You are here

Seminars

Title

Click on the title to read the abstract

Presenter

Date

Presentation

Enabling Scalable Cloud Service Choreographies Marco Aurélio Gerosa Carlos Eduardo Moreira dos Santos 15-Nov-2011 PDF

Detecting SLA Violations in Web Service Choreographies

Victoriano Alfonso Phocco Diaz

26-Ago-2011

PDF (in Portuguese)

Modeling with I*

Yanik Ngoko

11-Abr-2011

PDF

An Overview of NEXOF

Victor Williams Stafusa da Silva

4-Abr-2011

PDF

From Volunteer Computing to Web Service Choreographies in the Cloud

Fabio Kon

14-Mar-2011

PDF

CHOReOS: Large Scale Choreographies for the Future Internet

Thiago R. Colucci

09-Dec-2010

PDF

Introduction to ESB and PEtALS

Fernando Hattori

09-Dec-2010

PDF

Introduction to BPMN

Gustavo Ansaldi Oliva

11-Nov-2010

PDF

User-centric Dynamic Service Composition

Luis Ferreira Pires, University of Twente

04-Nov-2010

PDF

Web Services Orchestration and Choreography

Short-course presented on CBSoft 2010

Emilio Francesquini and Fabio Kon

27-Sep-2010

PDF (in Portuguese)

Hadoop MapReduce

Felipe Besson

07-Jul-2010

PDF

Choreographies

Alfonso Phocco Diaz and Glaucus Augustus Grecco Cardoso

31-May-2010

PDF

Orchestration - What is and how to do it?

Thiago R. Colucci

07-May-2010

PDF

Cloud Computing

Carlos Eduardo Moreira dos Santos

03-May-2010

PDF

Web Services - Basic Concepts

Emilio Francesquini

08-Jan-2010

PDF (in Portuguese)

 

 

 

 


Enabling Scalable Cloud Service Choreographies The Future Internet comprises the existence of Service Internet. Among the various challenges, there is a need for developing techniques for the combination of services. A usual technique, named Orchestration, involves the use of a central node that controls and executes the composition. A more recent alternative, known as Choreography, involves a model in which different services collaborate in a decentralized manner to provide a complex service. In this talk, we will provide an overview of Future Internet and its challenges, focusing on how service choreographies can deal with them. The Baile project, a joint research effort involving the University of São Paulo and HP Labs, investigates how to support service choreographies in this scenario. We will present our initial version of a cloud-enabled middleware architecture and our approach for choreography analysis towards dynamic adaptation support. In addition, we will present a prototype running a sample choreography with automated testing support on top of the Open Cirrus platform.

 

Modeling with I*

This talk surveys the I* modelling language. I* has been proposed by Eric Yu for requirements engineering. Its main objective  is to represent the

requirements of the system to develop in the social context ( actors, materials)
in which it will be embedded.  This global view gives an explanation of "why"
we need the system. It also helps software engineers in choice decisions for
building the system in providing all the possible alternatives and their
qualitative evaluation.
The talk is mainly based on two papers of Eric Yu which detail
the different I* artifacts for capturing requirements and the application of
I* for making social analysis.

 

 



 

 

From Volunteer Computing to Web Service Choreographies in the Cloud
(presented by Prof. Fabio Kon at HP Labs, Palo Alto)

In this seminar, Prof. Fabio Kon will present past, ongoing, and future research projects of the Distributed Systems Research group from the University of São Paulo (USP), Brazil. First, we will discuss the InteGrade, middleware for Opportunistic Grid Computing, developed at USP from 2004 to 2010. It provides a generic infrastructure for the execution of computationally-intensive, parallel applications on non-dedicated, existing machines from companies and academic institutions, such as desktops in instructional and grad student labs. InteGrade supports multiple programming models and provides several mechanisms for application execution in the presence of failures.

Since 2010, the group is studying Web Service Choreographies and is currently developing system support for the enactment of scalable choreographies in the context of Cloud Computing and the Internet of Things. This effort, sponsored by HP Brasil under the BAILE project and by the European Commission under the CHOReOS project, focuses on developing new tools and a middleware for the design, implementation, enactment, monitoring, and runtime reconfiguration of large-scale choreographies. In the seminar, we'll discuss the state of the art in the field and the research work begin carried out at USP with emphasis on the middleware engine for choreography enactment, the framework for development and testing, and support for dynamic reconfiguration and evolution of scalable web service choreographies.

 

 


 

 

CHOReOS: Large Scale Choreographies for the Future Internet

CHOReOS is a high-technology software project addressing one of the most acute challenges of the future Internet of Services: the design and  management of Ultra Large-scale Web Services Coordination. CHOReOS project has three use cases that will be implemented until it ends: Passenger-friendly airport, Mobile-enabled coordination of people, and DynaRoute. In this seminar, we explain all the three use cases.

 

 

 


 

 

Introduction to ESB and PEtALS

The Service-Oriented Architecture (SOA) uses loosely-coupled services to enable business agility, services reusability, and interoperability. The ESB is one of the most fundamental parts of a SOA architecture. As expected from a "bus" technology, ESB is responsible for networking, routing and message delivery while being at the same time transparent to its users.
In this seminar, we will present the open source ESB: PEtALS, supported by OW2 community.
PEtALS differs from other ESB especially thanks to its original distributed architecture, that allows PEtALS to handle a large number of platforms equally, and also because it fully complies with JBI specification and standards.

 

 


 

 

Introduction to BPMN

Business Process Modeling Notation (BPMN) is currently the industry most widely used standard for visual modeling of business processes - collection of related, structured activities that produce a specific service or product for a particular customer or group of customers. Business processes modeling is motivated by a number of reasons, among which we highlight: design, understanding, control and automation of processes, as well as support for effective and non-ambiguous communication.
According to the Object Management Group (OMG), the current maintainer of the standard, the main goal of the specification is to support business and technical users in process management activity. This is done through a graphical notation especially conceived for these users and complete enough to represent complex business processes semantics. More specifically, the language is based on flow charts and has a set of graphical elements that is similar to those offered by the activity diagram of the Unified Modeling Language (UML).
In this seminar, we will give an introduction to the BPMN language and will cover the following topics:

- Introduction to business processes
- BPMN 1.2 and its main graphical elements
- Mapping from BPMN to BPEL (Business Process Execution Language)
- New features of BPMN 2.0 (beta stage), with a focus on choreography and conversation models
- BPMN tools for process modeling

Basic references:

- OMG/Business Process Management Initiative
http://www.bpmn.org/
- Introduction to BPMN (Dr. Jim Marlow slides)
http://www.slideshare.net/jimarlow/introductiontobpmn005
- Business Process Modeling Notation
http://en.wikipedia.org/wiki/Business_Process_Modeling_Notation

 

 


 

 

User-centric Dynamic Service Composition

A composição dinâmica de serviços pode trazer muitos benefícios para os usuários de plataformas de serviços, como, por exemplo, uma melhor adequação dos serviços aos requisitos desses usuários e um tempo menor para a criação de novos serviços. Esses benefícios potenciais têm motivado uma intensa pesquisa nessa área. Entretanto a maior parte dos resultados encontrados na literatura tem aplicação limitada, pois se concentram no problema de achar ou resolver composições, ignorando quase que completamente as características dos usuários que irão utilizar os serviços compostos. Essa apresentação apresenta os resultados obtidos na Universidade de Twente no desenvolvimento do DynamiCoS (http://dynamicos.sourceforge.net), que é uma plataforma para composição dinâmica de serviços centrada nos usuários desses serviços. O DynamiCoS consiste de componentes que permitem a descoberta e composição de serviços partindo dos objetivos dos usuários, utilizando para isso ontologias de domínio. Essa apresentação também discute as dificuldades de se avaliar e comparar diferentes plataformas de composição de serviço e os resultados das avaliações da nossa plataforma. Discutimos também uma categorização dos usuários finais de serviços compostos que inspirou e motivou as nossas novas direções de pesquisa. A apresentação visa dar uma ideia global dos desafios em composição dinâmica de serviços e identificar tópicos para colaboração futura.

 

 


 

 

Web Services Orchestration and Choreography

Web Services technology has been gaining some ground over the last decade and has effectively become a trend both in the software industry and in several kinds of services available on the Internet. By using it, one can encapsulate software services into components accessible through the Internet in such a way that it is independent of hardware platforms, operating systems, and programming languages. This is accomplished via standard protocols and interfaces defined by the World Wide Web Consortium (W3C). In recent years, the need for combinations of simpler Web services into more sophisticated ones has increased significantly. This combination of Web services can be done by means of a centralized node that controls the composition process, which is known as Orchestration. Another alternative, that has been explored lately, is a model in which there is not a central point of coordination. In this alternative, known as Choreography, a set of services collaborate in a decentralized manner to provide the desired service. In this short course we will present the basic concepts of Web services and, then, will describe, in more detail, Web Services Orchestration and Choreography. This short course will cover fundamental concepts, current standards, and existing technologies.

 

 

 


 

 

Hadoop MapReduce

MapReduce is a programming model and a framework developed by Google for processing large-scale data sets (terabytes). Applications developed for this model are executed distributed and paralleled inherently, as issues from distributed computing (e.g., data partition, process scheduling, nodes communication, replication) are treated automatically. Apache Hadoop, an project composed by frameworks a project consisting of frameworks designed to ensure reliability and scalability in distributed computing and data storage, provides an OpenSource implementation of this model, the Hadoop MapReduce. In this seminar, an overview of the Apache Hadoop and some characteristics, architecture and modes of operation of  Hadoop MapReduce and HDFS (High Distributed File System) will be presented.

Further reading (course material from TU Dresden and Google papers; First three lectures): http://wwwse.inf.tu-dresden.de/index.php?language=English&site=courses&course=ss06vl01

 

 

 


 

 

Choreographies

Web Services choreography is a service composition strategy where the interactions between the partner services are defined from a global point of view. In an intuitive manner, service choreographies can be understood as follows: "Dancers dance following a global scenario without a single point of control". During runtime, each participant in the choreography executes its part of it (i.e. play its role) according to the behavior of other participants. In choreographies, a role specifies the expected messaging behavior of the participants that will play it in terms of the sequencing and timing of the messages that they can consume and produce. Just like web services orchestration, choreographies were and still are of interest to industry, with effort being spent on the standardization of specification languages such as WS-BPEL, WS-CDL, BPMN and others.

 

 


 

 

Orchestration - What is and how to do it?

Ten years ago the use of web services did not enjoy such a widespread usage due to the limited Internet bandwidth. Nowadays the Internet connection speeds evolved to a point where we can rely on real-time systems based on these services, with the advantage of a high degree of interoperability. One might consider that Web Service Orchestration is the use of standards to make web services interact with each other, aiming the creation of new of high-level business models. More formally, Web Service Orchestration describes how these services can interact with each other through message exchange, including the business logic and the execution order of the interactions. These interactions can bring applications and / or organizations together and result in a long-lived, transactional and multi-phased business model. This seminar will briefly introduce web services technology, this introduction will be followed by an explanation of what Web Services Orchestration is and the standards for its implementation, as well as some tools for this.

 


 

 

Cloud Computing

From the standpoint of an user, cloud computing provides a great amount of resources (processing, storage, bandwidth) without any initial cost, measured and charged in a pay-per-use fashion. There is no need to concern oneself with the acquisition, maintenance and upgrade of pieces of equipments or with infrastructure management. For a cloud developer, there is a vast and challenging field of research in many different areas. Resources must be multiplexed and shared to lower costs, virtual machines need to migrate from racks and containers so to avoid unnecessary energy consumption, heat generation and environmental impacts, to ensure data security firstly the hardware layer should be reliable. These are some of the problems that must be dealt with in a way that does not disrupt the application performance. In this presentation, a panorama of cloud computing will be shown, its architecture, characteristics, advantages as well as the hurdles and opportunities to its growth. We will also talk about the state-of-the-art implementations and research opportunities.

 

 


 

 

Web Services - Basic Concepts

An oversimplified and direct definition for a web service is a piece of software that, in a distributed and heterogeneous environment, allows communication and data exchange between software applications. Although the tasks performed by web services are not by themselves revolutionary - one could argue that CORBA has been doing similar system integrations for a really long time -, web services technology now enjoys a widespread market acceptance that has never been shared by its predecessors. This widespread acceptance has attracted the gaze of both industry and academia.
In this presentation, I will briefly show the applications and the usage of web services technologies. I will not talk about how someone should build a web service, but I will talk in brief about the main concepts and technologies involved, such as XML, SOAP and REST. I will also remark on the work experience of mine relating to web services during the integration of real systems as well as the integration problems.