ipyparallel

ipyparallel copied to clipboard

HTCondor engines dont stop when using SSHProxyEngineSetLauncher

2

I'm using SSHProxyEngineSetLauncher to start an ipcluster on a remote machine which uses condor. When stopping SSHProxyEngineSetLauncher condor_q still shows ipengines in the queue. I think this is because SSHProxyEngineSetLauncher...

feanor12

job priority or task priority

1

I want to set job priority so that post-submitted high-priority jobs can be executed first.If a job is running and ipcontroller send a new higher priority job,the running job will...

CAOLINAN

The config file `ipcontroller-client.json` appears to be required to start ipengine, Once more the values in the config file appear to *override* the values provided via command line flags like...

elibixby

segfault in ipengine on linux with windows controller

1

I'm running controller and the client on a windows host. I start an engine on a linux machine (using `ipengine --file=ipcontroller-engine.json`) and the engine connects fine. If I then start...

cycomanic

Safely cancelling jobs

1

Hi, Is there a way to safely cancelling jobs without needing to know if they have started, running or finished? I do multiple simulations in parallel, and many of them...

jlintusaari

Hi, following jupyter confirmation-on-exit, it would be great if ipcontroller and ipcluster were asking for confirmation when hitting Ctrl-C. It would prevent the user from killing by mistake their controller...

cphyc

The current groupby argument always displays the result of all engines. It is useful sometimes to only display the result of a single engine -- when the execution / result...

rainwoodman

Warning of unhandled iopub "data_message" in hub.save_iopub_message()

1

engine.datapub.publish_data() sends a "data_message", but this isn't recognized in hub.save_iopub_message(). Is this expected behavior? It produces only a warning, but it leaves me wondering if there's been a name change...

rjplevin

More complex MPI split communicator layouts?

6

At the moment we are manually launching a cluster with `ipcontroller` listening on an Infiniband over TCP/IP interface, and then launching 16 `ipengine` using say `mpiexec -n 16 ipengine`. All...

jhale

Strategies for profiling functions in ipyparallel

1

I have a function that is going slow and I would like to figure out what is holding it up. I have profiled the underlying non-parallel function and it goes...

jakirkham