BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Metadata Management - Giorgio Gianforme\, Università Roma Tre
DTSTART:20090319T113000Z
DTEND:20090319T120000Z
UID:TALK17390@talks.cam.ac.uk
CONTACT:Dr Fabien Petitcolas
DESCRIPTION:*Abstract*: The management of heterogeneous databases\, in int
 egrated or collaborative contexts\, always involves the need for solutions
  to data programmability issues. In general\, data programmability address
 es problems dealing with evolving scenarios: changes in a database which c
 ollaborates in a heterogeneous environment often imply a sequence of propa
 gating changes in related databases at any level\, model\, schema\, and da
 ta.\n\nIn this scenario there is the need to translate data and their desc
 riptions from one model (i.e. data model) to another. Even small variation
 s of models are often enough to create difficulties. For example\, while m
 ost database systems are now object-relational\, the actual features offer
 ed by different systems rarely coincide\, so data migration requires a con
 version. Every new database technology introduces more heterogeneity and t
 hus more need for translations.\n\nAccording to the model management propo
 sal\, these problems can be solved conveniently applying the ModelGen oper
 ator\, that can be defined as follows using our terminology: given two mod
 els M_1 and M_2 and a schema S_1 of M_1\, ModelGen translates S_1 into a s
 chema S_2 of M_2 that properly represents S_1.\nIn my PhD thesis I present
 ed our theoretical and practical contribution to the development of an eff
 ective implementation of a generic (i.e.model independent) platform for sc
 hema and data translation. We improve the expressive power of its supermod
 el\, that is the set of models handled and accuracy and precision of such 
 models representation. We show how it is possible to automatically reason 
 on models and schemas and how to and a suitable translation given a source
  and a target model exploiting a formal system\, proved to be sound and co
 mplete. Then we propose an extension of Datalog based on the use of hierar
 chies and a sort of polymorphism\, that provides a significant simplificat
 ion in the definition of translations and a higher level of reuse in the s
 pecification of elementary translations. Finally we present a new\, lightw
 eight\, runtime approach to the translation problem\, where translations o
 f data are performed directly on the operational system.\n\n*Biography*: G
 iorgio Gianforme is a research associate (“assegnista di ricerca”) at 
 Università Roma Tre. He  studied  in  Rome  at  Università  Roma  Tre\, 
  where  he  obtained  the  “Laurea”  (five  year  program\, equivalent
   to  a Master  degree)  in Computer  Engineering  in May  2005 with maxim
 um  grade  and  highest honours (110/110 cum laude). \n\nAfter graduation\
 , he  joined the database group of Università Roma Tre and then he starte
 d his Ph.D.  in Computer Engineering\,  co-funded  by Microsoft Research  
 through  their  European PhD  Scholarship Programme.  \n\nHe worked under 
 the supervision of Prof. Paolo Atzeni on mappings and transformations of s
 chemas and data in heterogeneous  settings and presented  several  results
  at  international  conferences and  in one of the top database journals. 
  He will defend his Ph.D. thesis on next 2nd of April.
LOCATION:Small Lecture Room\, Microsoft Research\, Roger Needham Building\
 , 7 J J Thomson Avenue\, Cambridge CB3 0FB
END:VEVENT
END:VCALENDAR