Algebraic Optimization of RDF Graph Pattern Queries on MapReduce - Large Scale and Big Data: Processing and Management - page 194

POLoad

LOLoad

MR Job 1-1 ( SJ 1) MR Job 1-2 ( SJ 2)

( p 1) ( p 2) ( p 3) ( p 4)

LOSplit M : M :

( p 1) ( p 2) ( p 3) ( p 4)

POFilter
(“ p 1”)

POFilter
(“ p 1”)

LOSplit
Output
(“ p 1”)

LOSplit
Output
(“ p 2”)

LOSplit
Output
(“ p 3”)

LOSplit
Output
(“ p 4”)

POLocal
Rearrange

(…) (…) (…)

POLocal
Rearrange

(…) (…) (…)

LOJoin LOJoin

POUnion

POUnion

POUnion POUnion

POJoin
Package

POJoin
Package

R : POJoinPackage R : POJoinPackage

MR Job 2 ( SJ 1 SJ 2)

POLocal
Rearrange

POLocal
Rearrange

POLocal
Rearrange

POLocal
Rearrange

LOJoin

M :

POJoinPackage

R : POJoinPackage

LOStore

POStore

Logical plan
(a)

Physical plan MapReduce plan

(b) (c)

FIGURE 6.4 (a-c) Translation from logical-to-physical-to-MapReduce plan in Pig.