Data Center Service

Service Description

HPC cooling is provided by a highly specialized N+1 closed loop cooling system. This closed loop cooling system isolates cold intake air from hot exhaust air at the cabinet level to improve efficiency and support high density computing.  The system utilizes in-cabinet environmental sensors to manage airflow volume and temperature. There are also in-plenum sensors that can activate a secondary/backup cooling source if necessary. The secondary/backup cooling is provided by N+1 60Ton CRAH units equipped with variable speed fans. These units provide room/row cooling through perforated floor tiles.

Chilled water is provided by the University’s Central Energy Plant with a backup IT chiller system located at the Data Center

Humidification is monitored and maintained in-line with industry standards for data center environments. The humidification is provided by multiple high pressure misting units and the entire system is monitored and managed through an automated control system.

Power is monitored and conditioned by redundant Data Center Uninterruptible Power Supply (UPS) units, a generator and all requisite switch gear, breakers and management/monitoring systems. Power is actively monitored for quality. The UPS can provide stand-in power for not less than 10 minutes during power outages.

The backup generator is configured to automatically start when there is a loss of commercial/utility power. The generator can start and provide power in less than 30 seconds.
The Data Center Operations Staff supports and monitors the data center infrastructure 7x24x365. The Data Center is staffed Monday through Friday, 8am-5pm. After hours, weekends and holidays are handled through on-call staffing.

The Data Center Operations Phone is (307) 766-2747. Voice mail for this number is monitored during and after business hours. The Data Center can be contacted via email at
Maintenance and Testing
The Data Center Operations Staff performs prescribed maintenance and testing of the Data Center Electrical and Mechanical Systems throughout the year. These activities are normally scheduled during weekends and holidays and are designed to not impact service.

On occasion, emergency maintenance and testing is necessary to repair faulty or failing equipment. These events and associated impacts will be reported to the IT HPC Support/Management Team and subsequently to the HPC community as quickly as possible.