Paulo Romero Martins Maciel |
|
PhD in Computer Science
|
|
Universidade Federal de Pernambuco (UFPE)
·
Centre of Informatics (CIn)
|
Skills (16)
-
0 Questions39 Followers
-
11 Questions380 Followers
-
0 Questions109 Followers
-
190 Questions17052 Followers
-
12 Questions263 Followers
-
6 Questions42 Followers
-
11 Questions896 Followers
-
58 Questions9381 Followers
-
3 Questions132 Followers
-
190 Questions17052 Followers
-
89 Questions14692 Followers
-
89 Questions14692 Followers
-
6 Questions42 Followers
-
190 Questions17052 Followers
-
190 Questions17052 Followers
-
89 Questions14692 Followers
Research experience
-
Jun 2001–
presentResearch: Universidade Federal de Pernambuco (UFPE)
Universidade Federal de Pernambuco (UFPE) · Centro de Informática (CIn) · MoDCSBrazil · RecifeMy main research interest is devoted to: Dependability and performance evaluation Energy consumption estimation Capacity Planning Cloud computing and data centers
Other
-
LanguagesEnglish
Portuguese -
Scientific MembershipsIEEE
Publications (161) View all
-
Article: Evaluating the power consumption of wireless sensor network applications using models.
[show abstract] [hide abstract]
ABSTRACT: Power consumption is the main concern in developing Wireless Sensor Network (WSN) applications. Consequently, several strategies have been proposed for investigating the power consumption of this kind of application. These strategies can help to predict the WSN lifetime, provide recommendations to application developers and may optimize the energy consumed by the WSN applications. While measurement is a known and precise strategy for power consumption evaluation, it is very costly, tedious and may be unfeasible considering the (usual) large number of WSN nodes. Furthermore, due to the inherent dynamism of WSNs, the instrumentation required by measurement techniques makes difficult their use in several different scenarios. In this context, this paper presents an approach for evaluating the power consumption of WSN applications by using simulation models along with a set of tools to automate the proposed approach. Starting from a programming language code, we automatically generate consumption models used to predict the power consumption of WSN applications. In order to evaluate the proposed approach, we compare the results obtained by using the generated models against ones obtained by measurement.Sensors 01/2013; 13(3):3473-500. · 1.74 Impact Factor -
Article: Performability modeling of electronic funds transfer systems
Carlos Araújo, Paulo Maciel, Armin Zimmermann, Ermeson Andrade, Erica Sousa, Gustavo Callou, Paulo Cunha[show abstract] [hide abstract]
ABSTRACT: Dependability is a paramount requirement for computer systems, since failures can cause a serious economic impact and jeopardize profitability. This paper considers electronic funds transfer (EFT) systems, where computing resources must be used efficiently in order to achieve high availability and reliability for a high quality of the offered service. The design of fault-tolerant computer systems requires their model-based performability evaluation. A stochastic Petri net model of EFT architecture and behavior is presented to evaluate the impact of availability and reliability issues on computational resources. An industrial case study shows the practical usability of the proposed models and techniques. KeywordsEFT system–Performance and reliability analysis–Modeling–Faults Model–Stochastic Petri netsComputing 04/2012; 91(4):315-334. · 0.70 Impact Factor -
Article: Performance and energy consumption estimation for commercial off-the-shelf component system design
[show abstract] [hide abstract]
ABSTRACT: Nowadays, component-based embedded real- time systems have been used to improve the system development as well as to keep cost down through the reuse of embedded software applications. Besides, the use of semi-formal models has been widely adopted in the embedded real-time system component and system life cycle due to their friendly and intuitive notations. However, the ever more complex systems of today require modeling methods that allow early detection of potential problems in the initial phases of development. This paper presents the mapping process of UML state machine diagram into a time Petri net with energy constraints so as to estimate execution time and energy consumption in early phases of the embedded real-time component development life cycle. The estimates obtained from the model show that the proposed approach is indeed a good approximation to the respective measures obtained from the real hardware platform. KeywordsComponent-Embedded systems-Energy consumption-Performance-UMLInnovations in Systems and Software Engineering 04/2012; 6(1):107-114. -
SourceAvailable from: Paulo Romero Martins Maciel
Chapter: Business and Environment Performance Evaluation in Supply Chains: A Formal Model-Driven Approach
09/2011; , ISBN: 978-953-307-250-0 -
SourceAvailable from: Paulo Romero Martins Maciel
Chapter: ISPN: Modeling Stochastic with Input Uncertainties Using an Interval-Based Approach
Se?rgio Galdino, Paulo Maciel09/2011; , ISBN: 978-953-307-708-6