Service, cloud, and fog computing  

This series of seminars aims at introducing some of the main aspects in service-based, cloud-based and fog computing by covering both modelling and analysis aspects (e.g., app deployment, management, QoS prediction) and implementation aspects (e.g., Docker, microservices, DevOps practices). The seminars will also illustrate some of the ongoing research carried on by the SoCC research group, and they will point to some open problems on which the possible cooperation of seminar attendees is welcome.

Introduction to service-based, cloud-based and fog computing Antonio Brogi 12/05/2017 11.00am Sala Seminari Ovest
Fault-aware management of cloud-based applications Jacopo Soldani 19/05/2017 11.00am Sala Seminari Ovest
Microservices and Docker Davide Neri 26/05/2017 12.00am Sala Seminari Ovest
Combining TOSCA and Docker Luca Rinaldi 01/06/2017 11.00am Sala Seminari Ovest
Predicting quality of service of software applications Ahmad Ibrahim 09/06/2017 11.00am Sala Seminari Ovest
App deployment through the fog Stefano Forti 16/06/2017 11.00am Sala Seminari Ovest
1-min pitches Students 05/10/2017 3.00pm Sala Seminari Est

Introduction to service-based, cloud-based and fog computing

by Antonio Brogi

Abstract. This seminar will open the series of seminars on service service-based, cloud-based and fog computing organized for the students our Ph.D. school in Computer Science. The objective of this seminar is to introduce some of the main aspects and challenges of service-based, cloud-based and fog computing, and to put in context the topics of the following seminars of the series.

Fault-aware management of cloud-based applications

by Jacopo Soldani

Abstract. Managing complex applications over heterogeneous clouds is one of the emerging problems in the cloud era. The OASIS Topology and Orchestration Specification for Cloud Applications (TOSCA) aims at solving this problem by providing a language to describe and manage complex cloud applications in a portable and vendor-agnostic way.
This seminar illustrates so-called fault-aware management protocols, which permit modelling the management behaviour of application components by taking into account the potential occurrence of faults. The seminar also shows how such protocols can be composed to analyse the behaviour of a multi-component application and to automate its management. We also illustrate a way to recover applications that are stuck because a fault was not properly handled and/or because a component is behaving differently than expected.

Microservices and Docker

by Davide Neri

Abstract. Microservice-based architecture, or simply microservices, is a method of developing software applications as a suite of independently deployable services in which each service runs a unique process and communicates through a well-defined, lightweight mechanism to serve a business goal. Docker is a container-based technology that permits shipping each microservice within an independent container. This seminar illustrates how microservice-based architectures can be deployed using Docker by illustrating a real case study.

Combining TOSCA and Docker

by Luca Rinaldi

Abstract. Docker is emerging as a simple yet effective solution for deploying and managing multi-component applications in virtualised cloud platforms. Application components can be shipped within portable and lightweight Docker containers toghether with their software dependency, permitting their execution on every Docker-based infrastructure. At the same time, the need for an enhanced support for orchestrating the management of the application components shipped within Docker containers is emerging.
In this seminar we show how TOSCA (Orchestration Specification for Cloud Applications) can be exploited to provide such an enhanced support, by proposing a representation for describing the components forming an application, as well as the Docker containers used to ship such components. We also present TosKer, an engine for orchestrating the management of multi-component applications based on the proposed TOSCA representation and on Docker.

Predicting quality of service of software applications

by Ahmad Ibrahim

Abstract. The ability to a priori predict the Quality of Service (QoS) of a service orchestration is crucial both in the design of service compositions and in the definition of their Service Level Agreements (SLAs). QoS prediction is difficult because of the different possible results of service invocations, and of the nondeterminism (alternatives,unbounded iterations) and complex structure (correlations, dependencies) of the workflows defining service orchestrations.
In this talk we will give an overview of PASO analyzer (Probabilistic Analyser of Service Orchestrations) to probabilistically predict the QoS of WS-BPEL service orchestrations. PASO relies on Monte Carlo simulations, and it improves the results provided by existing approaches by coping with complex structures, unbounded loops, fault handling, and unresponded service invocations. We will illustrate the practical usefulness of PASO by showing how it can be fruitfully exploited to predict the QoS of service orchestrations.

App deployment through the fog

by Stefano Forti

Abstract. Deploying composite applications over Fog infrastructures is challenging due to the heterogeneity and scale of such infrastructures. Application components must be provided with the software and hardware capabilities they need, while communications should meet the desired QoS (e.g., latency and bandwidth). On the other hand, different Fog and Cloud nodes provide different software and hardware capabilities, and actual communication links support different QoS over time. In this talk we will discuss some of the problems of determining how to “best” deploy composite applications over Fog infrastructures, and the benefits of semi-automated tools capable of supporting human-driven application management at the edge.

1-min pitches

by students

Abstract. In this final meeting each participant will make a 1 minute “pitch” to present 1 slide where she/he will identify and comment on a research challenge that she/he found interesting in the area of the seminar.