2nd Workshop on Cloud Computing in Cyprus: Opportunities and Challenges
Department of Computer Science, University of Cyprus & Microsoft Cyprus
Wednesday, June 3, 2015 from 11:00 AM to 5:00 PM (EEST), Nicosia, Cyprus
University of Cyprus, Department of Computer Science and Microsoft Cyprus organise the 2ndworkshop under the title: "Cloud Computing in Cyprus: Opportunities and Challenges". The workshop aims at presenting three tutorials on specific aspects of Big Data and Cloud Computing.
11:00 -11:10 Welcome Note
George Pallis, Assistant Professor, Department of Computer Science, University of Cyprus
11:10 -12:30 Cloud Application Management Framework
Marios D. Dikaiakos, Professor, Department of Computer Science, University of Cyprus
Chrystalla Sofokleous, CAMF contributor, Department of Computer Science, University of Cyprus
Nicholas Loulloudes, CAMF contributor, Department of Computer Science, University of Cyprus
12:30 -14:00 Lunch Break
14:00 -15:00 High Performance Computing, Machine Learning on Microsoft Azure
Alice Crohas, Solution Specialist, Education HQ, Microsoft Central Eastern Europe
Ingo Laue, Solution Specialist, Education HQ, Microsoft Central Eastern Europe
15:00 -16:00 Deep Analysis with Apache Flink
Andreas Kunft, TU Berlin
Cloud Application Management Framework
CAMF is a Cloud Application Management Framework through which users are able to define the description, deployment and management phases of their Cloud applications in a clean and intuitive graphical manner. In this tutorial we will talk about Cloud Computing, its history and its essential characteristics. We will also present ongoing research in the context of the CELAR FP7 project (www.celarcloud.eu) which aims to support elastic resource provisioning for Cloud applications. CAMF originates from the research activities within the CELAR project, and it is a newly established open-source technology project under the Eclipse Foundation (http://projects.eclipse.org/projects/technology.camf). This tutorial will teach participants how to deploy and manage applications over a Cloud provider using CAMF. Specifically, you will learn how to describe Cloud applications of different complexity (from single-tier to three-tier applications), how to deploy them over our OpenStack-compliant infrastructure, and finally how to monitor the live performance of your deployments via the built-in monitoring system of CAMF.
Marios D. Dikaiakos is a professor of computer science at the University of Cyprus and director of the University's Centre for Entrepreneurship. Dikaiakos has a PhD in computer science from Princeton University. His research interests focus on large-scale distributed computing systems.
Nicholas Loulloudes is a computer science doctoral candidate at the University of Cyprus, Nicosia. His research interests include elastic cloud computing, vehicular computing and complex networks. Loulloudes is an active member in the open source Eclipse ecosystem and currently the project lead of the Cloud Application Management Framework (CAMF) project.
Chrystalla Sofokleous works at the University of Cyprus, Nicosia, as a researcher in the FP7 EU project CELAR. Her research interests focus on the Web, cloud computing, and elastic resource provisioning. She has an MSc in Web Technology from the University of Southampton.
High Performance Computing, Machine Learning on Microsoft Azure Tutorial
Microsoft Azure is an open and flexible global cloud platform that supports any language, tool, or framework, and is ideally suited to researchers' needs across disciplines. In this tutorial, you will learn how Azure provides on-demand compute resources that enable you to run large parallel and batch compute jobs in the cloud. You can extend your on-premises HPC cluster to the cloud when you need more capacity, or run work entirely in Azure. You will also learn how Azure Machine Learning can facilitate and accelerate data analysis and scientific research by enabling you to perform machine learning over data in the cloud and deploy machine learning web services for community use.
Ingo Laue has studied Experimental Physics at Technical University at Brunswick, Germany and finished with his PhD. He has got 20 years IT experience in various job roles (IT engineer, consultant, presales, marketing) at EDS, Oracle and Microsoft.
Alice Crohas holds a dual MSc in Electrical Engineering from the University of Notre Dame (US) and Supelec (France), as well as an MBA from IMD (Switzerland). She has worked in China for Archos, a French manufacturer of Windows & Android devices, in project management and purchasing. She is also experienced in systems & software engineering, working for Thales on the automation of a subway line.
In their current Microsoft roles, Ingo and Alice are Solution Specialists for cloud services for Central and Eastern European customers in the Education market.
Deep Analysis with Apache Flink
The tutorial focuses on advanced data analysis in Apache Flink, a massive parallel data processing engine. Apache Flink originated from a research project at Technische Universität Berlin et.al. and combines database techniques with concepts of MapReduce-like engines. The expressive features of Flink, e.g. support for arbitrary DAG's and Iterations, make it well suited for complex tasks from various fields in data science. In the tutorial, we give an overview of the architecture and programming model of Flink. We show how to implement algorithms beyond simple text analysis from machine learning and graph analysis and the attendees have the opportunity to get hands-on experience in Flink programming.
Andreas Kunft is a research associate at Technische Universität Berlin in the Database Systems and Information Management Group (DIMA) since April 2014. Prior to that, he received his M.Sc. degree at TU Berlin and worked as an assistant at the compiler group. His research foci are massive parallel processing frameworks, optimization techniques, and compilers, were he is especially interested in combining optimizations applied to traditional database systems with techniques found in compilers to improve performance and stability of data flow engines.