<div class="adn ads" style="padding-bottom:20px;border-left-width:1px;border-left-style:solid;border-left-color:transparent;padding-left:4px;color:rgb(34,34,34);font-family:arial,sans-serif;font-size:medium;background-color:rgb(255,255,255)">
<div class="gs" style="margin-left:30px"><div id=":1mz" class="ii gt adP adO" style="font-size:12.727272033691406px;margin:5px 15px 0px 0px;padding-bottom:5px"><div id=":1n0">Dear Sir or Madam,<div><br></div><div>I have some questions about communication optimizations on shared memory in state-of-art mpich2.</div>
<div><br></div><div>1) When configure mpich2, does it support shared memory optimization in default? If so, can it select the communication pattern automatically according to hardware architecture? For example in a SMPs cluster, using shared memory optimization within a node and using socket cross nodes?</div>
<div><br></div><div>2) What about collective communication on shared memory? Is it implemented on top of the optimized point to point communication or use other mechanisms?</div><div><br></div><div>Best Regards,</div><div>
Shigang.</div><div class="yj6qo"></div></div></div><div class="hi" style="background-color:rgb(242,242,242);padding:0px;margin:0px;width:auto;border-bottom-left-radius:0px;border-bottom-right-radius:0px;background-repeat:initial initial">
</div></div><div class="ajx" style="clear:both"></div></div><div class="gA gt ac5" style="font-size:12.727272033691406px;background-color:rgb(255,255,255);padding:0px;margin:0px;width:auto;border-bottom-left-radius:0px;border-bottom-right-radius:0px;border-right-width:0px;border-bottom-width:0px;border-left-width:0px;border-top-style:none;color:rgb(34,34,34);font-family:arial,sans-serif">
<br class="Apple-interchange-newline"></div>