GRaCe: a relaxed approach to graph caching

Francesco De Fino

GRaCe: a relaxed approach to graph caching
Francesco De Fino
Caching approaches are widely used in graph query processing to optimize elaboration of user requests. Queries, typically expressed in SPARQL, must access heterogeneous datasets, such as Linked Data, in order to be executed, making query processing inefficient for multiple reasons. Caching framework are exploited in order to speed up the process storing previously executed requests together with information related to them, such as their results. However, most of the approaches rely on precise matching techniques, which reduces the number of possible hits in the cache for a given user request, forcing the query to be executed in a traditional way. In order to provide a more flexible solution, together with improving the query processing, we propose GRaCe, a caching framework which implements a relaxed graph matching technique that enables the system to provide more general solutions, at cost of reducing the query result precision.
Short Bio:
Francesco is a third year PhD student in Computer Science (supervisors: Barbara Catania, Giovanna Guerrini). He is graduated in Computer Science, his main research interest concerns query processing optimization for graph queries. Currently, he is working on approximate caching for optimizing query processing in Linked Data environment, facing source selection and result selection problems.
Thursday, 6th June 2019
DIBRIS, Valletta Puggia, Conference Hall (322) - Via Dodecaneso 35, Genova 


Sei qui: Home Ricerca Seminari di ricerca GRaCe: a relaxed approach to graph caching

Questo sito usa i cookie.