Skip to end of metadata
Go to start of metadata



NYU Research Technology Services (RTS) supports and encourages a model of hosting and managing clusters for research groups or departments in return for making unused cluster cycles available to the general NYU research community. These research groups and departments are our HPC Stakeholders, for whom NYU HPC manages hardware and provides priority access. Our current stakeholders are CGSBCNSCDSKussell Lab and CAOS.

If you are interested in becoming a stakeholder, please contact us at for details before you purchase your cluster. We can discuss your needs and work with you in the planning and purchase of hardware.

HPC Hosting and Equipment Life Cycle Policy

NYU IT data centers are secure, state-of-the-art facilities with 24/7 monitoring and redundant AC and power. These data centers are used for hosting ITS-owned, department-owned, and jointly-owned research computing clusters and storage. All HPC equipment in ITS data centers are taken out of service after 4 years. If used/refurbished equipment is put into the data center, this is measured from the original manufacture date rather than the date it was put into the data center.

Allocation of space and other scarce resources in ITS facilities is determined by the Research Faculty Advisory Group. Effective Summer 2013 there are no longer co-location fees for research clusters hosted in these facilities for individual researchers or departments. All requests to qualify for this service and/or to extend the management life of a cluster beyond 4 years (for a maximum of 1 additional year) must be approved by the Research Faculty Advisory Group.  Requests should be emailed to to submit to the Research Faculty Advisory Group and for requests to extend beyond 4 years up to 1 additional year should include the following:

  • Reason for extension request
  • Plans for replacement or retirement of cluster
  • Length of time of extension being requested

HPC Stakeholders Meetings

Stakeholders meet with HPC staff quarterly to discuss status updates, hardware upgrades, new services and provide feedback about the services they receive. Planned stakeholders meetings are announced in the front page of the HPC wiki.




  • No labels