Activities

2019

CPDLA 2019 – Computação Paralela, Distribuída e De Larga Escala
Sep. 2019
Universidade do Minho, Braga

There is an increasing interest in the professional and scientific communities towards the adequate employment of computing infrastructures. This requires the distribution of computational tasks across available resources. This session at the Portuguese INFORUM conference aims the discussion of all subjects concerning to concurrent, parallel, distributed and large-scale systems. We seek contributions, among others, on issues relating to execution environments for distributed and parallel programming, Grid and Clouds, distributed algorithms and architectures, and high-performance computing and e-science.

http://inforum.org.pt/INForum2019/sessoes/computacao-paralela-distribuida-e-de-larga-escala.html


OPTIM 2019 – International Workshop on
Optimization of Energy Efficient HPC & Distributed Systems

July. 2019
Dublin, Ireland

Energy efficiency optimization has gradually become one fundamental constraint and requisite in the design of High Performance Computing (HPC) and Distributed Systems, comprehending several prominent research domains embracing the processor architectures (homogeneous/heterogeneous many-core GPPs), the processing accelerators and co-processors (e.g., APUs, GPUs, FPGAs, etc.), the distributed infrastructures (clusters, grids, data centers, etc.), the distributed platforms (application models, framework runtimes and virtual machines), and the underlying network and communication subsystem. Moreover, even in the embedded domain, saving energy in battery-supplied devices is established as a fundamental issue in mobile, hand-held and wireless applications and in most pervasive systems.

Accordingly, due to the often imposed thermal, power and energy constraints, an effective minimization and optimization of the energy consumption surpasses the obvious economic impact of the energy cost, since it also directly affects other aspects of the overall infrastructure with a significant influence on its viability, including the power supply and its distribution, cooling mechanisms (cost and space overhead) and management techniques, etc.

These new challenges require innovative and effective solutions to attain an effective optimization and/or minimization of the power consumption in HPC and distributed systems, including energy-efficient architectures, scheduling and mapping algorithms, load-balancing and scalability studies, and communication protocols.

http://hpcs2019.cisedu.info/2-conference/workshops/workshop02-optim


Talks on MDE & AI
Apr. 2019
Quidgest, Lisboa

Model Driven Engineering – MDE is a development approach where models are set in the center of every engineering process. Its target is to guarantee significant rises in productivity, maintenance and interoperability. It’s increasingly more present in industry, not only in sectors as cyber-physical, as automotive and aircraft industry, but also in the IT sector, where low-code platforms are gaining popularity.

In the last decade, we have lived a paradigm change due to important advances in methodologies, processes and technologies, which guarantee the development of languages to specific domain and modelling technologies. Making language engineering easy.

In this presentation we will talk about the adoption of MDE and analyses the pros and cons, current obstacles; will make suggestions for a more effective adoption and challenge the academic community for future research.

https://quidgest.com/en/events/talks-on-model-driven-engineering-artificial-intelligence-approaches/


2018

QUATIC 2018
4-7 Sep. 2018
Faculdade de Ciências e Tecnologia, Universidade de Coimbra

The event served as a a forum for disseminating advanced methods, techniques and tools for supporting quality approaches to ICT engineering and management. Practitioners and researchers had the opportunity to exchange ideas and approaches on how to adopt a quality culture in ICT process and product improvement and to provide practical studies in varying contexts.

https://sites.google.com/view/quatic2018/home


CPDLA 2018 – Computação Paralela, Distribuída e De Larga Escala
Sep. 2018
Universidade de Coimbra

There is an increasing interest in the professional and scientific communities towards the adequate employment of computing infrastructures. This requires the distribution of computational tasks across available resources. This session at the Portuguese INFORUM conference aims the discussion of all subjects concerning to concurrent, parallel, distributed and large-scale systems. We seek contributions, among others, on issues relating to execution environments for distributed and parallel programming, Grid and Clouds, distributed algorithms and architectures, and high-performance computing and e-science.

http://inforum.org.pt/INForum2018/sessoes/computacao-paralela-distribuida-e-de-larga-escala.html


Low-code-Software-Development-Summer-School 2018
16-20 Jul
ISCTE-IUL, Lisboa

The goal of this Summer School is to provide training on different aspects of low-code software development, using the OutSystems low-code software development platform, as well as to foster research topics around the low-code software development paradigma. These topics include model-driven development, cloud-based software as a service, programming language engineering, software quality and software security.

Low-code development platforms allow the creation of application software through graphical user interfaces and configuration instead of traditional procedural computer programming. These platforms focus on design and development of databases, business processes, or user interfaces for the development of complex software such as business web or mobile applications.

http://lowcode2018.allofads.com/


2017

RE’17
Sep. 2017
Reitoria of Universidade Nova de Lisboa

The 25th IEEE International Requirements Engineering Conference (RE’17) offered an extensive program of interest to academia, government and industry. It included several distinguished keynote speakers and three conference days full of papers, panels, posters and demos. A series of exciting tutorials to develop skills in and advance awareness of requirements engineering practices is of particular interest to industry. Two days of workshops as well as doctoral symposium offer forums for participants to present cutting-edge techniques and approaches in particular fields.


CPDLA 2017 – Computação Paralela, Distribuída e De Larga Escala
Sep. 2017
Universidade de Coimbra

There is an increasing interest in the professional and scientific communities towards the adequate employment of computing infrastructures. This requires the distribution of computational tasks across available resources. This session at the Portuguese INFORUM conference aims the discussion of all subjects concerning to concurrent, parallel, distributed and large-scale systems. We seek contributions, among others, on issues relating to execution environments for distributed and parallel programming, Grid and Clouds, distributed algorithms and architectures, and high-performance computing and e-science.

Relevant topics include: Distributed systems and middleware, Distributed algorithms, replication, high availability and fault-tolerance, High-performance and scalable computing, Grid, utility and cloud computing, Large-scale and Big-Data application scenarios.


IoTGC
July 2017
Universidade da Madeira (UMa)

Internet of Things for the Global Community – IoTGC conference addressing: Advanced manufacturing and logistics with IoT, Application Specific IoT (ASIT), Artificial Intelligence in IoT, Big Data in IoT, Cloud Computing and IoT, IoT Applications, IoT Enabling Technologies, IoT Experimental Results and Deployment Scenarios, Internet of Things for Smart Cities,IoT Fundamentals, IoT Interactions, IoT in renewable energy, IoT Multimedia, IoT Societal Impacts, IoT Regulations, IoT Services, Security and Privacy for Internet of Things, Sensor and Sensing Technologies in IoT


2016

QUATIC 2016
Sep. 2016
Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa

The event served as a a forum for disseminating advanced methods, techniques and tools for supporting quality approaches to ICT engineering and management. Practitioners and researchers had the opportunity to exchange ideas and approaches on how to adopt a quality culture in ICT process and product improvement and to provide practical studies in varying contexts.


CPDLA 2016 – Computação Paralela, Distribuída e De Larga Escala
Sep. 2016
Instituto Superior Técnico, UL, Lisboa

There is an increasing interest in the professional and scientific communities towards the adequate employment of computing infrastructures. This requires the distribution of computational tasks across available resources. This session at the Portuguese INFORUM conference aims the discussion of all subjects concerning to concurrent, parallel, distributed and large-scale systems. We seek contributions, among others, on issues relating to execution environments for distributed and parallel programming, Grid and Clouds, distributed algorithms and architectures, and high-performance computing and e-science.

Relevant topics include: Distributed systems and middleware, Distributed algorithms, replication, high availability and fault-tolerance, High-performance and scalable computing, Grid, utility and cloud computing, Large-scale and Big-Data application scenarios.


Summer School on Distributed Computing
Summer school in Systems and Networking

Jun. 2016
Costa da Caparica

The co-located events address the broad areas of distributed computing, systems and networking. It consisted of two days of lectures by renowned researchers on selected topics. Students also had the opportunity to present a poster about their research and ample time to engage in discussion with all the participants.


2015

CPDLA 2015 – Computação Paralela, Distribuída e De Larga Escala
Sep. 2015
Universidade da Beira Interior, Covilhã

There is an increasing interest in the professional and scientific communities towards the adequate employment of computing infrastructures. This requires the distribution of computational tasks across available resources. This session at the Portuguese INFORUM conference aims the discussion of all subjects concerning to concurrent, parallel, distributed and large-scale systems. We seek contributions, among others, on issues relating to execution environments for distributed and parallel programming, Grid and Clouds, distributed algorithms and architectures, and high-performance computing and e-science.

Relevant topics include: Distributed systems and middleware, Distributed algorithms, replication, high availability and fault-tolerance, High-performance and scalable computing, Grid, utility and cloud computing, Large-scale and Big-Data application scenarios.


IT4ENERGY 2015
Apr. 2015
Fundação Gulbenkian, Lisboa

The event brought together experts and academics with different backgrounds, covering the Energy, Engineering and Computer Science, as well as industry experts operating in the energy, IT, automation and facility management. It aimed to establish itself as a forum for discussion on the topic of effective information technology for energy applications. It was focused on Policies and Best Practices for Energy Efficiency, counting with reputable key speakers on the field, as well as Information Technology for Energy Applications, with topics covering SOA in IEMS, Innovative Business Models and User Behaviour Transformation amongst others.

http://www.alfamicro.pt/it4energy-back-to-back-with-smart-campus-conference/


International Conference on Cloud Computing and Services Science (CLOSER 2015)
May 2015
Lisboa

Workshop on Emerging Software as a Service and Analytics – ESaaSA 2015

Keynote: Economics-inspired Resource and Energy Management for Cloud Environments.

Cloud infrastructures host different kinds of applications belonging to clients with different levels of service agreements. Their execution is supported by high language virtual machines and system-level virtual machines (VMs). Aiming to maximize revenue, by minimizing the operational expenditure, cloud providers often consolidate several VMs in a single server. This is particularly useful also for the emergent distributed clouds where physical resources, at each node, are not so abundant. However, this technique can lead to overcommitment of resources, and to undesirable performance degradation, if carried out in a non-informed way.

Recent work presented proposes new allocation mechanisms for the VMs used by Platform-as-a-Service (PaaS) and Infrastructure-as-a-Service (IaaS) clouds, and how these mechanisms are to be controlled. They are driven by classic economic notions such as yield, which expresses the return the provider has in applying a given resource allocation to the tenants’ workload, and utility functions, a relation of clients’ perceived usefulness to a given allocation.

For PaaS providers, a Java VM was extended with an integrated resource management API, heap resizing policies for yield maximization and concurrent checkpoint for migration of the execution state. Overall, these new mechanisms impose small penalties, measured in the execution of typical benchmarks, while allowing the use of application-tailored policies.

At the IaaS level, we present a novel cost model and new scheduling algorithms for system-level VMs, along with their implementation in a state-of-the-art cloud simulation framework. Simulations with synthetic and real-world traces, show that the utility-based scheduling allows more VMs to be allocated, thus allowing extra revenue per resource allocated, and shorter waiting times for clients, when comparing with a utility-oblivious redistribution of resources.

Scheduling algorithms are also extended to take into account energy efficiency. For this assessment the typical data center topologies are compared against community networks and community cloud scenarios, which are backed by peer-to-peer architectures.

http://closer.scitevents.org/ESaaSA.aspx?y=2015


1st Workshop Multi-Paradigm Modelling for Cyber-Physical Systems (MPM4CPS)
May 2015
Ordem dos Engenheiros, Lisboa

O Conselho Regional Sul de Engenharia Informática promoveu uma Conferência com o tema: Multi-Paradigm Modelling for Cyber-Physical Systems (MPM4CPS). A Ordem dos Engenheiros será a anfitriã deste evento que contará com 50 peritos de mais de 18 países Europeus, que se irão reunir para debater a temática da Modelação Multi-Paradigma para a construção de Sistemas Ciberfísicos, no âmbito de uma rede de Excelência COST ( IC1404 ), com o apoio da Comunidade Europeia. Este evento irá promover grupos de ação que trabalharão nos próximos 4 anos ao nível Europeu para construir bases sólidas, ferramentas, estratégias de educação, definição de perfis de peritos em CPS, e integração industrial de ponta, de modo a produzir “Saber Fazer Europeu”, numa área tida como unificadora e prioritária para investimento Europeu. A emergência de sistemas complexos com uma forte componente de software e de cariz multidisciplinar, requer cada vez mais uma abordagem sistemática, com metodologias e ferramentas para acompanhar a sua engenharia. Não é conhecida até à data, uma teoria unificadora que agregue os conhecimentos requeridos para a construção destes sistemas nas áreas da Mecânica, Eletrónica e Informática.

http://www.ordemengenheiros.pt/pt/agenda/conferencia-multi-paradigm-modelling-for-cyber-physical-systems-mpm4cps/


2nd DataStorm Big Data Summer School
July 2015
Instituto Superior Técnico, UL, Lisboa

Keynote: Big Data Infrastructures: Economics of Resources, Energy, Data, and Applications

Big data processing infrastructures encompass the full spectrum of cloud-based services: ranging from Infrastructure-as-a-Service (IaaS), to Platform-as-a-Service (Paas) and Data-as-a-Service (DaaS) both for storage as well as processing. Our recent work proposes new resource management and task scheduling mechanisms and algorithms, covering IaaS, PaaS, and DaaS. These are driven by economic inspired notions (such as yield, partial-utility and quality-of-data) that express the efficiency and effectiveness of applying resources to the tenants’ workload, taking into account the clients’ and workloads’ perceived usefulness to a given allocation.

At the IaaS level, system virtual machines (VMs), the cloud providers goals are to maximize revenue, by minimizing the operational expenditure, typically through consolidation of several VMs per physical server. We present a novel cost model and new scheduling algorithms for system-level VMs, along with their implementation in a state-of-the-art cloud simulation framework. Simulations with synthetic and real-world traces, show that the utility-based scheduling allows more VMs to be allocated, thus allowing extra revenue per resource allocated, and shorter waiting times for clients, when comparing with a utility-oblivious redistribution of resources. Scheduling algorithms are also extended to take into account energy efficiency. For this assessment the typical data center topologies are compared against community networks and community cloud scenarios, which are backed by peer-to-peer architectures.

Regarding the PaaS level, we propose new resource management and application monitoring mechanisms for the VMs used by Platform-as-a-Service (PaaS). For PaaS providers, a Java VM was extended with an integrated resource management API, heap resizing policies for yield maximization and concurrent checkpoint for migration of the execution state. Overall, these new mechanisms impose small penalties, measured in the execution of typical benchmarks, while allowing the use of application-tailored policies.

Regarding the DaaS level, we address the storage and processing layers, so that the underlying infrastructure and resource management should fit data and application behavior effectively and efficiently. In Big-data noSQL storage, such as HBase, the main requirement is high-availability, but classical consistency models necessarily degrade performance. Still, an important class of applications can tolerate and benefit from relaxed consistency by bounding inconsistent access in application-specific manner. We explore the spectrum between pessimistic and eventual consistency and provide a novel consistency model with Quality-of-Service tailored to geo-replication (Quality-of-Service for Consistency of Data).

In Big-data processing, specifically in Java frameworks such as Hadoop / Map-Reduce and workflows/dataflows of MR jobs, traditional WMS enforce strict temporal synchronization. In continuous processing, resources are wasted due to the small impact that input data often might have on the ?nal output. thus, no data- and application-aware reasoning is performed in order to evaluate impact of new executions on the ?nal data?ow output. This fails at delivering high resource efficiency for long-lasting applications while keeping costs low. We propose a novel dataflow model with Quality-of-Data (QoD) for continuous and incremental processing. The WMS framework enforcing the QoD model trades off results accuracy with resource savings, achieving resource efficiency, controlled performance, and task prioritization, targeting data-intensive data?ows, where processing steps communicate data via a noSQL database.

http://www.ds3.inesc-id.pt/#schedule


IEEExtreme 9.0
Oct. 2015
Faculdade de Ciências e Tecnologia, Universidade de Coimbra

IEEEXtreme is a global challenge in which teams of IEEE Student members—advised and proctored by an IEEE member, and often supported by an IEEE Student Branch—compete in a 24-hour time span against each other to solve a set of programming problems. Th 9th edition of IEEEXtreme in Portugal was held in Coimbra.

http://www.ieee-uc.org/ieeextreme/2015.html


Navtalks
Nov. 2015
Faculdade de Ciências, UL, Lisboa

Invited Talk: Cassowary: Middleware Platform for Context-Aware Smart Buildings with Software-Defined Sensor Networks

Smart devices sense the environment through their sensors and leverage the contextual information derived from the sensor readings to satisfy system requirements such as energy and carbon efficiency and user preferences. Smart buildings compose of smart devices, and local sensors of the devices and device controllers in a coordinated network. Software-Defined Networking (SDN) offers a centralized view of the entire networking data plane elements to a logically centralized controller. While smart buildings and ubiquitous computing are heavily researched, later advancements in networking are not exploited in achieving tenant-aware smart buildings.

http://www.navigators.di.fc.ul.pt/wiki/Navtalks


The Third OpenDaylight Lisboa Meetup
Nov. 2015
Faculdade de Ciências, UL, Lisboa

This meet served to discuss current research in Software-Defined Networking and recent OpenDaylight implementation developments.


First Cassandra Users Lisbon Meet-up
June 2015
Instituto Superior Técnico, UL, Lisboa

Fighting Fraud with Cassandra

Payment Fraud and data breaches are on a rise. Every year millions of credit-card numbers are compromised. In the recent Target breach between 70 and 110 million people saw their information stolen and purchases being made in their name. In this talk we will discuss how at Feedzai we leverage Machine Learning and Big-Data Techniques to make commerce safe. In particular, we’ll share our experience on leveraging Cassandra as a fast real-time NoSQL database, enabling us to score thousands of transactions per second with extremely short latencies.

Speaker: Diogo Ferreira, Feedzai.

Bio: Diogo Ferreira is a Product Architect and Team Leader at Feedzai. Diogo’s ninja skills include high performance computing, designing systems that scale, and the Java/Scala platform. Diogo is also known by his contributions to open-source world in the CyanogenMod and Compiz projects.


2014

QUATIC 2014 – The International Conference on the Quality of Information and Communications Technology
Sep. 2014
Guimarães

Keynote: Conceptual models versus ontologies

The overwhelming use of Internet emphasizes the need for efficient techniques aiming at facilitating the understanding of the information that is found over the Web. This need for understanding is mostly addressed nowadays through the development of ontologies. Still, rather than simply relying on most traditional taxonomic ontologies (e.g. Wordnet), advanced approaches tend to develop to more descriptive ontologies. These are intended to provide agreed descriptions of the data structures representing the complex organization of objects and links of interest within the targeted domain. We are very close here to the goal usually associated with conceptual modeling when in a database environment.
This talk analyzes the requirements for descriptive ontologies, and contrast the requirements to the functionality provided by some current representative approaches that have been proposed for ontology management. Selected approaches originate from research in artificial intelligence, knowledge representation, and database conceptual modeling.

Speaker: Stefano Spaccapietra, EPFL, Switzerland. Stefano Spaccapietra is a honorary professor at EPFL, Switzerland, where he has been heading the database laboratory for over 20 year. He has been in academic positions all along his career. He got his PhD from the University of Paris VI, in 1978, where he first had his master in Computer Science in 1969. At that time, he has been teaching file systems, later turned into teaching database systems. He moved to the University of Burgundy, Dijon, in 1983 to take a professor position at the Institute of Technology. He left Dijon for EPFL in 1988. Prof. Spaccapietra is a Fellow of the IEEE and recipient of the IFIP Silver Core Award and ER Award. He is an ER fellow. He has been Editor-in-chief of the Journal of Data Semantics (LNCS subline), Springer. He is member of the editorial boards of the Data and Knowledge Engineering Journal (Elsevier), the Internet and Web Information Systems Journal (Kluwer), the Revue Internationale de Géomatique (Hermes), and the Computing Letters Journal (CoLe), VSP/Brill. He was former Chair of the IFIP Working Group 2.6 “Databases” and of the ER Conferences Steering Committee. He also chaired many conference program committees. He co-authored several books, a large number of journal papers, and a very large number of conference papers.

Organized in the context of the IEEE Computer Society Distinguished Visitors Program (DVP) Keynote Speakers.

http://2014.quatic.org/keynotes


Conferência sobre Risco e grandes sistemas informáticos da Administração Pública – 1ª Sessão
Dec. 2014
Auditório da Sede da Região Sul, Ordem dos Engenheiros, Lisboa

O Conselho Nacional de Colégio de Engenharia Informática da Ordem dos Engenheiros em parceria com os Conselhos Regionais de Colégios de Engenharia Informática da Região Centro e da Região Sul da Ordem dos Engenheiros, está a organizar um Ciclo de Conferências sob o título “Ordem na Informática”, que pretende ser um fórum de discussão franca e aberta para debater publicamente os principais problemas com os quais a Engenharia Informática se tem deparado na atualidade. Este Ciclo será composto por seis conferências que decorrerão entre as três Secções Regionais da Ordem dos Engenheiros, sendo a primeira na Região Sul, com o título “Risco e grandes sistemas informáticos da Administração Pública”, já no próximo dia 9 de dezembro,pelas 17h00, no auditório da Sede da Região Sul em Lisboa.

http://www.ordemengenheiros.pt/pt/agenda/conferencia-sobre-risco-e-grandes-sistemas-informaticos-da-administracao-publica-1a-sessao/


CPDLA 2014 – Computação Paralela, Distribuída e De Larga Escala
Sep. 2014
Universidade do Porto

There is an increasing interest in the professional and scientific communities towards the adequate employment of computing infrastructures. This requires the distribution of computational tasks across available resources. This session at the Portuguese INFORUM conference aims the discussion of all subjects concerning to concurrent, parallel, distributed and large-scale systems. We seek contributions, among others, on issues relating to execution environments for distributed and parallel programming, Grid and Clouds, distributed algorithms and architectures, and high-performance computing and e-science.

Relevant topics include: Distributed systems and middleware, Distributed algorithms, replication, high availability and fault-tolerance, High-performance and scalable computing, Grid, utility and cloud computing, Large-scale and Big-Data application scenarios.


Conferência sobre Risco e grandes sistemas informáticos da Administração Pública – 1ª Sessão
Dec. 2014
Auditório da Sede da Região Sul, Ordem dos Engenheiros, Lisboa

The IEEE Computer Society is the world’s leading computing membership organization and the trusted information and career-development source for a global workforce of technology leaders including: professors, researchers, software engineers, IT professionals, employers, and students –our future technology leaders. Members have endless opportunities to network with the world’s foremost technology professionals, to lead the community by volunteering and mentoring, and to both gain exposure and contribute to the field as authors and reviewers.