To: | Ontology Summit 2014 discussion <ontology-summit@xxxxxxxxxxxxxxxx> |
---|---|
From: | Victor Chernov <vchernov@xxxxxxxxxxxxxx> |
Date: | Wed, 2 Apr 2014 19:03:53 +0400 |
Message-id: | <1002045149.20140402190153@xxxxxxxxxxxxxx> |
Colleagues, I wanted to give you some explanation why we found it expedient to participate in the Hackathon with this particular project: "Optimized SPARQL performance management via native API". Often the problem arises: it is impossible (or extremely hard) to get rather complex data structure by one SPARQL query, but the same time pretty simple to get all the data using a combination of simple queries, and then assemble this complex structure at the application level. This way may require hundreds of queries and consequently hundreds of calls to database. Using endpoint in such cases may be rather time-consuming due to the large overhead of HTTP access. In such cases it is good if manufacturer provides native API that allows running queries directly, bypassing HTTP. Objective of our Hackaton project was to find out the triplestores reaction speed to simple queries through native API. We’ve run SP2Bench only on one computer and not as a benchmark, but ensure that dataset (obtained from sp2bench generator) is loaded correctly. Rest of participants performed only simple queries. The experiments showed that quite often it makes sense to simplify queries by transferring a number of operations from database to application level. For example, comparing the execution time of Query 20 and Query 40, differing only by sorting the result, we have seen that sorting of one million records on database level, may take more than 4 seconds. Thus it would be worth to transfer sorting to application level, where it would take a fraction of a second. Regards, Victor mailto:vchernov@xxxxxxxxxxxxxx
_________________________________________________________________ Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/ Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/ Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx Community Files: http://ontolog.cim3.net/file/work/OntologySummit2014/ Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2014 Community Portal: http://ontolog.cim3.net/wiki/ (01) |
<Prev in Thread] | Current Thread | [Next in Thread> |
---|---|---|
|
Previous by Date: | Re: [ontology-summit] Hackathon Short Report: "Ontology Design Patterns and Semantic Abstractions in Ontology Integration", Mike Bennett |
---|---|
Next by Date: | Re: [ontology-summit] OntologySummit2014 session-12: Synthesis-II: Technical Tracks & Hackathon - Thu 2014.04.03, Peter Yim |
Previous by Thread: | Re: [ontology-summit] Ontology Summit 2014 Hackathon - Optimized SPARQL performance management via native API Status report, Kingsley Idehen |
Next by Thread: | [ontology-summit] Reference data for Anime and Manga - Report for Hackathon project, Victor Agroskin |
Indexes: | [Date] [Thread] [Top] [All Lists] |