Resource Documentation

ACES is a composable HPC testbed at Texas A&M offering dynamically reconfigurable GPU, FPGA, and AI processor configurations for data-intensive and AI-driven scientific research.

Anvil is a large-scale computing resource operated by Purdue University, offering a balanced mix of CPU, GPU, and AI-optimized nodes to support a wide range of research workloads.

Bridges-2 is a converged HPC, AI, and data platform at Pittsburgh Supercomputing Center, offering CPU, GPU, large-memory, and storage resources to support diverse scientific workloads.

CloudBank provides researchers and educators flexible access to commercial cloud platforms — including AWS, Google Cloud, Azure, and IBM Cloud — for both research computing and classroom use.

Delta offers a highly capable GPU-focused computing environment designed to support both GPU and CPU workloads.

NCSA's AI/ML-focused computing resource. Built for workloads that need large-memory GPUs and the ability to scale from one GPU to hundreds

Derecho is NCAR's large-scale HPC system combining high-core-count CPU nodes for parallel scientific computing with dedicated GPU nodes for AI, deep learning, and GPU-accelerated simulation.

Expanse is a flexible supercomputer at SDSC offering CPU and GPU compute alongside scalable storage and composable environments with integrated cloud provider access for diverse research workloads.

FABRIC is an NSF-funded, nationwide research infrastructure for at-scale experimentation in networking, cybersecurity, distributed systems, storage, machine learning, and science applications.

Granite is NCSA's long-term tape archive system providing high-capacity replicated storage for preserving large research datasets.

Jetstream2 is a cloud-based research platform offering on-demand virtual machines with CPU, GPU, and large-memory configurations for interactive computing and persistent research environments.

KyRIC is a large-memory computing resource at the University of Kentucky for data-intensive workloads that run on a single node, such as genome sequencing, processing of large text corpora, and big

Launch is a general-purpose computing cluster at Texas A&M offering CPU and GPU nodes supporting a range of research applications and scientific tools.

Neocortex is PSC's specialized AI resource, pairing Cerebras wafer-scale systems with a large shared-memory HPE Superdome Flex server.

The Open Science Grid aggregates distributed computing capacity from campuses and national labs nationwide into a single virtual cluster for high-throughput open science workloads.

The Open Storage Network is an NSF-funded, geographically distributed cloud storage resource providing S3-compatible object storage for research datasets, with allocations up to 50TB.

Ranch is TACC's long-term archival storage system combining a high-performance filesystem with tape-based backing store for preserving large scientific datasets generated on HPC systems.

REPACSS is a power-aware HPC and AI infrastructure resource operated by Texas Tech University, supporting large-scale simulations, AI training, and data analytics.

Sage is a national AI-at-the-edge testbed for deploying AI/ML applications on sensor-equipped nodes that process real-time data directly where it is collected.

Stampede3 is TACC's national supercomputer offering a diverse mix of CPU, GPU, and high-bandwidth memory nodes to support a broad range of open science workloads.

Voyager is SDSC's AI-focused computing system purpose-built for science and engineering research, supporting large-scale deep learning and AI-driven experimental and computational workflows.