The Faculty Owned Equipment (FOE) service supports researchers with grants that require the purchase of computing hardware. FOE allows researchers to place their own hardware within an HPC cluster.  Most FOE hardware is hosted in the cluster Lighthouse, but may be hosted in other clusters.

To use FOE, we require researchers:

  1. coordinate with ARC-TS prior to including hardware in a proposal or purchasing hardware;
  2. purchase hardware that is compatible with the rest of the cluster installation where the hardware will be running;
  3. obtain a subscription to FOE for each compute node.

A subscription to FOE provides access to the HPC infrastructure  — data center, staff, networking, storage and basic software — except the compute hardware.

FOE is available for:

  1. Researchers who receive funds from external funding agencies for purchasing equipment.
  2. Researchers who have workflow and/or infrastructure needs that differ significantly from those that can be met by other HPC services (i.e. Standard-Memory, Larger-Memory or GPU Flux allocations).

The hardware added to Lighthouse is for the exclusive use of the research group that added it and will not be part of the HPC node pool.

Budgeting for the Faculty Owned Equipment

The FOE subscription rate and node purchase charges allow for funding from a range of sources: federally funded research projects; general funds; departmental instructional funds; faculty discretionary and research incentive accounts; cost-sharing funds; and faculty start-up or retention package funds. All FOE purchase and subscription costs are charged to a U-M shortcode; no other payment method (cash, credit cards, etc.) is accepted.  The costs are determined by the FOE rate that applies to the purchasing researcher and the number and price of the compute nodes being added.  FOE use is billed monthly by ITS. The details of the FOE charges will appear on the normal monthly Statement of Activity.

Hardware purchases for FOE are a one-time expense.  If made from federal funds, the timing of the purchase relative to the end of the grant period must be consistent with federal regulations.  FOE subscription charges are a recurring expense.  If made from federal funds, subscription charges can not be prepaid to extend beyond the end of the grant period.

Order Service

The rate for a subscription to FOE is $113 per node per month.

A “no-software” (without commercial software) version of FOE is available for $97 per node per month on the Lighthouse HPC cluster.

As long as the subscription charges for FOE are paid, compute nodes will be maintained in the Flux Operating Environment until they are 4 years old. Keeping nodes older than 4 years in the Flux Operating Environment will be at the discretion of Advanced Research Computing – Technology Services (ARC-TS) and will not be run beyond 6 years.

The subscription rate has been set by ARC and approved by the U-M’s Office of Financial Analysis (OFA) and may be charged to federal grants. The hardware purchase price will be finalized at the time of purchase and will be processed by U-M purchasing.

To order Service:

Email arcts-support@umich.edu

Related Event

March 2 @ 11:00 am - March 3 @ 5:00 pm

XSEDE HPC Workshop: GPU Programming Using OpenAcc

OVERVIEW XSEDE, along with the Pittsburgh Supercomputing Center is pleased to present an OpenACC GPU programming workshop. OpenACC is the accepted standard using compiler directives to allow quick development of…

March 5 @ 3:30 pm - 5:00 pm

GIS Fundamentals – V (Spatial Database – PostGIS)

This is the fifth workshop in a series of workshops we are offering this semester on the fundamentals of GIS. Each workshop covers one or two key elements of GIS…

March 10 @ 1:00 pm - 3:00 pm

Software Development For Research: Version Control Principles

Software development and computer programming is increasingly a major part of scientific research. Projects can quickly grow, and it can be difficult to manage changes, especially if multiple programmers are…

March 16 @ 10:00 am - 12:00 pm

Introduction to Deep Neural Networks with Keras/TensorFlow

Deep Neural Networks (DNNs) are used as a machine learning method for both regression and classification problems. Keras is a high-level, Python interface running on top of multiple neural network…