|
Title:
|
ON THE DEPENDANCIES OF QUERIES EXECUTION TIME AND MEMORY CONSUMPTION IN C-SPARQL |
|
Author(s):
|
Yousra Chabchoub, Zakia Kazi-Aoul, Amadou Fall Dia, Rayane El Sibai |
|
ISBN:
|
978-989-8533-45-6 |
|
Editors:
|
Hans Weghorn |
|
Year:
|
2015 |
|
Edition:
|
Single |
|
Keywords:
|
C-SPARQL queries, semantic DSMS, RDF, Scalability. |
|
Type:
|
Full Paper |
|
First Page:
|
97 |
|
Last Page:
|
104 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Data stream community expressed a growing interest to semantic web technologies over the last few years. They opted to perform stream reasoning over semantic data streams in order to deal with the heterogeneous data streams produced by different sources. Modelling data streams with the standard RDF format guarantees applications interoperability and allows, therefore, wider on-line mining and reasoning perspectives. C-SPARQL is an efficient, well-known semantic DSMS (Data Stream Management Systems), also called RDF (Resource Description Framework) stream processor. We focus in this paper on C-SPARQL. Our objective is to evaluate its performance and scalability in term of query execution time. For this purpose, we considered several C-SPARQL queries with different complexities and evaluated their execution time and memory usage under different conditions: single and multiple streams, varying stream rate and time window length. An adapted choice of all these parameters is necessary to avoid losing stream data and guarantee therefore query results correctness. We discussed, in particular, the consequences of a long execution time queries (compared to the frequency of query executions). |
|
|
|
|
|
|
