[Mpi-21] Review of topology chapter 6

Richard Barrett rbarrett at [hidden]
Tue Mar 4 10:42:01 CST 2008

In general this section needs updating. It often reads like its trying to
convince the reader that the functionality is worthwhile. Would also benefit
from an update to citations in support of capabilities and features. Some
text is clunky.

Suggestions, corrections, observations, based on MPI 2.1, draft 2008-02-23.

1. p231, line 41: The citation supporting the performance potential of graph
defined topologies is from 1989. Are there more recent citations to
supplement? Similar question for other citations throughout the chapter.
2. p231, line 47: ³Šwith tremendous benefits for program readability². I
appreciate the optimistic tone, but perhaps a little less hype: ³with
potential benefitsŠ²
3. p232, line 6: ³The nodes stand forŠ². How about, ³represent²?
4. p232, line 7: ³MPI provides message-passing between any pair of processes
in a group.² This seems to imply that topologies are only applicable to
point-to-point communication. If so, we should clearly say so, earlier.
5. p232, line 17: Citation [9] is listed as ³To appear.²
6. p232, Lines 23-29. This seems like unnecessary rationalization. Unless it
can be substantiated (with a citation), it should be removed. How about,
³When the graph structure is regular and can be completely defined by the
number of dimensions and the number of processes in each coordinate
direction, such as rings, two- or higher dimensional grids, or tori, a
simpler description can be madeŠ² or the like. And then talk about support
for Cartesian topologies, which suddenly is mentioned on 6.4, line 3
7. p233, line 5. How about, ³Šare collective, and therefore must adhere to
that category¹s requirements.² Clunky, but the ideaŠ
8. p233, line 19: ³Similar functions are contained in EXPRESS and PARMACS.²
Can we update (or eliminate) this? Sounds like, ³All the other kids are
doing it.² :)
9. p233, line 22-33. Reads clunky. In particular, ³foo can be used toŠ²
Implies they are designed primarily for something else. And perhaps provide
a high level sentence regarding the content of the paragraph: ³A variety of
inquiry functions are provided.² (Which then calls for a bit of a
reorganization of the paragraph, since sub-setting functionality is
discussed as well. 
10. p234, line 27. ³choose a beneficial² rather than ³good². (Could be
great! Or ok, orŠ) 
11. p237, line 19-20. ³If a topology has been defined with one of the above
functionsŠ² How else? Instead: ³Information regarding a process topology can
be returned using  inquiry functions.² Or something like that.
12. p238, line 6. Is ³dimension² a verb?
13. p241, line 15. I prefer, ³Suppose this topology is associated with
communicator comm...²
14. p241, line 34. ³Šan MPI_SENDRECV², since Œa¹ and Œan¹ are selected based
on the vowel or consonant sound. Or are these fightin¹ words? :)

1. pg 237, line 26: (choice) s/b (status).
2. The other errata seems to have been satisfied.
Long example, p246. Any verification of correctness available? No complaints
in the errata.

  Richard Barrett
  Future Technologies Group, Computer Science and Mathematics Division, and
  Scientific Computing Group, National Center for Computational Science
  Oak Ridge National Laboratory

More information about the Mpi-21 mailing list