Cluster Computing Fee Structure & Consulting Requests
Cluster Computing Fee Structure (Tier Levels)
Access to shared high performance computing (HPC) at the University is organized into three tiers of use:
- Up-front leasing for a fixed period
- Pay-as-you-go use
- No-fee subsidized use (with restrictions)
Tier 1 is a "service contract" based lease plan that is available in four–year increments and is paid for upfront (non-refundable). Three types of HPC resources are available to investors:
|Usage Type||Price ($/Node-Equivalent*)|
*Node Equivalents are based on 2017 generation hardware: A CPU node has 32 CPU cores and 256 GB of RAM.
Tier 2 is a "pay as you go" plan where users are charged on a monthly basis for their usage. Two types of HPC resources are available through Tier 2.
*Tier 2 rates are adjusted quarterly.
High Performance Storage is a "pay as you go" plan where users are charged on a monthly basis for their usage.
Business & Industry Affiliates
The University's shared HPC resources are available to business, industry and off-campus affiliates as well. Access by external (non-University) users whose work connects to the University, such as a research collaboration, is coordinated through the Nevada Center for Applied Research and charged at higher rates (to be negotiated).
Consulting from the High Performance Computing Team
Our team is committed to collaborating with and advising researchers on complex research technology in areas including, but not limited to:
- High performance computing
- High speed networking
Submit inquiries for consulting to firstname.lastname@example.org. Please be prepared to provide a brief proposal that includes the following information:
- A list of researchers who will need accounts
- The faculty member(s) sponsoring the project
- The scientific background of your project for the proposed work
- The software pipeline for your project
- Programming language(s) used
- Parallelization mechanism ex: MPI, OpenMP, naive parallel
- Required libraries
- Number of concurrent CPUs per job
- RAM per job
- Total disk space