Re: [Paraview] About ParaView catalyst

2015-10-19 Thread Andy Bauer 
Hi Jun,

I'm a bit confused. Do you mean Cetus at Argonne? Also, are you talking
about the Intel compilers or the IBM XL compilers? I'm not familiar with
any Intel XL compilers.

Assuming that you're talking about the IBM XL compilers on a BG/Q, I know
that there were some issues with that but they may have been resolved in
the ParaView superbuild (http://www.paraview.org/Wiki/ParaView/Superbuild),
which includes Catalyst. If you try it out and have issues though, please
email the mailing list so that we can try and fix them.

FYI: ParaView has also been built with the Intel compilers on many
platforms but I'm not sure about a BG/Q.

Regards,
Andy

On Mon, Oct 19, 2015 at 3:54 PM, Jun Fang  wrote:

> Hello,
>
> Here I have some questions about the in-situ library support. Does anyone
> know if the ParaView catalyst is now supported by Intel XL compilers or
> not? Based on my recent scaling test on INL Cetus, the XL compilers are
> almost twice efficient compared to the GNU compilers for the identical
> source codes, compilation flags (-O2), simulation case set-up (16,385
> cores) and input parameters. Therefore, it becomes a tough decision for me
> to choose either "GNU+in-situ capability" or "XL without in-situ". The
> perfect situation is that XL compilers can also support insitu
> visualization (i.e. ParaView catalyst), but I am not sure about that.
>
> Thanks a lot,
> Jun
>
> --
> Jun Fang
> Department of Nuclear Engineering
> North Carolina State University
> Phone: (919) 592-2233
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview

Re: [Paraview] About ParaView catalyst

2015-10-19 Thread David E DeMarle 
Back in the day, with bluegene-P I was able to get ParaView compiled with
xlc working so it should be possible. (And before that others did it on
BGL.) However - I struggled with xl compilers on the mira@anl installs for
several days (see the dreaded -mminimal-toc error below). Despite great
help from the alcf support guys, and trying many many things, I eventually
gave up and went with the gnu compilers. But ... sometime after those few
terrible days that I do not care to repeat, Todd Gamblin at llnl
contributed a BGQ platform file to CMake. With that and latest cmake, there
is some hope that the xlc compile will just work.

/bgsys/drivers/ppcfloor/gnu-linux/powerpc64-bgq-linux/bin/ld:
/bgsys/drivers/V1R2M0/ppc64/comm/sys/lib/libpami.a(component-pami--CAPI.cnk.o)(.text._ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E[_ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E]+0xbc):
sibling call optimization to
`_ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E'
does not allow automatic multiple TOCs; recompile with -mminimal-toc or
-fno-optimize-sibling-calls, or make
`_ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E'
extern



David E DeMarle
Kitware, Inc.
R Engineer
21 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-881-4909

On Mon, Oct 19, 2015 at 5:13 PM, Andy Bauer  wrote:

> Hi Jun,
>
> I'm a bit confused. Do you mean Cetus at Argonne? Also, are you talking
> about the Intel compilers or the IBM XL compilers? I'm not familiar with
> any Intel XL compilers.
>
> Assuming that you're talking about the IBM XL compilers on a BG/Q, I know
> that there were some issues with that but they may have been resolved in
> the ParaView superbuild (http://www.paraview.org/Wiki/ParaView/Superbuild),
> which includes Catalyst. If you try it out and have issues though, please
> email the mailing list so that we can try and fix them.
>
> FYI: ParaView has also been built with the Intel compilers on many
> platforms but I'm not sure about a BG/Q.
>
> Regards,
> Andy
>
> On Mon, Oct 19, 2015 at 3:54 PM, Jun Fang  wrote:
>
>> Hello,
>>
>> Here I have some questions about the in-situ library support. Does anyone
>> know if the ParaView catalyst is now supported by Intel XL compilers or
>> not? Based on my recent scaling test on INL Cetus, the XL compilers are
>> almost twice efficient compared to the GNU compilers for the identical
>> source codes, compilation flags (-O2), simulation case set-up (16,385
>> cores) and input parameters. Therefore, it becomes a tough decision for me
>> to choose either "GNU+in-situ capability" or "XL without in-situ". The
>> perfect situation is that XL compilers can also support insitu
>> visualization (i.e. ParaView catalyst), but I am not sure about that.
>>
>> Thanks a lot,
>> Jun
>>
>> --
>> Jun Fang
>> Department of Nuclear Engineering
>> North Carolina State University
>> Phone: (919) 592-2233
>>
>> ___
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ParaView Wiki at:
>> http://paraview.org/Wiki/ParaView
>>
>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview
>>
>>
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview

Re: [Paraview] About ParaView catalyst

2015-10-19 Thread Jun Fang 
Hi Andy,

Thank you so much for the prompt reply and the corrections. Yes, you are
totally right, I made some stupid typos and mistakes. Cetus is a BG/Q HPC
machine at Argonne National Lab for small test cases (along with Mira for
production runs). The XL compiler I mentioned is from IBM, not Intel.

I will check the link you sent and see if I can resolve the compatibility
issue b/w XL compilers and catalyst.

Again, thank you so much.
Jun

On Mon, Oct 19, 2015 at 5:13 PM, Andy Bauer  wrote:

> Hi Jun,
>
> I'm a bit confused. Do you mean Cetus at Argonne? Also, are you talking
> about the Intel compilers or the IBM XL compilers? I'm not familiar with
> any Intel XL compilers.
>
> Assuming that you're talking about the IBM XL compilers on a BG/Q, I know
> that there were some issues with that but they may have been resolved in
> the ParaView superbuild (http://www.paraview.org/Wiki/ParaView/Superbuild),
> which includes Catalyst. If you try it out and have issues though, please
> email the mailing list so that we can try and fix them.
>
> FYI: ParaView has also been built with the Intel compilers on many
> platforms but I'm not sure about a BG/Q.
>
> Regards,
> Andy
>
> On Mon, Oct 19, 2015 at 3:54 PM, Jun Fang  wrote:
>
>> Hello,
>>
>> Here I have some questions about the in-situ library support. Does anyone
>> know if the ParaView catalyst is now supported by Intel XL compilers or
>> not? Based on my recent scaling test on INL Cetus, the XL compilers are
>> almost twice efficient compared to the GNU compilers for the identical
>> source codes, compilation flags (-O2), simulation case set-up (16,385
>> cores) and input parameters. Therefore, it becomes a tough decision for me
>> to choose either "GNU+in-situ capability" or "XL without in-situ". The
>> perfect situation is that XL compilers can also support insitu
>> visualization (i.e. ParaView catalyst), but I am not sure about that.
>>
>> Thanks a lot,
>> Jun
>>
>> --
>> Jun Fang
>> Department of Nuclear Engineering
>> North Carolina State University
>> Phone: (919) 592-2233
>>
>> ___
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ParaView Wiki at:
>> http://paraview.org/Wiki/ParaView
>>
>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview
>>
>>
>


-- 
Jun Fang
Department of Nuclear Engineering
North Carolina State University
Phone: (919) 592-2233
___
Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview

[Paraview] About ParaView catalyst

2015-10-19 Thread Jun Fang 
Hello,

Here I have some questions about the in-situ library support. Does anyone
know if the ParaView catalyst is now supported by Intel XL compilers or
not? Based on my recent scaling test on INL Cetus, the XL compilers are
almost twice efficient compared to the GNU compilers for the identical
source codes, compilation flags (-O2), simulation case set-up (16,385
cores) and input parameters. Therefore, it becomes a tough decision for me
to choose either "GNU+in-situ capability" or "XL without in-situ". The
perfect situation is that XL compilers can also support insitu
visualization (i.e. ParaView catalyst), but I am not sure about that.

Thanks a lot,
Jun

-- 
Jun Fang
Department of Nuclear Engineering
North Carolina State University
Phone: (919) 592-2233
___
Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview