 |
The Eleventh International Conference on Autonomic and Autonomous Systems ICAS 2015 May 24 - 29, 2015 - Rome, Italy |
| Tutorials |
T1. Dataconda
Prof. Dr. Michele Samorani, University of Alberta School of Business, Canada
T2. Network Function Virtualization and Software Defined Networking Cooperation
Prof. Dr. Eugen Borcoci, University „Politehnica” Bucharest, Romania
DETAILS
T1. Dataconda
Prof. Dr. Michele Samorani, University of Alberta School of Business, Canada
Dataconda is a software program, freely available to academics on www.dataconda.net, which solves classification and regression problems in a relational database, as opposed to a single table. The user selects a class attribute contained in a table of a relational database, and the software builds and selects predictors by exploring the whole database and aggregating information, without any user intervention. For example, Dataconda may find that the best predictor for “customer value” is the amount of money spent by the customer in cheap electronics, even if the user has not built any such attribute.
This tutorial will first cover some theoretical background on attribute generation, including existing approaches for data denormalization. Then, it will consider a simple “sales” database and show how to use Dataconda to find the drivers of product returns. Here, the discussion will focus on how to select the best settings for the specific problem at hand. Finally, the tutorial will discuss strengths and limitations of Dataconda, and it will end by mentioning how to extend some of its functionalities.
Attendees are encouraged to download the trial version of Dataconda before the tutorial.
T2. Network Function Virtualization and Software Defined Networking Cooperation
Prof. Dr. Eugen Borcoci, University „Politehnica” Bucharest, Romania
Pre-requisites: general knowledge on IP networking architectures, protocols and SDN technologies
The emerging Network Functions Virtualization (NFV) technologies aim to enhance the flexibility (in terms of design, deployment and management) of network services and to reduce the time to market. NFV decouples network services from the hardware delivering them. Therefore, different network - related functions, such as load balancing, network address translation (NAT), firewalling, intrusion detection, domain name service, caching, etc., can be delivered in software and deployed on general purpose servers. However, NFV exposes also important several challenges such as the network performance guarantees for virtual appliances, dynamic instantiation and migration, and efficient placement of Virtual Network Functions (VNF).
On the other side, recently proposed Software Defined Networking (SDN) architectures and technologies, together with vertical (e.g., OpenFlow) protocols constitute an approach of high interest for both research and industry communities. In SDN, the control plane and data planes are decoupled, while network intelligence is more centralized, thus offering a better and more flexible control of the network resources allocation, routing, traffic engineering, quality of services, etc. The data plane network forwarders become programmable via open interfaces. NFV can be seen as complementary to SDN; they both aim to improve the overall network management and increase flexibility and programmability, but having slightly different goals and different methods to achieve them. While SDN separates the control and forwarding planes thus offering a centralized network view, NFV is primarily focused on optimizing the network services themselves. SDN technology enables the flexible routing of traffic within an NFV infrastructure to improve the efficiency and maximize the overall agility of the network. Several major standardization organizations are active in both NFV/SDN areas: the European Telecommunications Standards Institute (ETSI) NFV Industry Specification Group, the Internet Engineering Task Force (IETF), and the Open Networking Foundation (ONF), as well as major industry- led open-source projects as OpenStack and OpenDaylight.
This tutorial compiles recent information on the main concepts of NFV, requirements, architecture, use cases and a summary of SDN. Then NFV/SDN cooperation is discussed. While the virtualization and deployment of network functions do not mandatory rely on SDN technologies, the NFV/SDN combination can create an overall greater value. (e.g., SDN can support NFV to enhance its performance, facilitate its operation and simplify the backward compatibility with legacy deployments). Samples and examples of relevant industry realizations are given. Some open issues and challenges are also emphasized.
