
CBI Infrastructure
To enable processing of investigators’ genomic data, we maintain numerous high-performance Linux-based compute resources, both locally in HCI and at CHPC.
HCI Resources
HCI maintains a number of Linux-based servers within our data center for use in processing bioinformatics data. For information on how to gain access to these compute resources, please contact us. These servers include the following:
- Uinta, an interactive server with 48 cores and 440 GB memory
- Alta, an interactive server with 32 cores and 512 GB memory
- Deadhorse, an interactive server with 80 cores and 470 GB memory
- An Illumina DRAGEN server with a dedicated FPGA processor for processing, de-multiplexing, and ORA compression of Illumina sequencing to support the High-Throughput Genomics Shared Resource
- A high performance 60 TB network file server for temporary, working storage
- A 128 TB archival network storage to support the GNomEx LIMS database
For the interactive HCI Linux servers, all machines mount the same network resources. Some common paths include the following.
/scratch
is for temporary file storage without storage quotas./home/BioApps
is a repository of bioinformatic applications not shared with CHPC nodes./tomato/dev/app
is a repository of bioinformatic applications shared with CHPC./tomato/dev/data
is a repository of genomic reference data shared with CHPC.- Some application packages are installed under Modules. Use
module avail
andmodule load
to list available modules and load a specific module, respectively. - A RStudio server is available for running R scripts and software. Email CBI for instructions on how to access it.
CHPC Resources
In addition to HCI resources, we own a number of compute nodes on clusters at the University of Utah Center for High Performance Computing (CHPC). Research labs that do not have dedicated compute allocations can request to access the nodes directly through a limited affiliate account at CHPC; please contact us for more information.
The nodes we own include the following:
- Ten nodes (32-48 cores, 64-128 GB RAM each) on the Kingspeak cluster
- Thirty-eight nodes (56-128 core, 128-256 GB RAM each) in the HIPAA/PHI-compliant protected environment Redwood cluster at CHPC. These are the nodes used for working with human-derived data.
We have common bioinformatic applications and genomic resources synced between HCI and CHPC. In the general environment at CHPC, you can find these at:
/uufs/chpc.utah.edu/common/home/hcibcore/atlatl/app
is a repository of bioinformatic applications./uufs/chpc.utah.edu/common/home/hcibcore/atlatl/data
is a repository of genomic reference data.
We also maintain our Module repository of applications and packages. To use this, add our directory to your environment with the following command:
module use /uufs/chpc.utah.edu/common/home/hcibcore/Modules/modulefiles
We keep similar resource directories in the restricted protected environment at CHPC; please contact us for details.