From a4102d0252236e85b2813160b4b11e3a19a00d62 Mon Sep 17 00:00:00 2001 From: mjkwiatkowski Date: Fri, 13 Feb 2026 12:46:54 +0100 Subject: feat: added some refinements to the introduction --- content/appendix.tex | 16 ---------------- content/intro.tex | 41 +++++++++++++++++++++++++++++++++++------ content/threats.tex | 5 +---- 3 files changed, 36 insertions(+), 26 deletions(-) delete mode 100644 content/appendix.tex (limited to 'content') diff --git a/content/appendix.tex b/content/appendix.tex deleted file mode 100644 index cc471e5..0000000 --- a/content/appendix.tex +++ /dev/null @@ -1,16 +0,0 @@ -% this file is called up by thesis.tex -% content in this file will be fed into the main document - -%: ----------------------- name of chapter ------------------------- -\chapter*{Appendix} % top level followed by section, subsection - - - - - - - -% --------------------------------------------------------------------------- -%: ----------------------- end of thesis sub-document ------------------------ -% --------------------------------------------------------------------------- - diff --git a/content/intro.tex b/content/intro.tex index 3eb6e09..58a759a 100644 --- a/content/intro.tex +++ b/content/intro.tex @@ -1,12 +1,41 @@ \chapter{Introduction}\label{s:intro} +Today's transportation systems, education and government largely depend on server-side services, which are hosted in datacentres~\cite{DBLP:journals/corr/IosupKLVG22}. +To facilitate the rising demand managers expand datacenters with new components and more heterogenous architectures (e.g., GPUs and NPUs)~\cite{DBLP:conf/date/MilojicicFDR21}. +However, in return datacenter complexity increases significantly. +To make better operational decisions despite the massive scale, new, promising technologies arise, such as datacenter Digital Twins. -\todo{ -This section includes some motivations behind the work, explicitly or implicitly -highlights the research question, provides a high-level explanation of the -solution, and describes the contributions. -} +\section{Context}\label{s:context} +Datacenters are one of the most important components of the digital society. +For example, over 25\% of professionals in the Netherlands depend on cloud services in their everyday work. +Faced with growing demand, this fraction will exceed 35\% by 2025~\cite{DBLP:journals/corr/IosupKLVG22}. +What is more, the surge of AI and Machine Learning workloads opens the need for versatile server architectures, pushing datacenter managers to meet customer expectations by adding more specialized hardware~\cite{DBLP:conf/date/MilojicicFDR21}. +In return, operating a modern datacenter with thousands of diversified servers presents a yet unsolved, non-trivial challenge that requires fast and well-informed decisions from on-site engineers. +To aid in datacenter management, operators turn to \gls{oda}, which is the process of analyzing monitoring data to gain insights into the system behavior. +For example, OMNI at \gls{nersc} and Wintermute at \gls{lrz} employ descriptive analytics to optimize power usage effectivenes~\cite{DBLP:conf/icppw/BourassaJBCJVS19} and prescriptive analysis for energy efficient scheduling~\cite{DBLP:conf/hpdc/NettiMGOTO020}. +Nonetheless, we observe a critical lack of predictive analysis capabilities~\cite{DBLP:conf/wosp/SumanCNTMI24} among the existing \gls{oda} frameworks. +In result, datacenter operators are often confronted with operational decisions with limited time to react, which can lead to missed \gls{sla}. -\lipsum[1-2] +``Lab-built, preproduction, or early hardware does \textit{not} work as defined, does \textit{not} work reliably and does \textit{not} stay the same from day to day'', +according to Frederick P. Brooks. +A solution is a dependable simulator of the system~\cite{DBLP:books/daglib/Brooks0080747}. +A novel improvement on simulation is a datacenter \gls{dt}~\cite{DBLP:journals/computer/AthavaleBBMMPS24}. +\section{Problem statement}\label{s:problem-statement} + +\section{Research Questions}\label{s:research-questions} + + + +\section{Research Methodology}\label{s:research-methodology} + +\section{Thesis Contributions}\label{s:thesis-contributions} + +\section{Plagiarism Declaration}\label{s:plagiarism-declaraion} +I hereby declare that this thesis is my own independent work and writing. +The thesis does not contain any material copied from other sources (person, Internet, or AI), and has not been submitted for assessment elsewhere. + +\section{Societal Impact}\label{s:societal-impact} + +\section{Thesis Structure}\label{s:thesis-structure} diff --git a/content/threats.tex b/content/threats.tex index 3025b7f..f01c3fb 100644 --- a/content/threats.tex +++ b/content/threats.tex @@ -1,9 +1,6 @@ \chapter{Threats To Validity}\label{sec:threats} -\todo{Report about each type of threat to the validity of the experiment, according to the classification framework proposed by Wohlin \etal \cite{wohlin12}. -} - \section{Internal Validity} \section{External Validity} \section{Construct Validity} -\section{Conclusion Validity} \ No newline at end of file +\section{Conclusion Validity} -- cgit v1.2.3