Re: [mpi3-coll] Comments on nonblocking collectives section
Torsten and Jesper: Re:
sorry for my late input. I've had time to read the document, and have found something. I refer to the version of 21st January. Thanks for your review - I'll comment below!
I'm clipping items on which I have no comments.
p. 49, line 26: "the performance of many systems..." -> "the performance of many applications can be improved by overlapping communication and computation, and many systems enable this." fixed
p. 49, line 28: "utilize overlap -> "exploit overlap" fixed
Reasonable. I see a split infinitive there also: to utilize overlap and avoid synchronization => to exploit overlap and to avoid synchronization
p. 49, line 32: drop "the use of" (grammar: "the use of" is no mechanism) nb collective communication doesn't increase performance by itself, only the correct *use* of them leads to higher performance, cf. MPI-2.1 p47:45. I changed "the use of" to "to use" which might be better English.
That does sound better. You could delete "mechanism" in "An alternative mechanism" earlier in the sentence since "to use collective communication" no longer seems like a mechanism but an implementation choice but it is not critical.
p.49, line 35: "a collective communication, which..." -> "a collective communication operation, that..." ("that" better than "which" here?) don't know - I'll leave it as is since Bronis didn't comment on this before
Proper English uses "that" with no comma for required clauses -- clauses without which the sentence does not make sense -- and "which" preceded by a comma for all others (usually apository but can have other functions). This clause does not seem required to me so I would go with ", which" but I could see an argument for "that". In any event, ", that" would be incorrect.
p.50, line 35: "In particular," -> "This means that" fixed
Hmm. I think at this point neither of these phrases really are needed. I would just use a colon instead of breaking the two sentences: Blocking and nonblocking collective operations are different collective operations: if some process calls a collective operation, all other processes in the communicator must eventually call the same collective operation, and no other collective operation in between. You might insert "blocking or nonblocking" between "same" and "collective" but I don't think it is critical.
p.50, line 48: "collective operations is required" -> "collective operation semantics is required" I don't understand this proposal and I think the current text is precise.
I agree with Torsten. Bronis
participants (1)
-
Bronis R. de Supinski