Mapping to Processor Subsets

 

This extension allows objects to be directly distributed to processor subsets by allowing a processor subset to be specified where a processor could be named, e.g., in a DISTRIBUTE directive. The specified subset must be a proper subset of the named processor arrangement.

The syntax of the extended dist-target is as follows:

H806 extended-dist-target

is processors-name [ ( section-subscript-list ) ] or * processors-name [ ( section-subscript-list ) ] or *

• The section-subscripts in the section-subscript-list may not be vector-subscripts and are restricted to be either subscripts or subscript-triplets.
• In the section-subscript-list, the number of section-subscripts must equal the rank of the processor-name.
• Within a DISTRIBUTE directive, each section-subscript must be a specification-expr.
• Within a DISTRIBUTE or a REDISTRIBUTE directive, if both a dist-format-list and a dist-target appear, the number of elements of the dist-format-list that are not ``*'' must equal the number of subscript-triplets in the named processor arrangement.
• Within a DISTRIBUTE or a REDISTRIBUTE directive, if a dist-target appears but not a dist-format-list, the rank of each distributee must equal the number of subscript-triplets in the named processor arrangement.

!Example 1
!HPF$ PROCESSORS P(10)
      REAL A(100)
!HPF$ DISTRIBUTE A(BLOCK) ONTO P(2:5)

!Example 2
!HPF$ PROCESSORS Q(10,10)
      REAL A(100,100)
!HPF DISTRIBUTE B(BLOCK,BLOCK) ONTO Q(5:10,5:10)

In Example 1, the array A is distributed by block across the processors P(2) to P(5) while in the second example, the array B is distributed across the lower right quadrant of the processor array Q.

Advice to users.This extension is most useful in conjunction with the tasking construct, see Section 9.4, which allows multiple independent phases of a computation to execute simultaneously on different subsets of processors. A similar situation arises when the code uses multiple data structures which can be computed in parallel where the computation on each individual object also exhibits parallelism, e.g., the multiple blocks in a multi-block grid used in some fluid dynamics calculation. Here, the individual blocks have to be distributed over subsets of processors to exploit both levels of parallelism. (End of advice to users.)