From d72cc014651bf52313b036800d92530c78f9afce Mon Sep 17 00:00:00 2001 From: carey Date: Thu, 3 Feb 2022 14:09:13 +0000 Subject: [PATCH] update --- .DS_Store | Bin 10244 -> 10244 bytes code/.DS_Store | Bin 8196 -> 8196 bytes lectures/12_combining_MPI_and_OpenMP.html | 20 ++++++++++---------- lectures/12_combining_MPI_and_OpenMP.org | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) diff --git a/.DS_Store b/.DS_Store index e729e2f890bfed8921641eedd8e7c230764b64fd..c42e5a712512c171fe810fab64916dd5340567cd 100644 GIT binary patch delta 106 zcmZn(XbISGT5z(7aH)uqnSqXiv8h3=jzYDi5s+$qiESs4{nVA8VK_5E+ delta 100 zcmV-q0Gt1WP=rvh%Mk&&lQ0l*C>eWuGB`6VAT2R5GJPx{dpR;NEFd&FGJSms2o)nQ zJ5pb3Y;A6Di<8R}9|4S$gc7fl%Mo=5H8nUaATcyJldKRwlT;iLv;PwY46_Fy`UJE8 G6axbs)gN&H diff --git a/code/.DS_Store b/code/.DS_Store index a2782b3915c4eafccb522e6448c9846964d8b5cc..c6ff96090d954128d0c07a41028b6ed6ce7a4093 100755 GIT binary patch delta 59 zcmZp1XmQwJC@{H0xKzZ*%s@xM*wmm_N1@u%2*@@uHk&LcVm!G-P<%71AQSs$Hi@4s OlS70xH*1RCVFCc@*bpHA delta 83 zcmZp1XmQwJD8RUL@&my_4aw?iBMW041zkf!qgow>YD*&n9R(9hquN?d4pC)&>!A4T noZP(p?#T)w@{B!`3k0P$vj{S=Z)TVH!7|xORBN+_=xrtd1Hl*f diff --git a/lectures/12_combining_MPI_and_OpenMP.html b/lectures/12_combining_MPI_and_OpenMP.html index 24f3d04..eea95e5 100644 --- a/lectures/12_combining_MPI_and_OpenMP.html +++ b/lectures/12_combining_MPI_and_OpenMP.html @@ -27,18 +27,18 @@
-

Combining MPI and OpenMP

Dr Carey Pridgeon

Created: 2021-11-12 Fri 13:37

+

Combining MPI and OpenMP

Dr Carey Pridgeon

Created: 2022-02-03 Thu 13:42

-

Combining MPI and OpenMP

-
+

Combining MPI and OpenMP

+
-

+

  • While MPI could be used as a parallelism library, it isn't a particularly efficient one.
    • @@ -55,7 +55,7 @@ the nodes properly.
-

+

  • On this module I've only covered OpenMP in any detail, but any parallelism library would work in combination with MPI.
    • @@ -71,7 +71,7 @@ computational time.
-

+

  • There's no inherent advantage, in terms of speed alone, between OpenMP and TBB, as they both falter at the same number of cores due to the inter core @@ -89,12 +89,12 @@ nature.
-

+

  • To integrate any Parallelism library, or a Concurrency library if you wanted to doesn't take anything like the detail you might think.
  • To continue, open the source code in -code/distributed/demonstration_code/mpi_omp and we'll go through this +code/distributed/demonstration_code/OpenMP_and_MPI and we'll go through this example program together.
@@ -109,9 +109,9 @@ example program together.
-

Obligatory XKCD

+

Obligatory XKCD

-
+

multi.png

diff --git a/lectures/12_combining_MPI_and_OpenMP.org b/lectures/12_combining_MPI_and_OpenMP.org index 54f6202..d9b3f45 100755 --- a/lectures/12_combining_MPI_and_OpenMP.org +++ b/lectures/12_combining_MPI_and_OpenMP.org @@ -53,7 +53,7 @@ - To integrate any Parallelism library, or a Concurrency library if you wanted to doesn't take anything like the detail you might think. - To continue, open the source code in - *code/distributed/demonstration_code/mpi_omp* and we'll go through this + *code/distributed/demonstration_code/OpenMP_and_MPI* and we'll go through this example program together.