Search our client documentation below, optionally filtered by one or more systems.
Supercomputer:
Search Documentation
Search Documentation
Ascend
The following are technical specifications for Quad GPU nodes.
- Number of Nodes
-
24 nodes
- Number of CPU Sockets
-
48 (2 sockets/node)
- Number of CPU Cores
-
2,304 (96 cores/node)
- Cores Per Node
-
96 cores/node (88 usable cores/node)
- Internal Storage
-
12.8 TB NVMe internal storage
Ascend
We use Slurm syntax for all the discussions on this page. Please check how to prepare slurm job script if your script is prepared in PBS syntax.
Memory limit
Ascend
Compilers
C, C++ and Fortran are supported on the Ascend cluster. Intel, oneAPI, GNU Compiler Collectio (GCC) and AOCC are available. The Intel development tool chain is loaded by default. Compiler commands and recommended options for serial programs are listed in the table below. See also our compilation guide.
Ascend
The Next Gen Ascend (hereafter referred to as “Ascend”) cluster is now running on Red Hat Enterprise Linux (RHEL) 9, introducing several software-related changes compared to the RHEL 7/8 environment used on the Pitzer and original Ascend cluster. These updates provide access to modern tools and libraries but may also require adjustments to your workflows.
Ascend
- These are the public key fingerprints for Ascend:
ascend: ssh_host_rsa_key.pub = 2f:ad:ee:99:5a:f4:7f:0d:58:8f:d1:70:9d:e4:f4:16
ascend: ssh_host_ed25519_key.pub = 6b:0e:f1:fb:10:da:8c:0b:36:12:04:57:2b:2c:2b:4d
ascend: ssh_host_ecdsa_key.pub = f4:6f:b5:d2:fa:96:02:73:9a:40:5e:cf:ad:6d:19:e5
Ascend
The Ascend cluster, which includes 24 Quad GPU nodes deployed in 2022 and 250 Dual GPU nodes deployed this spring, is expected to be made available to all OSC clients on March 31, 2025, tentatively.
Who is eligible to participate in the Early User Program?
Ascend, Cardinal, Pitzer
AlphaFold 3 developed by DeepMind and Isomorphic Labs, is an advanced artificial intelligence system that predicts the 3D structures of proteins and their interactions with other molecules, including DNA, RNA, ligands, and ions.
Owens
After eight years of service, the Owens high performance computing (HPC) cluster will be decommissioned over the next two months. Clients currently using Owens for research and classroom instruction must migrate jobs to other OSC clusters during this time.
Overview
Estimating GPU memory (VRAM) usage for training or running inference with large deep learning models is critical to both 1. requesting the appropriate resources for running your computation and 2. optimizing your job once it is setup. Out-of-memory (OOM) errors can be avoided by requesting appropriate resources and by better understanding memory usage during the job using memory profiling tools described here.
Ascend, Cardinal, Pitzer
This page outlines how to use the Jupyter interactive app on OnDemand.
Launching Jupyter App
