Hi Matt, I was able to get my MPI problem fixed by using subcommunicators. I modified enable_parallelism() to take an optional argument that specifies the MPI communicator yt should use. If none is specified, it defaults to COMM_WORLD. Seems to be working pretty well so far. Thanks for the advice, Dan 2014-12-03 19:16 GMT-05:00 Matthew Turk :

Hi Dan,

I've done this before by modifying how the global communicator is created. The way they work in yt is through pushing new communicators onto a stack, which can vary based on which processors you're on. I don't think I ever wrote it up in a PR, but I think it could be done without too much work.

...

Is there a way to restrict yt to only use a subset of the available processes when it's running in parallel? I have a set of ipython engines running, and I'm trying to launch a yt script on a subset of them, but it just hangs. It works fine if I launch the same script on all the

On Wed, Dec 3, 2014 at 11:47 AM, Daniel Fenn wrote: processes.

...

Printing out MPI.COMM_WORLD.Get_size() from the script gives the total number of engines, not the size of the subset I'm trying to use.

I haven't done a lot of parallel programming, but I would guess that the running processes are waiting on one that's not running a yt script,

since

...

they can see the total number of processes available, but not the size of the subset of processes that are actually running a yt script.

Does anyone have any thoughts on how to do this?

Thanks,

Dan

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

Hi Cameron, Yeah, no problem. I'm not exactly sure how to do that. Do I have to fork yt first, or can I use my existing installation since it's already using Mercurial? I'm familiar with subversion, since that's what we use in our research group, but I'm a bit hazy on how the whole thing is set up with yt. When I installed yt using the installation script, it created a yt-hg directory. Is that a checked-out version of the code that I can commit? Thanks! Dan 2014-12-03 20:59 GMT-05:00 Cameron Hummels :

Hey Daniel,

This is great that you got this working, and it sounds like something from which others could benefit! Would you mind submitting this change to the main codebase in a pull request?

Cameron

On Wed, Dec 3, 2014 at 6:46 PM, Daniel Fenn wrote:

...

Hi Matt,

I was able to get my MPI problem fixed by using subcommunicators. I modified enable_parallelism() to take an optional argument that specifies the MPI communicator yt should use. If none is specified, it defaults to COMM_WORLD. Seems to be working pretty well so far.

Thanks for the advice,

Dan

2014-12-03 19:16 GMT-05:00 Matthew Turk :

Hi Dan,

...

I've done this before by modifying how the global communicator is created. The way they work in yt is through pushing new communicators onto a stack, which can vary based on which processors you're on. I don't think I ever wrote it up in a PR, but I think it could be done without too much work.

...

Is there a way to restrict yt to only use a subset of the available processes when it's running in parallel? I have a set of ipython engines running, and I'm trying to launch a yt script on a subset of them, but it just hangs. It works fine if I launch the same script on all the

On Wed, Dec 3, 2014 at 11:47 AM, Daniel Fenn wrote: processes.

...

Printing out MPI.COMM_WORLD.Get_size() from the script gives the total number of engines, not the size of the subset I'm trying to use.

I haven't done a lot of parallel programming, but I would guess that

the

...

running processes are waiting on one that's not running a yt script, since they can see the total number of processes available, but not the size of the subset of processes that are actually running a yt script.

Does anyone have any thoughts on how to do this?

Thanks,

Dan

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

There is also some discussion on how to do it in the developer documentation: http://yt-project.org/docs/dev/developing/developing.html#making-and-sharing... Thanks for your hard work on this, and don't hesitate to ask for more help! Cameron On Thu, Dec 4, 2014 at 8:05 AM, Matthew Turk wrote:

Hi Dan,

Yup, you can commit in that directory, and push to a fork. Create a fork on bitbucket, then do your changes and commit, and issue this command:

hg push https://bitbucket.org/yourusername/yt

Then you can issue a pull request via their web interface. You can also edit your hgrc to add an alias, but that can happen later.

-Matt

On Thu, Dec 4, 2014 at 8:08 AM, Daniel Fenn wrote:

...

Hi Cameron,

Yeah, no problem. I'm not exactly sure how to do that. Do I have to fork yt first, or can I use my existing installation since it's already using Mercurial? I'm familiar with subversion, since that's what we use in our research group, but I'm a bit hazy on how the whole thing is set up with yt. When I installed yt using the installation script, it created a yt-hg directory. Is that a checked-out version of the code that I can commit?

Thanks!

Dan

2014-12-03 20:59 GMT-05:00 Cameron Hummels :

...

Hey Daniel,

This is great that you got this working, and it sounds like something from which others could benefit! Would you mind submitting this change to the main codebase in a pull request?

Cameron

On Wed, Dec 3, 2014 at 6:46 PM, Daniel Fenn wrote:

...

Hi Matt,

I was able to get my MPI problem fixed by using subcommunicators. I modified enable_parallelism() to take an optional argument that

specifies

...

the MPI communicator yt should use. If none is specified, it defaults to COMM_WORLD. Seems to be working pretty well so far.

Thanks for the advice,

Dan

2014-12-03 19:16 GMT-05:00 Matthew Turk :

...

Hi Dan,

I've done this before by modifying how the global communicator is created. The way they work in yt is through pushing new communicators onto a stack, which can vary based on which processors you're on. I don't think I ever wrote it up in a PR, but I think it could be done without too much work.

On Wed, Dec 3, 2014 at 11:47 AM, Daniel Fenn wrote:

...

Is there a way to restrict yt to only use a subset of the available processes when it's running in parallel? I have a set of ipython engines running, and I'm trying to launch a yt script on a subset of them, but it just hangs. It works fine if I launch the same script on all the processes.

Printing out MPI.COMM_WORLD.Get_size() from the script gives the total number of engines, not the size of the subset I'm trying to use.

I haven't done a lot of parallel programming, but I would guess that the running processes are waiting on one that's not running a yt script, since they can see the total number of processes available, but not the size of the subset of processes that are actually running a yt script.

Does anyone have any thoughts on how to do this?

Thanks,

Dan

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org

On Thu Dec 04 2014 at 4:23:06 PM Daniel Fenn wrote:

Ok, thanks for the help--that ended up being pretty easy. Something else I'm confused about is why the same files appear in two different directories. There's the hg-yt directory in YT_DIR/src,

The yt-hg directory is a clone of the yt mercurial repository. It gets created by the install script.

and then there's YT_DIR/yt, which seems to be the exact same thing. What's the difference between them?

I don't have this directory in my install script build. Is there a chance you created this directory at some point?

Another question--will the fork that I created be updated with future changes to the code, or do I have to manually update it, and, if so, how?

You will need to manually keep it up to date. You can do this by clicking the "sync now" link on the overview page for your fork on the bitbucket web interface, or by manually pushing changes to your fork using "hg push". You can configure mercurial so that your default push location is your fork, see "hg help paths". That said, I don't think it's important to keep your fork up to date. I generally have my paths setup so I pull directly from the main yt_analysis/yt repository and then push to my fork only when I want to issue pull requests. If your fork is out of date, you when you push the changes for your pull request, you will also push the changes that will bring it up to date.

Thanks for your help,

Dan

2014-12-04 11:23 GMT-05:00 Cameron Hummels :

There is also some discussion on how to do it in the developer

...

documentation:

http://yt-project.org/docs/dev/developing/developing.html#making-and-sharing...

Thanks for your hard work on this, and don't hesitate to ask for more help!

Cameron

On Thu, Dec 4, 2014 at 8:05 AM, Matthew Turk wrote:

...

Hi Dan,

Yup, you can commit in that directory, and push to a fork. Create a fork on bitbucket, then do your changes and commit, and issue this command:

hg push https://bitbucket.org/yourusername/yt

Then you can issue a pull request via their web interface. You can also edit your hgrc to add an alias, but that can happen later.

-Matt

...

Hi Cameron,

Yeah, no problem. I'm not exactly sure how to do that. Do I have to fork yt first, or can I use my existing installation since it's already using Mercurial? I'm familiar with subversion, since that's what we use in our research group, but I'm a bit hazy on how the whole thing is set up with yt. When I installed yt using the installation script, it created a yt-hg directory. Is that a checked-out version of the code that I can commit?

Thanks!

Dan

2014-12-03 20:59 GMT-05:00 Cameron Hummels :

...

Hey Daniel,

This is great that you got this working, and it sounds like something from which others could benefit! Would you mind submitting this change to

...

...

main codebase in a pull request?

Cameron

On Wed, Dec 3, 2014 at 6:46 PM, Daniel Fenn wrote:

...

Hi Matt,

I was able to get my MPI problem fixed by using subcommunicators. I modified enable_parallelism() to take an optional argument that

specifies

...

the MPI communicator yt should use. If none is specified, it defaults to COMM_WORLD. Seems to be working pretty well so far.

Thanks for the advice,

Dan

2014-12-03 19:16 GMT-05:00 Matthew Turk :

> Hi Dan, > > I've done this before by modifying how the global communicator is > created. The way they work in yt is through pushing new communicators > onto a stack, which can vary based on which processors you're on. I > don't think I ever wrote it up in a PR, but I think it could be done > without too much work. > > On Wed, Dec 3, 2014 at 11:47 AM, Daniel Fenn wrote: > > Is there a way to restrict yt to only use a subset of the available > > processes when it's running in parallel? I have a set of ipython > > engines > > running, and I'm trying to launch a yt script on a subset of

...

...

...

> > it > > just hangs. It works fine if I launch the same script on all the > > processes. > > > > Printing out MPI.COMM_WORLD.Get_size() from the script gives the total > > number of engines, not the size of the subset I'm trying to use. > > > > I haven't done a lot of parallel programming, but I would guess

...

...

...

> > the > > running processes are waiting on one that's not running a yt

On Thu, Dec 4, 2014 at 8:08 AM, Daniel Fenn wrote: the them, but that script,

...

...

...

> > since > > they can see the total number of processes available, but not the size > > of > > the subset of processes that are actually running a yt script. > > > > Does anyone have any thoughts on how to do this? > > > > Thanks, > > > > Dan > > > > _______________________________________________ > > yt-users mailing list > > yt-users@lists.spacepope.org > > http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org > > > _______________________________________________ > yt-users mailing list > yt-users@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

On Thu Dec 04 2014 at 5:08:18 PM Daniel Fenn wrote:

There's a good chance I created that directory at some point. That cuts down on my confusion a good deal. One final question--I'm trying to test out some of the recent changes that were made--specifically pull requests #1328 and #1331. I thought I would just have to do hg update in the hg-yt directory.

The update worked fine, and it now says I'm at version 3.1-dev, but when I check the log, I don't see any mention of those pull requests. Did I miss them (very possible), or have they not been added to the main repository? I notice that they both still say "open". If that's the case, is there a way that I can download the changed files?

To test a PR, you're going to need to pull it in locally from the issuer's fork of yt, then update to the tip of their PR. I'm looking at our documentation, and I don't think we cover this specific workflow, so this is something that should probably be added. Here what I do. Go to the pull request page on bitbucket, and click over to the commits tab. I.e. this page for PR 1328: https://bitbucket.org/yt_analysis/yt/pull-request/1328/delay-running-callbac... That lists all of the changesets that are part of the pull request. You're going to need to copy/paste the changeset hash for the tip of the pull request, in this case it's c62d7d0. Next, in your local clone of yt, do: $ hg pull -r c62d7d0 https://bitbucket.org/ngoldbaum/yt Note that I'm pulling from the PR issuer's repository. Pull requests are not part of the main repository until they have been merged. Next, just in case any compiled code changes (i.e. a cython .pyx file), run: $ python setup.py develop It's good to get into the habit of doing this, since you will run into weird behavior if any compiled files changed and they didn't get recompiled. Now the changeset in the tip of the pull request is the "active" change. You can test the pull request. To do back to "mainline" yt, you can issue the following command: hg update Where is the changeset hash corresponding to the current tip of the "yt" branch on the main yt bitbucket repository. I usually look at the following page to figure out what that is: https://bitbucket.org/yt_analysis/yt/commits/branch/yt You can also automate this workflow using the remotebranches extension for mercurial: http://mercurial.selenic.com/wiki/RemoteBranchesExtension If at any time you get confused about where you are on the changeset graph, take a look at the output of "hg log -G". You can also use a gui like tortoisehg or sourcetree, if you find that more comfortable. Best, Nathan

Thanks for taking the time to answer all my questions.

Dan

2014-12-04 19:30 GMT-05:00 Nathan Goldbaum :

...

On Thu Dec 04 2014 at 4:23:06 PM Daniel Fenn wrote:

...

Ok, thanks for the help--that ended up being pretty easy. Something else I'm confused about is why the same files appear in two different directories. There's the hg-yt directory in YT_DIR/src,

The yt-hg directory is a clone of the yt mercurial repository. It gets created by the install script.

...

and then there's YT_DIR/yt, which seems to be the exact same thing. What's the difference between them?

I don't have this directory in my install script build. Is there a chance you created this directory at some point?

...

Another question--will the fork that I created be updated with future changes to the code, or do I have to manually update it, and, if so, how?

You will need to manually keep it up to date. You can do this by clicking the "sync now" link on the overview page for your fork on the bitbucket web interface, or by manually pushing changes to your fork using "hg push". You can configure mercurial so that your default push location is your fork, see "hg help paths".

That said, I don't think it's important to keep your fork up to date. I generally have my paths setup so I pull directly from the main yt_analysis/yt repository and then push to my fork only when I want to issue pull requests. If your fork is out of date, you when you push the changes for your pull request, you will also push the changes that will bring it up to date.

...

Thanks for your help,

Dan

2014-12-04 11:23 GMT-05:00 Cameron Hummels :

There is also some discussion on how to do it in the developer

...

documentation:

http://yt-project.org/docs/dev/developing/developing.html#making-and-sharing...

Thanks for your hard work on this, and don't hesitate to ask for more help!

Cameron

On Thu, Dec 4, 2014 at 8:05 AM, Matthew Turk wrote:

...

Hi Dan,

Yup, you can commit in that directory, and push to a fork. Create a fork on bitbucket, then do your changes and commit, and issue this command:

hg push https://bitbucket.org/yourusername/yt

Then you can issue a pull request via their web interface. You can also edit your hgrc to add an alias, but that can happen later.

-Matt

...

Hi Cameron,

Yeah, no problem. I'm not exactly sure how to do that. Do I have to fork yt first, or can I use my existing installation since it's already using Mercurial? I'm familiar with subversion, since that's what we use in our research group, but I'm a bit hazy on how the whole thing is set up with yt. When I installed yt using the installation script, it created a yt-hg directory. Is that a checked-out version of the code that I can commit?

Thanks!

Dan

2014-12-03 20:59 GMT-05:00 Cameron Hummels :

> Hey Daniel, > > This is great that you got this working, and it sounds like something from > which others could benefit! Would you mind submitting this change to the > main codebase in a pull request? > > Cameron > > On Wed, Dec 3, 2014 at 6:46 PM, Daniel Fenn wrote: >> >> Hi Matt, >> >> I was able to get my MPI problem fixed by using subcommunicators. I >> modified enable_parallelism() to take an optional argument that specifies >> the MPI communicator yt should use. If none is specified, it defaults to >> COMM_WORLD. Seems to be working pretty well so far. >> >> Thanks for the advice, >> >> Dan >> >> 2014-12-03 19:16 GMT-05:00 Matthew Turk : >> >>> Hi Dan, >>> >>> I've done this before by modifying how the global communicator is >>> created. The way they work in yt is through pushing new communicators >>> onto a stack, which can vary based on which processors you're on. I >>> don't think I ever wrote it up in a PR, but I think it could be done >>> without too much work. >>> >>> On Wed, Dec 3, 2014 at 11:47 AM, Daniel Fenn wrote: >>> > Is there a way to restrict yt to only use a subset of the available >>> > processes when it's running in parallel? I have a set of ipython >>> > engines >>> > running, and I'm trying to launch a yt script on a subset of

...

>>> > it >>> > just hangs. It works fine if I launch the same script on all the >>> > processes. >>> > >>> > Printing out MPI.COMM_WORLD.Get_size() from the script gives

...

>>> > number of engines, not the size of the subset I'm trying to use. >>> > >>> > I haven't done a lot of parallel programming, but I would guess

...

>>> > the >>> > running processes are waiting on one that's not running a yt

...

>>> > since >>> > they can see the total number of processes available, but not

On Thu, Dec 4, 2014 at 8:08 AM, Daniel Fenn wrote: them, but the total that script, the size

...

>>> > of >>> > the subset of processes that are actually running a yt script. >>> > >>> > Does anyone have any thoughts on how to do this? >>> > >>> > Thanks, >>> > >>> > Dan >>> > >>> > _______________________________________________ >>> > yt-users mailing list >>> > yt-users@lists.spacepope.org >>> > http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org >>> > >>> _______________________________________________ >>> yt-users mailing list >>> yt-users@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org >> >> >> >> _______________________________________________ >> yt-users mailing list >> yt-users@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org >> > > > > -- > Cameron Hummels > Postdoctoral Researcher > Steward Observatory > University of Arizona > http://chummels.org > > _______________________________________________ > yt-users mailing list > yt-users@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org >

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________

yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org