Are you looking for a collaborative computing environment that is easy to use and can handle data science projects with large RAM and GPUs ?

Texas Virtual Data Library (Texas ViDaL) computing servers are a new onsite free super computing infrastructure managed by TAMU HPRC and might be a good fit for your next data science project. If you have projects you want to put on ViDaL, please contact Dr. Hye-Chung Kum (kum at tamu dot edu) for more information.

What it offers

ViDaL 2.0 (Released Feb. 3, 2025)

• Both Windows and Linux OS are available

• Large memory nodes (3TB & 384GB)

• 128 processing cores

• GPU nodes with 2 NVIDIA H100 NVL GPUs

• securely maintained in the West Campus Data Center by the TAMU HPRC and meets compliance requirements approved by TAMU’s chief privacy and security officers

ViDaL 1.0 (no longer available)

• Both Windows and Linux OS are available

• Large memory nodes (1.5TB & 192GB)

• 2PB Fast Disk

• 120GB SSD scratch space (local disk)

• NVIDIA V100 GPUs

• securely maintained in the West Campus Data Center by the TAMU HPRC and meets compliance requirements approved by TAMU’s chief privacy and security officers

Who might benefit

• Undergraduates doing research in data science

• Data science projects with staff that are not well trained in programming and linux

• Research that would benefit from having a collaborative disk space to share data and programs

• Data science projects that do not fit well with traditional high performance computing environments

• Research that requires HIPAA compliance

• Research that involves proprietary data with legal agreement (e.g., licensed data)

Types of Compliant Computing Available

ViDaL 2.0 (released Feb 3, 2025)

• Large RAM nodes: 4 nodes with 3TB RAM each

• Regular RAM nodes: 10 nodes with 384GB RAM each

• GPU nodes: 4 GPU nodes with two NVIDIA H100 NVL 94GB GPUs

ViDaL 1.0 (no longer available)

• Large RAM nodes: 4 nodes with 1.5TB Ram each

• Regular RAM nodes: 16 nodes with 192GB Ram each

• GPU nodes: 4 GPU nodes, each with 192 GB Ram and two NVIDIA V100 GPUs

Please acknowledge ViDaL whenever appropriate

• “This work was supported in part by the Texas Virtual Data Library (ViDaL) funded by the Texas A&M University Research Development Fund”

• Please report any proposals or papers that benefit from vidal to us over email (admin@vidal.tamu.edu). This is important for tracking and reporting purposes for future funding to keep vidal affordable to us.