Graduation Semester and Year

Spring 2024

Language

English

Document Type

Thesis

Degree Name

Master of Science in Mechanical Engineering

Department

Mechanical and Aerospace Engineering

First Advisor

Dr. Dereje Agonafer

Second Advisor

Dr. Abdolhossein Haji-Sheikh

Third Advisor

Dr. Satyam Saini

Abstract

As data centers evolve to support increasing server densities, the inadequacies of traditional air cooling in managing thermal loads have become evident. This is primarily due to an increased demand for powerful CPU and GPU-based platforms that can run Artificial Intelligence (AI) and Machine Learning (ML) workloads amongst others. Consequently, single-phase immersion cooling has gained prominence as an efficient, sustainable alternative, significantly reducing operational costs and energy consumption. This method has shown a remarkable ability to cut cooling energy by up to 90% and overall data center energy use by 50%. The current investigation uses Particle Image Velocimetry (PIV) to quantify the flow profiles in a functional 1U Coyote Pass server (M50CYP), cooled using an immersion fluid Polyalphaolefin 6 (PAO6), a synthetic hydrocarbon lubricant known for its thermal advantages, to cool a server with an inlet temperature set at 40 degrees Celsius. PG-25 water-based coolant circulates in a coolant distribution unit, coupled with a counterflow heat exchanger to optimize heat dissipation. The cooling system consists of a single-phase immersion-cooled 2U tank filled with PAO as a dielectric fluid.

This study aims to evaluate the effectiveness of single-phase immersion cooling in maintaining optimal server temperatures. The focus is on analyzing the 3D velocity vector field in critical components such as CPUs and heatsinks using Tomographic PIV. Tomographic PIV enables the measurement of fluid velocity by reconstructing a three-dimensional structure from a series of two-dimensional images and tracking the motion of particles suspended in the fluid. Also, this investigation examines the thermal performance of CPUs to gain insights into the cooling system's overall effectiveness.

Disciplines

Mechanical Engineering

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.