Delta User Documentation

12 racks in a machine room, with overhead power, with red to orange "DELTA" wrap on the rack doors.

Introduction

Delta is a dedicated, ACCESS-allocated resource designed by HPE and NCSA, delivering a highly capable GPU-focused compute environment for GPU and CPU workloads. In addition to offering a mix of standard and reduced precision GPU resources, Delta also offers GPU-dense nodes with both NVIDIA and AMD GPUs. Delta provides high performance node-local SSD scratch filesystems, as well as both standard Lustre and relaxed-POSIX (not yet implemented) parallel filesystems spanning the entire resource.

Envisioned to lead change across the dimensions of computing, storage, and usability, Delta employs standards-based, fully customized science gateways with per-domain interfaces provided alongside the traditional command-line batch system interface. Enabled science gateways seamlessly integrate Delta into the cyberinfrastructure ecosystem alongside other resources to enable complex, multi-system workflows. To advance applications, the Delta project team continues the applications-focused, outcomes-oriented approach to drive the transition to GPU-accelerated applications alongside gateway-based productivity advances. Integrated with the NSF’s Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS) ecosystems, Delta leverages the substantial portfolio of services and support offered therein and together deliver unprecedented advances in researcher productivity.

Delta supports the ACCESS core software stack, including remote login, remote computation, data movement, science workflow support, and science gateway support toolkits.

Status Updates and Notices

Getting Started with High-Performance Computing

There are no specific prerequisite courses or high-performance computing (HPC) experience required before using Delta. However, if you are unfamiliar with using an HPC cluster, it is highly recommended that you take NCSA’s short tutorial Using an HPC Cluster for Scientific Applications before continuing.

Browse NCSA’s HPC-Moodle for a full list of HPC training opportunities including self-paced tutorials and training events (in-person and virtual).