Giuseppe Attardi

You are here

31 Oct 2016

Giuseppe Attardi

Full professor at the University of Pisa

 

Giuseppe Attardi is full professor at the University of Pisa.
He has been visiting professor at MIT, Berkeley, Sony Research and Yahoo! Research. He has been director of the department Computing Centre and has been on the board of Centro SerRA and GARR, where he
currently coordinates the Cloud and Storage Department. In these roles he has been active in the development of networks, both at the local and national level.

 

He designed the green datacenter of his university at S. Piero a Grado. He developed Omega, an ontology formalism for the Semantic Web; the first window system at MIT; the garbage collectors for Java; Arianna, the first Italian search engine; the dependency parser DeSR; WikiExtractor, a tool for extracting texts from the Wikipedia.

 

His current research is about Deep Learning techniques for natural language analysis, sentiment analysis, knowledge extraction and language inference. He received research grants from Google, Yahoo!, Sun, Microsoft, Nvidia and HP. He has founded several spin-offs: DELPHI, Ideare, Link.it, Comprendo, KiwiLocal and WebSays.


When: Day 2 - 15th November, Session 10, 14:00 - 16:00

A Reference Architecture for a Federated Cloud for Research

Abstract. Deploying a cloud infrastructure allows sharing resources through virtualization and allows simpler provisioning of storage and computing resources exposed as cloud-based services.
This document presents the high level reference architecture for a federated cloud where different research organizations can contribute portions of their private cloud and cooperate to stand up a common cloud services platform.
Users of the federation will be able to combine resources from any cloud for deploying their services.
The architecture is based on the OpenStack cloud technology, developed by a large world-wide open source community.
Building a cloud remains a complex task, which requires the integration of several disparate components.
To simplify the process, to reduce the required man-power, to reduce risks and to keep services always up to date, we exploit tools for automating the steps needed to create, implement and support a production-ready OpenStack cloud.
The cloud federation can provide a playground where the research community can experiment and share best practices and solutions.