Lighthouse Configuration

Hardware

Login

While the hardware provided for Lighthouse may vary, the login nodes and process remains uniform.  All Lighthouse users should login to lighthouse.arc-ts.umich.edu in order to access their allocation. Detailed instructions are available in the Lighthouse User Guide.

Networking

The compute nodes are all interconnected with InfiniBand networking. In addition to the InfiniBand networking, there is a gigabit Ethernet network that also connects all of the nodes. This is used for node management and NFS file system access.

Storage

The high-speed scratch file system is based on Lustre v2.5 and is a DDN SFA10000 backed by the hardware described in this table:

Server Type

Network Connection

Disk Capacity (raw/usable)

Dell R610 40Gbps InfiniBand 520 TB / 379 TB
Dell R610 40Gbps InfiniBand 530 TB / 386 TB
Dell R610 40Gbps InfiniBand 530 TB / 386 TB
Dell R610 40Gbps InfiniBand 520 TB / 379 TB

Totals

160 Gbps

2100 TB / 1530 TB

See our list of storage options to find the best fit for your needs.

Operation

Computing jobs on Lighthouse are managed completely through Slurm.  See the Lighthouse User Guide for directions on how to submit and manage jobs.

Software

There are three layers of software on Lighthouse:

Operating Software

The Lighthouse cluster runs CentOS 7. We update the operating system on Lighthouse as CentOS releases new versions and our library of third-party applications offers support. Due to the need to support several types of drivers (AFS and Lustre file system drivers, InfiniBand network drivers and NVIDIA GPU drivers) and dozens of third party applications, we are cautious in upgrading and can lag CentOS’s releases by months.

Compilers and Parallel and Scientific Libraries

Lighthouse supports the Gnu Compiler Collection, the Intel Compilers, and the PGI Compilers for C and Fortran. The Lighthouse cluster’s parallel library is OpenMPI, and the default versions are 1.10.7 (i686) and 3.1.2 (x86_64), and there are limited earlier versions available.  Lighthouse provides the Intel Math Kernel Library (MKL) set of high-performance mathematical libraries. Other common scientific libraries are compiled from source and include HDF5, NetCDF, FFTW3, Boost, and others.

Please contact us if you have questions about the availability of, or support for, any other compilers or libraries.

If you have questions, please send email to hpc-support@umich.edu.

To Order Lighthouse Service:

Email hpc-support@umich.edu before including hardware in a proposal or purchasing hardware.  You will need to purchase a subscription to Lighthouse for each compute node.

Related Events

July 19 @ 3:00 pm - 4:30 pm

More Mixed Models

In the R world, lme4 is a great package for mixed model estimation, and the most widely used for such models.  For standard settings, few tools will do the trick…

July 22 @ 12:00 am - July 23 @ 12:00 am

GlobusWorld Tour

Globus is coming to Michigan July 22-23 to present one of their Globus Tour Workshops.  The purpose is present to developers and researchers on the use of their tools (Globus…

July 22 @ 9:00 am - 11:00 am

Transitioning from Flux to Great Lakes

Lecture w/ Q&A   No Registration required. Drop-ins only. Thanks!

July 25 @ 1:00 pm - 2:30 pm

Spin Class: No Sweat Reports in R

If you use R, there’s a decent chance you are already familiar with using Rmarkdown files and the knitr package to create reports documenting your analyses. A lesser known but…