SCTi

Non-Profit – CFD Modeling

Home »  Non-Profit – CFD Modeling

Customer: A national Canadian health related organization.

Issue: The data centre was undergoing major upgrades and redesign including equipment layout. The objective was to extend the life of the existing infrastructure for approximately 5 years and try to increase overall energy efficiency. Major question that needed to be answered was, could the existing cooling infrastructure meet the growing heat loads that were projected for the next five years.

Challenge: The organization was investing a considerable amount of money to recondition the data centre, however it was not known what the life expectancy would be. The cooling infrastructure was aging and the room had a number of significant limitations such as low ceiling and minimal underfloor plenum height. Furthermore, the aging CRAC units could present a potential for failure and an analysis of the impact of this was required.

Process: SCTi developed a detailed model of the facility for Stage 1 implementation. By inputting this model into a Computational Fluid Dynamics (CFD) study we were able to model the heat loads, cooling capacity, air flow and air bypass to show how the redesigned data centre would operate. This enabled us to highlight where any potential hotspots may be and what the impact of a CRAC failure would be to the operation of the data centre. With the input from data centre staff, additional models were built simulating higher heat loads and equipment configurations, representative of planned growth over the next five years.

Solution: Based on the CFD analysis, SCTi provided the client with detailed simulations showing the air supply and exhaust temperature at each rack, volume of air flow to each rack, supply air paths and return air paths. Data centre staff were able to see where air flow was limited enabling them to plan on equipment configurations which would avoid these potential problem areas. This analysis was completed for the new design along with a number of designs based on expected growth patterns for the next 5 years. In each case, we were able to show where potential hot spots may arise due to air flow, what the capacity of the CRAC units would be to cool the facility, what the impact would be of the failure of a CRAC unit and whether aisle containment would be required.

Implementation: The client now knows the growth limits of the data centre and what action needs to be taken to reach the projected growth peak. This provides them with a clear timeline for planning a new facility while still operating economically within the current data centre.

Result: Based on the simulation results the client is able to optimize their cooling capacity by turning down one of the units and increasing set points to conserve energy. This action could result in energy savings 20% or more.