A number of DLCI groups operate dedicated computational resources for their own communities, and commercial options are also available.
TIG Shared Computing
TIG shared computing, operated by the CSAIL infrastructure group, consists of an Openstack cluster and a Slurm cluster for general use by members of CSAIL. The Openstack environment supports full virtual machines. The Slurm cluster supports Singularity as a container engine for Docker containers. Additional compute and storage resources can be purchased by PIs to support group specific needs.
Further information and support is available at help@csail.mit.edu.
LNS Computing
The Laboratory for Nuclear Science in Physics operates computing resources that are available to researchers within LNS.
Further information and support is available from pra@mit.edu.
Kavli Computing
The MIT Kavli Institute operates a cluster for astrophysics research. The cluster uses the Slurm resource scheduler and is available for use by Kavli researchers.
Koch Bioinformatics
Koch operates a bioinformatics facility which specializes in processing needs of computational biologists.
subMIT
Open to all members of the Physics Department, subMIT provides specialized physics computing support along with access to a login pool connected to a local SLURM cluster and HTCondor access to large-scale external resources, including the Open Science Grid, CMS Tier 2/3, and the Lattice QCD cluster. In addition to traditional CPU nodes, the internal SLURM cluster includes specialized hardware such as GPUs, high-density (hundreds of cores per node), high-memory CPU nodes, fast NVMe scratch space, and a 100 Gbit/s network. Further information is available at https://submit.mit.edu/ or submit-help@mit.edu.
SuperCloud
The SuperCloud system is a collaboration with MIT Lincoln Laboratory on a shared facility that is optimized for streamlining open research collaborations with Lincoln Laboratory (e.g., AIA, BW, CQE, Haystack, HPEC, ISN). The facility prioritizes access for Lincoln Laboratory collaborators. Further information and support is available at supercloud@mit.edu.
Features:
- Compute power: The latest SuperCloud system has more than 16,000 x86 CPU cores and more than 850 NVidia Volta GPUs in total.
- Hardware access: Hardware access is through the Slurm resource scheduler that supports batch and interactive workload and allows dedicated reservations.
- Portal: A custom, web-based portal supporting Jupyter notebooks is available.
- Software: A wide range of standard software is available and the Docker compatible Singularity container tool is supported.
Commercial Cloud Based Resources
Many commercial cloud-based resources are available, and some are commonly used at MIT.
- MIT Cost Object Cloud Accounts: Provides a central location for creating standard AWS, Google or Microsoft commercial cloud provider accounts that are tied to cost objects and projects. Enables tracking of expenses for different projects.
- Google Colab: Provides a free base service that can be very useful for modest workloads, for example in classes.
- Binder: Provides free virtual machines that can be flexibly configured by providing a Github repository with software setup instructions.
- Code Ocean: Highly customizable cloud based virtual machine system. Machines in Code Ocean are direct charged and performance is more predictable than free services.
The ORCD team is also experimenting with a simplified AWS system called RONIN, designed to support university research and teaching needs. If you are interested in testing RONIN, contact ORCD.
Major Cloud Provider Credits Program
Most cloud providers provide useful compute and data credits programs for educational and research needs.
- AWS Programs for Research and Education: Provides cloud credit grants for both research and education projects.
- Open Data on AWS: Hosts open data for sharing with others. AWS has a process for applying to have datasets considered for hosting.
- Azure student development resources: Provides Azure cloud credits for education.
- Azure Open Datasets: Hosts standard datasets for general use including machine learning. Additional datasets for inclusion can be nominated.
- Google Cloud for Students, Faculty, and Researchers: Offers free credits and technical resources for education and research.