Examples of org.apache.pig.newplan.optimizer.PlanOptimizer

Package org.apache.pig.newplan.optimizer

Examples of org.apache.pig.newplan.optimizer.PlanOptimizer

org.apache.pig.newplan.optimizer.PlanOptimizer
The core class of the optimizer. The basic design of this class is that it is provided a list of RuleSets. RuleSets represent all of the optimizer rules that can be run together. The rules in the RuleSet will be run repeatedly until either no rule in the RuleSet passes check and calls transform or until maxIter iterations (default 500) has been made over the RuleSet. Then the next RuleSet will be moved to. Once finished, a given RuleSet is never returned to. Each rule is has two parts: a pattern and and associated transformer. Transformers have two important functions: check(), and transform(). The pattern describes a pattern of node types that the optimizer will look to match. If that match is found anywhere in the plan, then check() will be called. check() allows the rule to look more in depth at the matched pattern and decide whether the rule should be run or not. For example, one might design a rule to push filters above join that would look for the pattern filter(join) (meaning a filter followed by a join). But only certain types of filters can be pushed. The check() function would need to decide whether the filter that it found was pushable or not. If check() returns true, the rule is said to have matched, and transform() is then called. This function is responsible for making changes in the logical plan. Once transform is complete PlanPatcher.patchUp will be called to do any necessary cleanup in the plan, such as resetting schemas, etc.

    }


    private LogicalPlan migrateAndOptimizePlan(String query) throws Exception {
        PigServer pigServer = new PigServer( pc );
        LogicalPlan newLogicalPlan = Util.buildLp(pigServer, query);
        PlanOptimizer optimizer = new MyPlanOptimizer( newLogicalPlan, 3 );
        optimizer.optimize();
        return newLogicalPlan;
    }

View Full Code Here

        String query = "a = load 'd.txt' as (id, v1, v2);" +
        "b = filter a by v1==NULL;" +        
        "store b into 'empty';";  
        LogicalPlan newLogicalPlan = buildPlan(query);
               
        PlanOptimizer optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query  = "a = load 'd.txt' as (id, v1, v2);" +
        "b = filter a by v1==NULL;"  +       
        "store b into 'empty';";  
        LogicalPlan expected = buildPlan(query);
        
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // no schema
        query = "a = load 'd.txt';" +
        "b = foreach a generate $0, $1;" +
        "store b into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a = load 'd.txt';"+
        "b = foreach a generate $0, $1;"+
        "store b into 'empty';";  
        expected = buildPlan(query);

View Full Code Here

        String query = "a = load 'd.txt' as (id, v1, v2);" +
        "b = foreach a generate id;"+        
        "store b into 'empty';";  
        LogicalPlan newLogicalPlan = buildPlan(query);
               
        PlanOptimizer optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a = load 'd.txt' as (id);" +
        "b = foreach a generate id;"+        
        "store b into 'empty';";  
        LogicalPlan expected = buildPlan(query);
        
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // with filter
        query = "a = load 'd.txt' as (id, v1, v5, v3, v4, v2);"+
        "b = filter a by v1 != NULL AND (v2+v3)<100;"+
        "c = foreach b generate id;"+
        "store c into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a = load 'd.txt' as (id, v1, v3, v2);" +
        "b = filter a by v1 != NULL AND (v2+v3)<100;" +
        "c = foreach b generate id;" +
        "store c into 'empty';"; 
        expected = buildPlan(query);
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // with 2 foreach
        query = "a = load 'd.txt' as (id, v1, v5, v3, v4, v2);" +
        "b = foreach a generate v2, v5, v4;" +
        "c = foreach b generate v5, v4;" +
        "store c into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a = load 'd.txt' as (v5, v4);" +
        "b = foreach a generate v5, v4;" +
        "c = foreach b generate v5, v4;" +
        "store c into 'empty';"; 
        expected = buildPlan(query);
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // with 2 foreach
        query = "a = load 'd.txt' as (id, v1, v5, v3, v4, v2);" +
        "b = foreach a generate id, v1, v5, v3, v4;" +
        "c = foreach b generate v5, v4;" +
        "store c into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a = load 'd.txt' as (v5, v4);" +
        "b = foreach a generate v5, v4;" +
        "c = foreach b generate v5, v4;" +
        "store c into 'empty';"; 
        expected = buildPlan(query);
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // with 2 foreach and filter in between
        query = "a =load 'd.txt' as (id, v1, v5, v3, v4, v2);" +
        "b = foreach a generate v2, v5, v4;" +
        "c = filter b by v2 != NULL;" +
        "d = foreach c generate v5, v4;" +
        "store d into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' as (v5, v4, v2);" +
        "b = foreach a generate v2, v5, v4;" +
        "c = filter b by v2 != NULL;" +
        "d = foreach c generate v5, v4;" +
        "store d into 'empty';";  
        expected = buildPlan(query);
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // with 2 foreach after join
        query = "a =load 'd.txt' as (id, v1, v2, v3);" +
        "b = load 'c.txt' as (id, v4, v5, v6);" +
        "c = join a by id, b by id;" +       
        "d = foreach c generate a::id, v5, v3, v4;" +
        "store d into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' as (id, v3);" +
        "b = load 'c.txt' as (id, v4, v5);" +
        "c = join a by id, b by id;" +  
        "d = foreach c generate a::id, v5, v3, v4;" +
        "store d into 'empty';";  
        expected = buildPlan(query);
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // with BinStorage, insert foreach after load
        query = "a =load 'd.txt' using BinStorage() as (id, v1, v5, v3, v4, v2);" +        
        "c = filter a by v2 != NULL;" +
        "d = foreach c generate v5, v4;" +
        "store d into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' using BinStorage() as (id, v1, v5, v3, v4, v2);" +
        "b = foreach a generate v5, v4, v2;" +
        "c = filter b by v2 != NULL;" +
        "d = foreach c generate v5, v4;" +
        "store d into 'empty';";  
        expected = buildPlan(query);
        assertTrue(expected.isEqual(newLogicalPlan));
        
       // with BinStorage, not to insert foreach after load if there is already one
        query = "a =load 'd.txt' using BinStorage() as (id, v1, v5, v3, v4, v2);" +    
        "b = foreach a generate v5, v4, v2;" +
        "c = filter b by v2 != NULL;" +
        "d = foreach c generate v5;" +
        "store d into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' using BinStorage() as (id, v1, v5, v3, v4, v2);" +
        "b = foreach a generate v5, v2;" +
        "c = filter b by v2 != NULL;" +
        "d = foreach c generate v5;" +
        "store d into 'empty';";  
        expected = buildPlan(query);
        assertTrue(expected.isEqual(newLogicalPlan));
        
       // with BinStorage, not to insert foreach after load if there is already one
        query = "a =load 'd.txt' using BinStorage() as (id, v1, v5, v3, v4, v2);" +    
        "b = foreach a generate v5, v4, v2, 10;" +
        "c = filter b by v2 != NULL;" +
        "d = foreach c generate v5;" +
        "store d into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' using BinStorage() as (id, v1, v5, v3, v4, v2);" +
        "b = foreach a generate v5, v2, 10;" +
        "c = filter b by v2 != NULL;" +
        "d = foreach c generate v5;" +

View Full Code Here

        String query = "a =load 'd.txt' as (id, v1, m:map[]);" +
        "b = foreach a generate id, m#'path';" +        
        "store b into 'empty';";  
        LogicalPlan newLogicalPlan = buildPlan(query);
               
        PlanOptimizer optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' as (id, m:map[]);" +
        "b = foreach a generate id, m#'path';" +        
        "store b into 'empty';";  
        LogicalPlan expected = buildPlan(query);
        
        assertTrue(expected.isEqual(newLogicalPlan));
        
        LOLoad op = (LOLoad)newLogicalPlan.getSources().get(0);
        Map<Integer,Set<String>> annotation = 
                (Map<Integer, Set<String>>) op.getAnnotation(MapKeysPruneHelper.REQUIRED_MAPKEYS);
        assertEquals(annotation.size(), 1);
        Set<String> s = new HashSet<String>();
        s.add("path");
        assertEquals(annotation.get(2), s);
        
        // foreach with join
        query = "a =load 'd.txt' as (id, v1, m:map[]);" +
        "b = load 'd.txt' as (id, v1, m:map[]);" +
        "c = join a by id, b by id;" +
        "d = filter c by a::m#'path' != NULL;" +
        "e = foreach d generate a::id, b::id, b::m#'path', a::m;" +        
        "store e into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' as (id, m:map[]);" +
        "b = load 'd.txt' as (id, m:map[]);" +
        "c = join a by id, b by id;" +
        "d = filter c by a::m#'path' != NULL;" +

View Full Code Here

        "c = foreach b generate id, FLATTEN(v);" +    
        "d = foreach c generate id, v::s2;" +    
        "store d into 'empty';";  
        LogicalPlan newLogicalPlan = buildPlan(query);
               
        PlanOptimizer optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' as (id, v:bag{t:(s1,s2,s3)});" +
        "b = filter a by id>10;" +
        "c = foreach b generate id, FLATTEN(v);" +    
        "d = foreach c generate id, v::s2;" +

View Full Code Here

        "b = filter a by v1>10;" +
        "c = foreach b generate id;" +        
        "store c into 'empty';";  
        LogicalPlan newLogicalPlan = buildPlan(query);
               
        PlanOptimizer optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' as (id, v1);" +
        "b = filter a by v1>10;" + 
        "c = foreach b generate id;" +      
        "store c into 'empty';";  
        LogicalPlan expected = buildPlan(query);
        
        assertTrue(expected.isEqual(newLogicalPlan));
        
        // join with foreach
        query = "a =load 'd.txt' as (id, v1, v2);" +
        "b = load 'd.txt' as (id, v1, v2);" +
        "c = join a by id, b by id;" +
        "d = filter c by a::v1>b::v1;" +
        "e = foreach d generate a::id;" +        
        "store e into 'empty';";  
        newLogicalPlan = buildPlan(query);
               
        optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        optimizer.optimize();
        
        query = "a =load 'd.txt' as (id, v1);" +
        "b = load 'd.txt' as (id, v1);" +
        "c = join a by id, b by id;" +
        "d = foreach c generate a::id, a::v1, b::v1;" +

View Full Code Here

                          " e = distinct d; " +
                          " generate group, e; };" +
            "f = foreach c generate group ;" +
            "store f into 'empty';";
        LogicalPlan newLogicalPlan = buildPlan(query);
        PlanOptimizer optimizer = new MyPlanOptimizer(newLogicalPlan, 3);
        try {
            optimizer.optimize();
        } catch (Exception e) {
            //PIG-2968 throws ConcurrentModificationException
            e.printStackTrace();
            fail("Unexpected Exception: " + e);
        }

View Full Code Here

        Rule r = new SplitFilter("SplitFilter");
        Set<Rule> s = new HashSet<Rule>();
        s.add(r);
        List<Set<Rule>> ls = new ArrayList<Set<Rule>>();
        ls.add(s);
        PlanOptimizer optimizer = new MyPlanOptimizer(plan, ls, 3);
        optimizer.optimize();
        
        Assert.assertEquals(plan.getPredecessors(filter).get(0), join);
        Operator next = plan.getSuccessors(filter).get(0);
        Assert.assertEquals(LOFilter.class, next.getClass());        
        next = plan.getSuccessors(next).get(0);
        Assert.assertEquals(LOStore.class, next.getClass());
        
        // run push up filter rule
        r = new PushUpFilter("PushUpFilter");
        s = new HashSet<Rule>();
        s.add(r);
        ls = new ArrayList<Set<Rule>>();
        ls.add(s);
        optimizer = new MyPlanOptimizer(plan, ls, 3);
        optimizer.optimize();
        
        // both filters should be moved up to be after each load
        next = plan.getSuccessors(load1).get(0);
        Assert.assertEquals(next.getClass(), LOFilter.class);
        Assert.assertEquals(plan.getSuccessors(next).get(0), join);
        
        next = plan.getSuccessors(load2).get(0);
        Assert.assertEquals(next.getClass(), LOFilter.class);
        Assert.assertEquals(plan.getSuccessors(next).get(0), join);
        
        Assert.assertEquals(plan.getSuccessors(join).get(0), store);
        
        // run merge filter rule
        r = new MergeFilter("MergeFilter");
        s = new HashSet<Rule>();
        s.add(r);
        ls = new ArrayList<Set<Rule>>();
        ls.add(s);
        optimizer = new MyPlanOptimizer(plan, ls, 3);
        optimizer.optimize();
        
        // the filters should the same as before, nothing to merge
        next = plan.getSuccessors(load1).get(0);
        Assert.assertEquals(next.getClass(), LOFilter.class);
        Assert.assertEquals(plan.getSuccessors(next).get(0), join);

View Full Code Here

    }


    private LogicalPlan migrateAndOptimizePlan(String query) throws Exception {
      PigServer pigServer = new PigServer(pc);
        LogicalPlan newLogicalPlan = Util.buildLp(pigServer, query);
        PlanOptimizer optimizer = new NewPlanOptimizer( newLogicalPlan, 3 );
        optimizer.optimize();
        return newLogicalPlan;
    }

View Full Code Here

    private void assertQuerySame(String origQuery, String optimizedQuery) throws Exception {
        LogicalPlan newLogicalPlan = Util.buildLp(pigServer, origQuery);


        SchemaTupleBackend.initialize(ConfigurationUtil.toConfiguration(pc.getProperties(), true),
                pc);
        PlanOptimizer optimizer = new MyPlanOptimizer(newLogicalPlan, 10);
        optimizer.optimize();


        LogicalPlan expected = Util.buildLp(pigServer, optimizedQuery);


        assertTrue(expected.isEqual(newLogicalPlan));

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.pig.newplan.optimizer.PlanOptimizer

org.apache.pig.newplan.OperatorPlan

org.apache.pig.test.TestConstantCalculator

org.apache.pig.test.TestFilterSimplification

org.apache.pig.test.TestMergeForEachOptimization

org.apache.pig.test.TestNewPartitionFilterPushDown

org.apache.pig.test.TestNewPlanColumnPrune

org.apache.pig.test.TestNewPlanFilterAboveForeach

org.apache.pig.test.TestNewPlanFilterRule

org.apache.pig.test.TestNewPlanPushDownForeachFlatten

org.apache.pig.test.TestNewPlanPushUpFilter

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.