com.bigdata.rdf.sparql.ast.eval.rto.TestRTO_LUBM Maven / Gradle / Ivy
Show all versions of bigdata-rdf-test Show documentation
/**
Copyright (C) SYSTAP, LLC DBA Blazegraph 2006-2016. All rights reserved.
Contact:
SYSTAP, LLC DBA Blazegraph
2501 Calvert ST NW #106
Washington, DC 20008
[email protected]
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
/*
* Created on Sep 4, 2011
*/
package com.bigdata.rdf.sparql.ast.eval.rto;
import java.util.Properties;
import com.bigdata.rdf.axioms.NoAxioms;
import com.bigdata.rdf.sail.BigdataSail;
/**
* Data driven test suite for the Runtime Query Optimizer (RTO) using LUBM data
* and queries based on LUBM.
*
* @author Bryan Thompson
* @version $Id: TestBasicQuery.java 6440 2012-08-14 17:57:33Z thompsonbry $
*/
public class TestRTO_LUBM extends AbstractRTOTestCase {
// private final static Logger log = Logger.getLogger(TestRTO_LUBM.class);
/**
*
*/
public TestRTO_LUBM() {
}
/**
* @param name
*/
public TestRTO_LUBM(String name) {
super(name);
}
@Override
public Properties getProperties() {
// Note: clone to avoid modifying!!!
final Properties properties = (Properties) super.getProperties().clone();
properties.setProperty(BigdataSail.Options.TRIPLES_MODE, "true");
properties.setProperty(BigdataSail.Options.AXIOMS_CLASS,
NoAxioms.class.getName());
return properties;
}
/**
* LUBM Q2 on the U1 data set.
*
* Note: There are no solutions for this query against U1. The optimizer is
* only providing the fastest path to prove that. In practice, this amounts
* to proving that one of the joins fails to produce any results.
*
* Note: For U50 there are 130 solutions and it discovers the join order
* [3, 5, 6, 4, 1, 2].
*/
public void test_LUBM_Q2() throws Exception {
final TestHelper helper = new TestHelper(//
"rto/LUBM-Q2", // testURI,
"rto/LUBM-Q2.rq",// queryFileURL
"src/test/resources/data/lehigh/LUBM-U1.rdf.gz",// dataFileURL
"rto/LUBM-Q2.srx"// resultFileURL
);
/*
* Verify the expected join ordering.
*/
final int[] expected = new int[] { 5, 6, 1, 2, 3, 4 };
assertSameJoinOrder(expected, helper);
}
/**
* LUBM Q8 on the U1 data set.
*
* Note: For U50 there are 6463 solutions and it discovers the join order:
* [4,1,3,2,5].
*/
public void test_LUBM_Q8() throws Exception {
final TestHelper helper = new TestHelper(//
"rto/LUBM-Q8", // testURI,
"rto/LUBM-Q8.rq",// queryFileURL
"src/test/resources/data/lehigh/LUBM-U1.rdf.gz",// dataFileURL
"rto/LUBM-Q8.srx"// resultFileURL
);
/*
* Verify that the runtime optimizer produced the expected join path.
*/
final int[] expected = new int[] { 1, 4, 3, 2, 5 };
assertSameJoinOrder(expected, helper);
}
/**
* LUBM Q9 on the U1 data set.
*
* Note: For U50 there are 8627 solutions and it discovers the join order:
* [2,5,6,4,1,3].
*
* Note: When random sampling is used, several join plans are possible
* including:
*
*
* {4, 2, 3, 5, 0, 1}
* {3, 0, 5, 4, 1, 2}
* {1, 4, 3, 5, 0, 2} - this solution has the best performance on U50.
*
*
* These all appear to be good join plans even though the last one is is
* definitely better.
*/
public void test_LUBM_Q9() throws Exception {
final TestHelper helper = new TestHelper(//
"rto/LUBM-Q9", // testURI,
"rto/LUBM-Q9.rq",// queryFileURL
"src/test/resources/data/lehigh/LUBM-U1.rdf.gz",// dataFileURL
"rto/LUBM-Q9.srx"// resultFileURL
);
/*
* Verify that the runtime optimizer produced the expected join path.
*/
final int[] expected = new int[] { 5, 3, 4, 6, 1, 2 };
assertSameJoinOrder(expected, helper);
}
}