0.25
0.5
0.75
1.25
1.5
1.75
2
Intelligent clients for replicated Triple Pattern Fragments
Published on Jul 10, 2018504 Views
Following the Triple Pattern Fragments (TPF) approach, intelligent clients are able to improve the availability of the Linked Data. However, data availability is still limited by the availability of T
Related categories
Chapter list
Intelligent Clients for Replicated Triple Pattern Fragments00:00
Introduction00:10
Server Availability00:33
SPARQL Query load-balancing between Replicated RDF Datasets00:59
Problem01:33
Related Work02:00
Triple Pattern Fragments02:03
Linked Data Replication02:48
Client-side load-balancing03:51
Ulysses approach04:19
Query evaluation over replicas04:25
Servers throughputs change over time04:58
Where to send Triple Pattern Queries?05:22
Ulysses: a replication-aware intelligent TPF client05:43
Partial replication model06:10
Ulysses replication-aware source selection06:46
How to get server throughput?07:18
Computing Server throughput - 107:26
Computing Server throughput - 208:24
Computing Server throughput - 308:44
Hard to compare: normalize! - 109:12
Hard to compare: normalize! - 209:23
Computing TPF servers capabilities - 109:36
Computing TPF servers capabilities - 209:52
Computing TPF servers capabilities - 310:01
Ulysses in action - 110:28
Ulysses in action - 210:34
Weighted random access of TPF servers - 111:16
Weighted random access of TPF servers - 211:26
Weighted random access of TPF servers - 311:34
Experimental Study11:54
Experimental setup - 112:02
Experimental setup - 212:40
Ulysses balance the load according to servers processing capabilities - 113:22
Ulysses balance the load according to servers processing capabilities - 213:52
Ulysses balance the load according to servers processing capabilities - 314:24
Ulysses improves query execution time under the load15:06
Ulysses tolerates server failures16:10
Ulysses in real-life16:56
Conclusion17:15
Future works17:40
Inteligent Clients for Replicated Triple Pattern Fragments18:22