SaGe: Web Preemption for Public SPARQL Query Services

Abstract

To provide stable and responsive public SPARQL query services, data providers enforce quotas on server usage. Queries which exceed these quotas are interrupted and deliver partial results. Such interruption is not an issue if it is possible to resume queries execution afterward. Unfortunately, there is no preemption model for the Web that allows for suspending and resuming SPARQL queries. In this paper, we propose SaGe: a SPARQL query engine based on Web preemption. SaGe allows SPARQL queries to be suspended by the Web server after a fixed time quantum and resumed upon client request. Web preemption is tractable only if its cost in time is negligible compared to the time quantum. The challenge is to support the full SPARQL query language while keeping the cost of preemption negligible. Experimental results demonstrate that SaGe outperforms existing SPARQL query processing approaches by several orders of magnitude in term of the average total query execution time and the time for first results.

SMART-KG: Hybrid Shipping for SPARQL Querying on the Web

“…4, Virtuoso behaves very similar to SaGe with the difference that i) it shows no timeout and similar performance for 10 clients, but ii) its execution time is slightly degrading with 80 clients. Given these results and in line with previous studies [36], in the following sections, we focus on comparing the performance of smart-KG with the shipping strategies of TPF and SaGe only. Increasing KG Size.…”

“…Finally, SaGe [36] is a SPARQL query engine tailored to address the undesirable starvation of simple queries waiting for long running queries to release the server resources. To this aim, SaGe proposes a preemptive execution model.…”

“…We configured Virtuoso and SaGe to run with 4 workers [36]. We omit the comparison with brTPF [27], as existing evaluations report that the performance lie between TPF and SaGe [36], and our preliminary tests show scalability problems of the brTPF server implementation [1] for the WatDiv-100M and Watdiv-1000M graphs. Hardware Setup.…”

RDF-TR: Exploiting structural redundancies to boost RDF compression

“…HDT has been widely adopted by the Semantic Web community because of its simplicity, its compression levels and its performance for data retrieval operations. It is worth noting that HDT is successfully deployed in client-side query processors, such as Triple Pattern Fragments 4 [50] and SAGE 5 [35], indexing/reasoning systems like HDT-FoQ [31] or WaterFowl [11], or recommender systems [19] among others. However, its encoding of the graph topology is quite simple and further compression could be achieved.…”

Processing SPARQL Aggregate Queries with Web Preemption

“…However, processing aggregate queries with TPF generates tremendous data transfer and delivers poor performance. Recently, the Web preemption approach [12] relies on a preemptable server that suspends queries after a quantum of time and resumes them later. The server supports joins, projections, unions, and some filters operators.…”