[MITgcm-devel] [MITgcm-support] asyncio

[Dimitris Menemenlis]

Hi Martin, the asyncio code is not ready for main trunk yet.
As it stands, it requires some manual adjustments and hard-coding
(http://mitgcm.org/viewvc/MITgcm/MITgcm_contrib/llc_hires/llc_4320/code-async/readme.txt),
it is not yet “MITgcm-pkg-ed”, and I think that it does not have a very elegant ee-death.

If there is a lot of demand for this capability, and if Jean-Michel approves,
I can put this on todo list but it would not happen any time soon.

My preference would be to wait for one or two more folks to take the
MITgcm_contrib asyncio code for a spin, demonstrating need and usefulness,
before we add to main branch.

Dimitris Menemenlis

P.S.:  I did not write this code.  (I am not sufficiently good programmer for this.)
The code was contributed by Bron Nelson and Chris Henze at NASA Ames,
with lots of help from Chris Hill.

> On Oct 18, 2016, at 10:03 PM, Martin Losch <Martin.Losch at awi.de> wrote:
> 
> Hi Dimitris,
> changing to devel-list:
> 
> does it make sense to add this to the main trunk? We all are and will be running large scale setups with lots of I/O (for me it’s not a bottle-neck, but I start to feel it), so this asynchronous I/O would be great. How much work is it and how specific is it to the file system that you are using?
> 
> Martin