<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<p>Thank you Chuck and I appreciate your detailed reply very much!</p></blockquote><div>You welcome :-) <br><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<p>I have to say that I'm a rookie in the field of parallel computing and visualization.</p></blockquote><div>The only way to gain knowledge and experience is by doing. It will pass and you'll get more comfortable with the various techniques the more you use it over time.<br><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><p>In my case, the dataset consists only of random numbers and it's in a single partition. I will try to generate a dataset in parallel in a few days and render it as you told me. </p></blockquote><div>5M point's is actually a pretty small dataset that should be easy to handle from a rendering standpoint, i.e. youshould be fine handling it on a single node. Most of your time is likely spent in I/O and data conversion rather than rendering. How are you getting your data into ParaView, i.e. what file format are you using? There is likely a better way that my address most of the performance issues you are having and remove the need for parallel rendering entirely.<br></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><p>Actually, my ultimate task is using ParaView Catalyst to implement in situ visualization of a simulation that generates large datasets, but I'm still in the very beginning stage of learning the Catalyst User's Guide.</p></blockquote><div>If that's the case then you won't have the IO time with Catalyst that youre seeing with the ParaView application as the data will end up getting passed directly in memory.<br><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><p>I know there are two GPUs in my rendering server but I'm not sure whether they work for the rendering because they are NVIDIA TESLA C2050, which is designed for scientific computing.</p></blockquote><div>The C2050s should support the fuill OpenGL stack and they will certainly give you better performance than the CPU based rendering. Which driver version are you using (check with the nvidia-smi command)? If it's new enough, you may be able to use EGL which would be ideal.<br></div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><p>I've tried many times to build ParaView 5.2.0 on my server, but finally I failed to get it to work. It seems that it has been built successfully but whenever I connect to it from the 5.2.0 paraview GUI on my desktop, it crashes, and on the server</p></blockquote><div>This is a seperate issue and I'd be happy to help you get 5.2 up in running. If you'd still like to pursue it, please start a new thread by posting a message with the actual error you're getting.<br></div><div> </div><br></div>- Chuck<br></div></div>