Hi Pantelis,

Are you saying the query runs successfully against some other older Virtuoso 
SPARQL endpoint but it returning no results against this newer 3214 build you 
are using from Oct 2015 ? If so what is the version and build data of this 
earlier build.

You need to profile the query and see were the time is being take ie in 
compilation or execution , by running it from isql with the “set profile on” 
option run to generate a query profile as detailed at:
        
        
http://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtTipsAndTricksAanalyzingSPARQLQuery

There are also some suggestion on how to optimise queries detailed at:

        
http://virtuoso.openlinksw.com/dataspace/doc/dav/wiki/Main/VirtQueryOptDiagnostic

I would suggest particularly trying the “Enable_Joins_Only = 1” option in the 
“[Flags]” section  of the INI  which is the new default for the latest 3215 
builds released this month, I would also recommend you update to ...

Best Regards
Hugh Williams
Professional Services
OpenLink Software, Inc.      //              http://www.openlinksw.com/
Weblog   -- http://www.openlinksw.com/blogs/
LinkedIn -- http://www.linkedin.com/company/openlink-software/
Twitter  -- http://twitter.com/OpenLink
Google+  -- http://plus.google.com/100570109519069333827/
Facebook -- http://www.facebook.com/OpenLinkSoftware
Universal Data Access, Integration, and Management Technology Providers

> On 29 Dec 2015, at 09:08, Pantelis Natsiavas <natsia...@gmail.com> wrote:
> 
> The [SPARQL] settings section of my virtuoso.ini file initially had a 
> value for MaxQueryCostEstimationTime. I have just commented it out and 
> now I don't get the error anymore.
> 
> However, I actually get no response on my query. I use the web interface 
> and when I click "Execute" button, I get no response. The web interface 
> does not give me a sign of executing my query. I only get a timeout 
> after half an hour.
> 
> Could you suggest any way to improve performance by avoiding or changing 
> the "OPTIONAL" clauses?
> 
> Once more, thank you very much.
> 
> Kind regards,
> Pantelis Natsiavas
> 
> 
> On 28/12/2015 05:56 μμ, Hugh Williams wrote:
>> Hi Pantelis,
>> 
>> Do you have “MaxQueryCostEstimationTime” set in the “[SPARQL]” section of 
>> the INI file, which is normally commented out by default ? As the 
>> "MaxQueryCostEstimationTime” should never be set in the INI file in 
>> production use as the Cost Optimizer estimates are generally not accurate 
>> and this is more a development/debugging param. Thus the only one that 
>> should be set to control query execution time is “MaxQueryExecutionTime".
>> 
>> Best Regards
>> Hugh Williams
>> Professional Services
>> OpenLink Software, Inc.      //http://www.openlinksw.com/
>> Weblog   --http://www.openlinksw.com/blogs/
>> LinkedIn --http://www.linkedin.com/company/openlink-software/
>> Twitter  --http://twitter.com/OpenLink
>> Google+  --http://plus.google.com/100570109519069333827/
>> Facebook --http://www.facebook.com/OpenLinkSoftware
>> Universal Data Access, Integration, and Management Technology Providers
>> 
>>> On 28 Dec 2015, at 12:13, Pantelis Natsiavas<natsia...@gmail.com>  wrote:
>>> 
>>> Hi everybody.
>>> 
>>> I use a Virtuoso instance installed on ubuntu (Virtuoso version: 
>>> 07.20.3214, Build: Oct 14 2015), in order to execute sparql queries. My 
>>> queries are quite complex and make heavy use of the "OPTIONAL" pattern. 
>>> While I have used the same queries on another virtuoso endpoint (not 
>>> controlled by me) I cannot seem to use them on mine. I keep getting the 
>>> message
>>> 
>>> "Virtuoso 42000 Error The estimated execution time 0 (sec) exceeds the 
>>> limit of 18000 (sec)."
>>> 
>>> As you see, I have changed the timeout parameters in the virtuoso.ini file 
>>> and put an irrational value (18000 seconds!!!). Still, it seems that the 
>>> engine would refuse to execute my query.
>>> 
>>> My questions are:
>>> 1. Is the value "0" mentioned as execution time a special value or is it 
>>> just a bug? Is it logical to assume that the virtuoso estimates that the 
>>> query would require infinite time to execute?
>>> 2. Are there any other configurations that I could do in order to improve 
>>> situation?
>>> 3. Is there anything that I should keep in mind when I use the "OPTIONAL" 
>>> sparql query pattern? I mean, is there a documented (even informally) way 
>>> of improving the performance of queries using OPTIONAL clauses?
>>> 
>>> Kind regards,
>>> Pantelis Natsiavas
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> Virtuoso-users mailing list
>>> Virtuoso-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/virtuoso-users
> 
> 
> ------------------------------------------------------------------------------
> _______________________________________________
> Virtuoso-users mailing list
> Virtuoso-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/virtuoso-users

Attachment: smime.p7s
Description: S/MIME cryptographic signature

------------------------------------------------------------------------------
_______________________________________________
Virtuoso-users mailing list
Virtuoso-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/virtuoso-users

Reply via email to