Delta GPU

2FA/MFA RP account needed

GPU nodes with NVIDIA A40, A100, or H200 cards, plus AMD MI100 nodes. Used for AI training and inference and other GPU-accelerated compute.

GET AN ACCOUNT ON DELTA GPU

NCSA Project Management

Jobs

Queue specifications

Metrics updated 2026-06-08

Name Purpose CPUs GPUs RAM Jobs
30 days
Wait Time
30-day trend
Wall Time
30-day trend
gpuA100x4 Default A100 queue with 4 GPUs per node for standard GPU jobs 64 4 x NVIDIA A100 per node 256GB 87,818
gpuA100x4 wait time: average 25.5 hours, range 0 to 82.3 hours over 30 days 25.5
gpuA100x4 wall time: average 1.5 hours, range 0.2 to 4.9 hours over 30 days 1.5
gpuA100x4-interactive Interactive A100 sessions for debugging and testing 64 4 x NVIDIA A100 per node 256GB 56,117
gpuA100x4-interactive wait time: average 6.9 hours, range 0.6 to 19.4 hours over 30 days 6.9
gpuA100x4-interactive wall time: average 0.5 hours, range 0.2 to 0.8 hours over 30 days 0.5
gpuA100x4-preempt Preemptible A100 jobs 64 4 x NVIDIA A100 per node 256GB 199
gpuA100x4-preempt wait time: average 10.8 hours, range 0 to 890.8 hours over 30 days 10.8
gpuA100x4-preempt wall time: average 0.2 hours, range 0 to 0.7 hours over 30 days 0.2
gpuA100x8 Large-scale multi-GPU workloads using 8 A100 GPUs 128 8× NVIDIA A100 ~2TB 4,359
gpuA100x8 wait time: average 9.3 hours, range 0.6 to 36.3 hours over 30 days 9.3
gpuA100x8 wall time: average 2.5 hours, range 0.2 to 8.3 hours over 30 days 2.5
gpuA100x8-interactive Interactive large-scale A100 jobs for development/testing 128 8× NVIDIA A100 ~2TB 562
gpuA100x8-interactive wait time: average 0.4 hours, range 0 to 3.8 hours over 30 days 0.4
gpuA100x8-interactive wall time: average 0.5 hours, range 0 to 1 hours over 30 days 0.5
gpuA40x4 Moderate GPU workloads 64 4× NVIDIA A40 256GB 58,034
gpuA40x4 wait time: average 63.0 hours, range 0.1 to 220.3 hours over 30 days 63.0
gpuA40x4 wall time: average 2.8 hours, range 0.2 to 6.1 hours over 30 days 2.8
gpuA40x4-interactive Interactive A40 usage for testing and development 64 4× NVIDIA A40 256GB 28,589
gpuA40x4-interactive wait time: average 3.9 hours, range 0.1 to 20.8 hours over 30 days 3.9
gpuA40x4-interactive wall time: average 0.3 hours, range 0.1 to 0.5 hours over 30 days 0.3
gpuA40x4-preempt Preemptible A40 jobs for flexible, lower-priority workloads 64 4× NVIDIA A40 256GB 7,493
gpuA40x4-preempt wait time: average 8.7 hours, range 0 to 29.2 hours over 30 days 8.7
gpuA40x4-preempt wall time: average 0.1 hours, range 0 to 1.3 hours over 30 days 0.1
gpuH200x8 High-end GPU queue for memory-intensive workloads 96 8× NVIDIA H200 ~2TB 5,731
gpuH200x8 wait time: average 6.8 hours, range 0.4 to 20.4 hours over 30 days 6.8
gpuH200x8 wall time: average 2.9 hours, range 0.1 to 4.7 hours over 30 days 2.9
gpuH200x8-interactive Interactive H200 sessions for development and tuning 96 8× NVIDIA H200 ~2TB 2,307
gpuH200x8-interactive wait time: average 0.3 hours, range 0 to 1.3 hours over 30 days 0.3
gpuH200x8-interactive wall time: average 0.3 hours, range 0.1 to 0.8 hours over 30 days 0.3
gpuMI100x8 AMD GPU workloads optimized for MI100 architecture 128 8× AMD MI100 (+1 MI210) ~2TB 241
gpuMI100x8 wait time: average 0.0 hours, range 0 to 0.4 hours over 30 days 0.0
gpuMI100x8 wall time: average 0.6 hours, range 0 to 12 hours over 30 days 0.6
gpuMI100x8-interactive Interactive AMD GPU sessions for development/testing 128 8× AMD MI100 (+1 MI210) ~2TB 27
gpuMI100x8-interactive wait time: average 0.0 hours, range 0 to 0.1 hours over 30 days 0.0
gpuMI100x8-interactive wall time: average 0.2 hours, range 0 to 0.6 hours over 30 days 0.2

Software

No software usage data is currently reported for Delta GPU in XDMoD.

SEE ALL SOFTWARE AVAILABLE ON DELTA GPU

Explore in SDS

EXPLORE SOFTWARE AVAILABLE ON DELTA

EXPLORE SOFTWARE

Datasets

Name Description
alphafold

Protein structure prediction datasets used for bioinformatics research with AlphaFold.

models(ollama)

Pretrained large language models used with Ollama for local inference and AI experimentation.


Storage

File System

Directory Path Quota Purge Backup Notes
HOME /u 100 GB 750,000 files per users. None 30 days Area for software, scripts, job files, and so on. Not intended as a source/destination for I/O during jobs.
PROJECTS /projects 500GB. Up to 25TB by request. None None Area for shared data for a project, common data sets, software, results, and so on.
WORK-HDD /work/hdd 1000GB. Up to 100TB by request. None None Area for computation, largest allocations, where I/O from jobs should occur. (this is now your scratch volume)
WORK-NVME /work/nvme NVME space is available upon request. None None Area for computation, NVME is best for lots of small I/O from jobs.
TMP /tmp 0.74 (CPU) or 1.5TB (GPU). Shared or dedicated depending on node usage by jobs. Purged after each job None Locally attached disk for fast small file I/O.

File Transfer

Supported Methods Data Transfer Node URL
SCP dt-login[01-04].delta.ncsa.illinois.edu
RSYNC dt-login[01-04].delta.ncsa.illinois.edu
GLOBUS | RECOMMENDED https://www.globus.org/globus-connect-personal