System status and downtime

UNM - IT Alerts

View alerts from UNM Information Technology

CARC - System Status

Cluster Login and Website Status

Hopper (Ganglia)

Xena (Ganglia)

Wheeler (Ganglia)

Taos (Ganglia)

CARC - Network Performance

View current network performance

Scheduled Maintenance/Downtime 

 None at this time

 

Updates

Preparing for Xena Retirement on February 28th

(as of Jan 23, 2026)

 

We would like to share three important updates from CARC:
1) Easley compute nodes are entering production
2) the Xena cluster will be retired on February 28th, and
3) we have introduced a cost model for non–Main Campus PIs to support moving work from Xena to Hopper and Easley.

We recognise that these changes may have a significant impact, particularly for long time Xena users, and we want to ensure that support is available throughout this transition.

Xena Retirement

The Xena cluster will be retired on February 28, 2026.

After nearly a decade of service, Xena has supported a wide range of research and teaching activities across UNM. We are grateful to the many users who have relied on Xena over the years and understand that its retirement represents a meaningful change for some workflows. CARC users should transition to the newer Hopper and Easley clusters.

Please ensure that all data you wish to retain is removed from Xena’s local scratch filesystem before the retirement date:

  • Xena scratch: /xena/scratch

The following filesystems will not be affected:

  • CARC scratch: /carc/scratch
  • Home directories: ~ or /users/

If you would like assistance planning your transition from Xena, please attend CARC office hours. These sessions are a good opportunity to discuss data migration, workflow changes, access options, or concerns specific to your project:

Easley Now in Production for Compute Workloads

The Easley cluster has completed its open beta period and is now production ready for compute workloads. We sincerely thank all users who participated in testing and provided feedback during the beta phase.

  • Login: ssh easley.alliance.unm.edu

Easley scratch storage is expected to enter production in approximately six weeks.

Compared to Xena, Easley offers substantially newer capabilities, including:

  • Modern Intel Xeon Gold CPUs
  • NVIDIA H100 (80GB VRAM) and L40s (40GB VRAM) GPUs (replacing Xena’s K40 12GB VRAM GPUs)
  • Large memory nodes with 2 TB RAM
  • 200 Gb/sec InfiniBand interconnect compared with Xena’s 56 Gb/sec network.

Access for Non-Main Campus PIs

For CARC PIs whose primary affiliation is not main campus, a cost model is now in place to enable access to the Easley and Hopper systems.

For more information, please contact help@carc.unm.edu or attend CARC office hours.

Data Transfer and Conda Environment Checks

A short tutorial video covering recommended methods for uploading and downloading data to and from CARC systems is available here:
https://youtu.be/2UphEzHOHGM

If you have used conda on Xena, you may wish to check whether any environments were installed under /xena/scratch and need to be migrated. You can list your conda environments and their locations by running:

conda info --envs

We understand that transitions like this can be disruptive, and the CARC team is committed to helping users move forward smoothly. Ultimately these changes place far more computational power into the hands of our users along with additional storage capacity.