Skip to content

Overview of RCC's HPC Systems

The University of Chicago Research Computing Center (RCC) hosts a diverse range of professionally managed high-performance computing clusters, forming the backbone of the RCC's advanced computational infrastructure. Notable clusters include Midway2, Midway3, Midway3-AMD, MidwaySSD, DaLI, Beagle3, KICP, GM4, MidwayR family, and Skyway.

Midway 3 ecosystem

Shared computing partitions

Midway3, and Midway3-AMD

In 2021, Midway3 and Midway3-AMD were introduced as flagship clusters for multi-purpose scientific computing. It is recommended that new users opt for Midway3 due to its latest hardware and software modules.

Restricted computing partitions


Funded by an NIH grant, Beagle3, part of Midway3 ecosystems, supports biomedical research with cutting-edge HPC resources, specializing in modeling and large-scale simulations of molecular structures. Launched in February 2022, Beagle3 features 44 compute nodes with Intel CPUs and NVIDIA GPUs.


MidwaySSD, part of the Midway3 ecosystem, serves computationally intensive Social Science Division research and educational purposes.


Skyway allows users to seamlessly burst computing workloads through Midway 2 and 3 ecosystems to remote commercial cloud platforms like Amazon Web Services (AWS) and Google Cloud: Cloud Computing Services (GCP). It simplifies cloud computing tasks without requiring users to manage cloud resources.

CPP nodes

Cluster Partnership Program (CPP) or private nodes.

SDE MidwayR ecosystems (R1, R2, and R3)

SDE MidwayR

MidwayR, within the Secure Data Enclave, provides a secure computing environment for research with high-security standards. It features a setup similar to Midway but tailored for secure data protection.

Midway 2 ecosystem

Shared computing partitions


Midway2, succeeding the original Midway in 2016, boasts advanced features.

Restricted computing partitions


DaLI, the Data Lifecycle Instrument, as a part of the Midway2 ecosystem, facilitates data management and sharing, offering a unified workflow for acquiring, transferring, processing, and storing experimental and observational data. It enhances data lifecycle management, sharing, and publication, supporting outreach and education.


GM4, a GPU-enabled cluster, is part of the Midway2 ecosystem and facilitates multiscale materials modeling and machine learning, supporting diverse molecular dynamics and continuum simulations. Funded by the NSF under the MRI program, GM4 promotes collaborative efforts in algorithm and software development.

CPP nodes

Cluster Partnership Program (CPP) or private nodes.