<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Hi Alaa,<br>
<br>
Some information is still missing.<br>
- Number of processes per node<br>
- The word document shows you are comparing three algorithms, but
you only mentioned two of them. What is the third one ?<br>
- What is the size of small/medium/large arrays in your experiments
?<br>
- I see Bcast speedup is better than Send in Figure (c), but the
numbers in the table do not show that. What is the number of
processes corresponding to each result in the table ? <br>
<br>
Since we do not support MPICH on Windows anymore, perhaps you would
like to use MS-MPI for better performance. If you still observe the
same issue, you can discuss with the Microsoft MPI team.<br>
<a class="moz-txt-link-freetext" href="https://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_can.27t_I_build_MPICH_on_Windows_anymore.3F">https://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_can.27t_I_build_MPICH_on_Windows_anymore.3F</a><br>
<br>
Min<br>
<div class="moz-cite-prefix">On 9/7/16 8:59 AM, alaa nashar wrote:<br>
</div>
<blockquote cite="mid:1975972539.311838.1473256789920@mail.yahoo.com" type="cite">
<div style="color:#000; background-color:#fff;
font-family:HelveticaNeue, Helvetica Neue, Helvetica, Arial,
Lucida Grande, sans-serif;font-size:24px">
<div id="yui_3_16_0_ym19_1_1473243871493_32628"><span>Hi Min</span></div>
<div id="yui_3_16_0_ym19_1_1473243871493_32620"><span id="yui_3_16_0_ym19_1_1473243871493_32619">Thanks a lot for
your fast response.</span></div>
<div id="yui_3_16_0_ym19_1_1473243871493_32621">The file
attached contains the required information. Regards</div>
<div id="yui_3_16_0_ym19_1_1473243871493_32621"><br>
</div>
<div id="yui_3_16_0_ym19_1_1473243871493_32621">Alaa</div>
<div id="yui_3_16_0_ym19_1_1473243871493_32621"><br>
</div>
<div class="qtdSeparateBR"><br>
<br>
</div>
<div class="yahoo_quoted" style="display: block;">
<div style="font-family: HelveticaNeue, Helvetica Neue,
Helvetica, Arial, Lucida Grande, sans-serif; font-size:
24px;">
<div style="font-family: HelveticaNeue, Helvetica Neue,
Helvetica, Arial, Lucida Grande, sans-serif; font-size:
16px;">
<div dir="ltr"><font size="2" face="Arial"> On Wednesday,
September 7, 2016 4:21 PM, msi <a class="moz-txt-link-rfc2396E" href="mailto:msi@anl.gov"><msi@anl.gov></a>
wrote:<br>
</font></div>
<br>
<br>
<div class="y_msg_container">
<div id="yiv9595793598">
<div>
<div>Hi Alaa,</div>
<div><br>
</div>
<div>Generally, for small number of processes, the
performance of MPI_Bcast is usually equal to
MPI_Isend/MPI_Recv, for large number of processes,
broadcast should be better.</div>
<div><br>
</div>
<div>Could you please try the lateat version of
MPICH on you system ? MPICH2 is very old. </div>
<div><br>
</div>
<div>Please also give us following information.</div>
<div>Number of processes</div>
<div>Number of processes per node</div>
<div>The execution time of each algorithm</div>
<div><br>
</div>
<div>Min</div>
<div><br>
</div>
<div><br>
</div>
<div><br>
</div>
<div id="yiv9595793598composer_signature">
<div style="font-size:9px;">Sent via my cell
phone.</div>
</div>
<br>
<br>
-------- Original message --------<br>
From: alaa nashar <a class="moz-txt-link-rfc2396E" href="mailto:nashar_al@yahoo.com"><nashar_al@yahoo.com></a> <br>
Date: 9/7/2016 7:04 AM (GMT-06:00) <br>
To: <a class="moz-txt-link-abbreviated" href="mailto:discuss@mpich.org">discuss@mpich.org</a> <br>
Subject: [mpich-discuss] MPI_Send and MPI_Bcast <br>
<br>
<div style="color:#000;background-color:#fff;font-family:HelveticaNeue,
Helvetica Neue, Helvetica, Arial, Lucida Grande,
sans-serif;font-size:24px;">
<div id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21673"><span></span></div>
<div id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21674"> <span style="font-family:Helvetica;font-size:18pt;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21961">Dear
all</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21962"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21963"> </span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21964"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21965">I have
implemented the following two algorithms on my
home Ethernet LAN containing 2/
and three heterogeneous devices using MPICH2.</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21967"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21968"> </span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21970"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21971">Algorithm1:</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21973"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21974">1- Root
process reads the contents of an input array.</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21976"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21977">2- It
then sends the array data to all other
processes.</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21979"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21980">3- All
processes including the root process perform
specific</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21982"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21983">
computations on their array copies.</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21985"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21986">4- Once
the computations within each process are
finished,</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21988"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21989">
the generated arrays are directly written by
the process that</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21991"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21992">
performed the computations to separate files.</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21994"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21995"> </span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21997"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21998">Algorithm2:</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22000"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22001">Same as
Algorithm 1 except step 2 is replace by:</span></div>
<div dir="LTR" style="text-indent:6pt;direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22003"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22004">2- Root process
broadcast the array data to all other
processes.</span></div>
<div dir="LTR" style="text-indent:6pt;direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22006"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22007"> </span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22009"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22010">The
implementation of both algorithms work fine
and give the expected results
but Algorithm2 that uses MIPI_Bcast is slower
than Algorithm1 that uses
MPI_Send</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22012"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22013"> </span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22015"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22016">For my
knowledge, MIPI_Bcast is faster than MPI_Send.</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22018"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22019">Please,
would you guide me to know if there is any
conflict of misunderstand.</span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22021"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22022"> </span></div>
<div dir="LTR" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22024"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22025">Thanks
a lot</span></div>
<div dir="LTR" style="text-indent:6pt;direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22027"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22028"> </span></div>
<div dir="LTR" style="text-indent:6pt;direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22030"><span style="font-size:18pt;font-family:Helvetica;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22031">Alaa</span></div>
<div dir="ltr" style="direction:ltr;unicode-bidi:embed;" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_22033">
</div>
<div class="yiv9595793598signature" id="yiv9595793598yui_3_16_0_ym19_1_1473243871493_21675"><br>
</div>
</div>
</div>
</div>
<br>
<br>
</div>
</div>
</div>
</div>
</div>
</blockquote>
<br>
</body>
</html>