# ACCESS-rAM3 'Flagship' Experiments

**URL:** <https://forum.access-hive.org.au/t/access-ram3-flagship-experiments/4580>\
**Category:** rAM3/Regional Nesting Suite\
**Created:** [5 June 2025 01:40 UTC](https://forum.access-hive.org.au/t/access-ram3-flagship-experiments/4580 "2025-06-05T01:40:27Z")\
**Posts on this page:** 1\
**Showing post:** 15

<div class="post-metadata">

**Author:** ![Paul.Gregory](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.access-hive.org.au/paul.gregory/32/2566_2.png) [@Paul.Gregory](https://forum.access-hive.org.au/u/Paul.Gregory)\
**Post date:** [22 July 2025 08:07 UTC](https://forum.access-hive.org.au/t/access-ram3-flagship-experiments/4580/15 "2025-07-22T08:07:13Z")

</div>

An update. I managed to get the I/O server running. It appears there is a conflict with Fortran unit numbers when running the I/O server with netCDF output enabled.

@cbengel provided a branch of rAM3 which used Fields Files outputs, which removed the conflict between Fortran unit numbers.

See here for further details: [Issues with configuring the UM I/O server - #6 by Paul.Gregory](https://forum.access-hive.org.au/t/issues-with-configuring-the-um-i-o-server/4653/6)

The default ROMIO\_HINTS settings also provided a useful speedup.

I compiled a UM executable for the Sapphirerapids nodes on the `normalsr` queue, see : [Building the UM to run on Gadi's Sapphire nodes - #4 by Paul.Gregory](https://forum.access-hive.org.au/t/building-the-um-to-run-on-gadis-sapphire-nodes/4698/4)

That also provided a useful speedup, but I’m still a fair way behind AUS2200 in terms of kSU/day using the same domain decomposition. While the AUS2200 configuration is similar to rAM3, it’s not exact, so I want to run AUS2200 for a day or two and examine the I/O server logs to compare against the rAM3 setup.

---

_[View the full topic](https://forum.access-hive.org.au/t/access-ram3-flagship-experiments/4580)._