<div dir="ltr">Hello,<div><br></div><div style>I have some more fundamental questions. The graphics machine that I am planning to use has 128 processors. I have attached the /proc/cpuinfo and /proc/meminfo output at the end of this email. I've installed the binary version of paraview right now for testing. </div>
<div style><br></div><div style>If I simply execute the command "paraview" on command line, will it not tap the full capacity of the machine? </div><div style>When I run simply "paraview", does the machine just use processor out of the 128 available or it uses all the available resources. Will it be more efficient to run paraview using mpi on this machine (in which case I guess I should install paraview from source)?</div>
<div style><br></div><div style>My regrets, but this is a different machine than that I mentioned in my very first email. Here I am not using any clusters.</div><div style><br></div><div style>Pradeep</div><div style><br>
</div><div style><br></div><div style>Information of the last processor in /proc/cpuinfo</div><div style><br></div><div style>--------------------------------------------</div><div style><div>processor<span class="" style="white-space:pre"> </span>: 127</div>
<div>vendor_id<span class="" style="white-space:pre"> </span>: GenuineIntel</div><div>cpu family<span class="" style="white-space:pre"> </span>: 6</div><div>model<span class="" style="white-space:pre"> </span>: 46</div>
<div>
model name<span class="" style="white-space:pre"> </span>: Intel(R) Xeon(R) CPU X7560 @ 2.27GHz</div><div>stepping<span class="" style="white-space:pre"> </span>: 6</div><div>cpu MHz<span class="" style="white-space:pre"> </span>: 2266.812</div>
<div>cache size<span class="" style="white-space:pre"> </span>: 24576 KB</div><div>physical id<span class="" style="white-space:pre"> </span>: 7</div><div>siblings<span class="" style="white-space:pre"> </span>: 16</div>
<div>
core id<span class="" style="white-space:pre"> </span>: 11</div><div>cpu cores<span class="" style="white-space:pre"> </span>: 8</div><div>apicid<span class="" style="white-space:pre"> </span>: 247</div><div>fpu<span class="" style="white-space:pre"> </span>: yes</div>
<div>fpu_exception<span class="" style="white-space:pre"> </span>: yes</div><div>cpuid level<span class="" style="white-space:pre"> </span>: 11</div><div>wp<span class="" style="white-space:pre"> </span>: yes</div><div>
flags<span class="" style="white-space:pre"> </span>: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx rdtscp lm constant_tsc ida nonstop_tsc pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr sse4_1 sse4_2 popcnt lahf_lm</div>
<div>bogomips<span class="" style="white-space:pre"> </span>: 4533.66</div><div>clflush size<span class="" style="white-space:pre"> </span>: 64</div><div>cache_alignment<span class="" style="white-space:pre"> </span>: 64</div>
<div>address sizes<span class="" style="white-space:pre"> </span>: 44 bits physical, 48 bits virtual</div><div>power management: [8]</div><div>-----------------------------------------------------------</div><div><br></div>
<div style>Output of /proc/meminfo</div><div style><br></div><div style>------------------------------------------------------</div><div style><div>MemTotal: 1054411612 kB</div><div>MemFree: 555700848 kB</div><div>
Buffers: 360100 kB</div><div>Cached: 430018964 kB</div><div>SwapCached: 0 kB</div><div>Active: 24301128 kB</div><div>Inactive: 429373584 kB</div><div>HighTotal: 0 kB</div><div>HighFree: 0 kB</div>
<div>LowTotal: 1054411612 kB</div><div>LowFree: 555700848 kB</div><div>SwapTotal: 2097136 kB</div><div>SwapFree: 2096068 kB</div><div>Dirty: 6252 kB</div><div>Writeback: 0 kB</div><div>
AnonPages: 23296664 kB</div><div>Mapped: 222688 kB</div><div>Slab: 43752860 kB</div><div>PageTables: 56756 kB</div><div>NFS_Unstable: 0 kB</div><div>Bounce: 0 kB</div><div>CommitLimit: 529302940 kB</div>
<div>Committed_AS: 25295660 kB</div><div>VmallocTotal: 34359738367 kB</div><div>VmallocUsed: 403800 kB</div><div>VmallocChunk: 34359330523 kB</div><div>HugePages_Total: 0</div><div>HugePages_Free: 0</div><div>
HugePages_Rsvd: 0</div><div>Hugepagesize: 2048 kB</div><div>--------------------------------------------------------------</div></div></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/1/24 Pradeep Jha <span dir="ltr"><<a href="mailto:pradeep.kumar.jha@gmail.com" target="_blank">pradeep.kumar.jha@gmail.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi David,<div><br></div><div>thanks for the detailed response. I am very new to using Paraview in parallel and still not very familiar with its terminologies. I want to make sure if I understood you correctly.</div>
<div><br></div><div><ol><li>You are saying that I can install Paraview from source in my home directory on the remote machine? </li><li>If I want to use the cluster, installing Paraview from source on the remote (login) mode, such that the computer nodes can access it, should be sufficient?<br>
</li><li>If possible I would like to visualize a data of around 800 GB. A machine that I can use for visualization has the following specifications:
<p><font size="1"><b>PRIMQUEST 1800E</b></font></p>
<p><font size="1"><b>CPU：8CPU (XeonTM X7560 2.26GHz (8core))</b></font></p>
<p><font size="1"><b>MEM：1TB</b></font></p><p>Do you have any intuitive ideas if this will be sufficient?</p></li><li>What is the oldest version of Python that Paraview is compatible with? Is this version fine?
<p><font size="1"><b>Python 2.4.3 (#1, Apr 14 2011, 20:41:59) </b></font></p><p><font size="1"><b>[GCC 4.1.2 20080704 (Red Hat 4.1.2-50)]</b></font></p><p><font size="1"><b><br>
</b></font></p><p>I am sorry but I am very new to parallel visualization and I am having a bit of a tough time putting it all together :)</p><p>Thanks </p><span class="HOEnZb"><font color="#888888"><p>Pradeep</p><p>
<br></p><p><br></p></font></span></li></ol></div>
<div><br></div></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><br><div class="gmail_quote">2013/1/17 David E DeMarle <span dir="ltr"><<a href="mailto:dave.demarle@kitware.com" target="_blank">dave.demarle@kitware.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div>On Thu, Jan 17, 2013 at 2:12 AM, Pradeep Jha<br>
<<a href="mailto:pradeep.kumar.jha@gmail.com" target="_blank">pradeep.kumar.jha@gmail.com</a>> wrote:<br>
> Hello,<br>
><br>
> I am trying to use PV for the first time in parallel. I have some basic<br>
> questions.<br>
><br>
> I have installed PV 3.98 from source on my local machine (Mac OSX). I don't<br>
> have the root permissions for the remote system I want to use for<br>
<br>
</div>You shouldn't need root permissions. You can build paraview in your<br>
home directory (or anywhere that the remote machine can read from) and<br>
connect to that. You generally don't even have to run make install.<br>
<div><br>
> visaulization. So I installed the binary form of PV 3.98 on the remote<br>
<br>
</div>Our binaries typically don't have an MPI in them that will work on the<br>
cluster's network. What we have is more or less just for making use of<br>
the cores on a single node. So you typically have to compile paraview<br>
on the cluster and enable MPI to use it in distributed memory parallel<br>
settings. It should work for you on the remote node, but not the<br>
cluster.<br>
<div><br>
> system. I set up the server on that machine using the command:<br>
><br>
> $mpiexec -n 4 pvserver --use-offscreen-rendering<br>
><br>
> And then I am trying to connect to this server from the PV on my Mac. But it<br>
> is giving me the following error:<br>
</div>> -----------------------------------------------------------------------------------------------------------------------------------------/Users/pradeep/softwares/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx,<br>
<div>> line 332<br>
> vtkTCPNetworkAccessManager (0x7f953b63aa30): Failed to connect to<br>
> localhost:11150. Client-Server Handshake failed. Please verify that the<br>
> client and server versions are compatible with each other<br>
><br>
</div><div>> Any ideas what is going wrong?<br>
<br>
</div>This is most often happens when the client and server are mismatching<br>
versions (3.14 connecting to 3.10 etc), but various network problems<br>
can result in the same thing.<br>
<div><br>
><br>
> Also, this remote system has 16 processors. But for bigger jobs, I can<br>
> submit a job on a cluster using this remote system using a shell script<br>
> where it can use over 100 processors. But I cannot install anything on this<br>
</div>> cluster. ...<br>
<br>
Same comment as before, you only have to have a paraview built<br>
somewhere that the compute nodes can access, so you shouldn't have to<br>
install it.<br>
<br>
> ... Can I use this remote cluster for using paraview faster?<br>
><br>
<br>
A word of caution - more nodes does not necessarily mean faster.<br>
ParaView's parallel architecture allows it to process larger data and<br>
larger data as more disks and memory are added - not the same sized<br>
data faster and faster using more CPUs and GPUs are added (accept in<br>
the infrequent case when processing and rendering are the<br>
bottlenecks).<br>
<div><br>
> The structure is something like this: Mac ---> Remote system (16 processors)<br>
> ----> Cluster with over 100 nodes.<br>
<br>
</div>You'll need to set up an ssh tunnel through the remote (login) node to<br>
the cluster in your job submission to do that.<br>
<br>
> Thanks,<br>
> Pradeep<br>
><br>
<span><font color="#888888"><br>
<br>
David E DeMarle<br>
Kitware, Inc.<br>
R&D Engineer<br>
21 Corporate Drive<br>
Clifton Park, NY 12065-8662<br>
Phone: <a href="tel:518-881-4909" value="+15188814909" target="_blank">518-881-4909</a><br>
</font></span></blockquote></div><br></div>
</div></div></blockquote></div><br></div>