotentes e independentes a nós de uma nuvem computacional com suporte às plataformas CUDA e OpenCL. Além disso, para evitar degredação de desempenho devido a fenômenos como a transferência de dados e \emph{cache misses}, nós introduzimos uma heurística para a seleção da fração dos dados que deve ser de fato considerada.

10.5753/eradsp.2020.16892

Cite

Citations (0)

Efficiency and scalability of multi-lane capsule networks (MLCN)

Journal of Parallel and Distributed Computing (2021)

Vanderson Martins do Rosário Maurício Breternitz Edson Borin

10.1016/j.jpdc.2021.04.010

Cite

Citations (7)

Addressing JavaScript JIT Engines Performance Quirks: A Crowdsourced Adaptive Compiler

Lecture notes in computer science (2014)

Rafael Auler Edson Borin Peli de Halleux Michał Moskal Nikolai Tillmann

Unobtrusive JavaScript

Just-in-time compilation

Crowdsourcing

10.1007/978-3-642-54807-9_13

Cite

Citations (8)

Quantifying and detecting HPC resource wastage in cloud environments

William F. C. Tavares Marcio Roberto Miranda Assis Edson Borin

Among many details that users need to consider when using cloud computing, the care not to waste resources requires more attention by administrators and new users. When the application does not fully utilize the provisioned resource, the end-of-the-month bill is unnecessarily increased. Several studies have developed solutions to avoid wastage using predictive techniques. Nonetheless, these approaches require applications' to have predictive behavior and depend on pre-executions or history data. To circumvent these limitations, we explore how a reactive solution can be used to detect and contain wastage. More specifically, we discuss several important issues that arise when quantifying resource wastage caused by HPC resource wastage on the cloud and propose a reactive strategy to quantify, detect, and contain resource wastage in this context. This solution is designed so that it can be applied in environments with expert and non-expert users with no prior knowledge about the applications.

Provisioning

Resource Management

10.1109/sbac-padw53941.2021.00017

Cite

Citations (5)

Efficient datapath merging for partially reconfigurable architectures

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (2005)

Nahri Moreano Edson Borin Cid C. de Souza Guido Araújo

Reconfigurable systems have been shown to achieve significant performance speedup through architectures that map the most time-consuming application kernel modules or inner loops to a reconfigurable datapath. As each portion of the application starts to execute, the system partially reconfigures the datapath so as to perform the corresponding computation. The reconfigurable datapath should have as few and simple hardware blocks and interconnections as possible, in order to reduce its cost, area, and reconfiguration overhead. To achieve that, hardware blocks and interconnections should be reused as much as possible across the application. We represent each piece of the application as a data-flow graph (DFG). The DFG merging process identifies similarities among the DFGs, and produces a single datapath that can be dynamically reconfigured and has a minimum area cost, when considering both hardware blocks and interconnections. In this paper we present a novel technique for the DFG merge problem, and we evaluate it using programs from the MediaBench benchmark. Our algorithm execution time approaches the fastest previous solution to this problem and produces datapaths with an average area reduction of 20%. When compared to the best known area solution, our approach produces datapaths with area costs equivalent to (and in many cases better than) it, while achieving impressive speedups.

Datapath

Control reconfiguration

Reconfigurable Computing

Speedup

High-Level Synthesis

Benchmark (surveying)

Data-flow analysis

10.1109/tcad.2005.850844

Cite

Citations (110)