\documentclass[12pt, handout]{beamer} \input{style/style.tex} \begin{document} \frame{\titlepage \centering \footnotesize Online slides: \url{https://www.overleaf.com/read/tknhxmqfgtdy\#87413e}} \begin{frame}\frametitle{Motivation} \begin{tcolorbox}[title=Context] 21\textsuperscript{st} century datacenters are primarily heterogeneous~\cite{DBLP:conf/date/MilojicicFDR21} and modern computational needs of AI drive managers to diversify datacenters even more~\cite{DBLP:journals/computer/AthavaleBBMMPS24}. In result datacenters become extremely complex and hard to operate with millions of CPU's, GPU's etc. \end{tcolorbox} \begin{center} \includegraphics[width=\linewidth]{images/datacenter_complexity.pdf} \end{center} \tiny \textbf{Figure 1.1:} Society depends on datacenters to keep running, and therefore we cannot afford to let these systems break down or experience significant performance-related issues. With millions of servers in the largest datacenters, real-time management becomes very difficult. Left to right: a Google datacenter, server racks, Ada Lovelace AD102 GPU architecture. \end{frame} \begin{frame}\frametitle{Problem Statement I} \begin{tcolorbox}[title=We need tools to tackle datacenter complexity] We need Datacenter Digital Twins (DCDT) to be better able to detect and solve issues in this critical infrastructure~\cite{DBLP:journals/computer/AthavaleBBMMPS24}. However, DCDT's are still actively developed, and lack crucial features such as predictive analytics~\cite{DBLP:usdoe/report/AP26894} to \emph{e.g.,} prevent unexpected job failures. \end{tcolorbox} \input{images/table.tex} \end{frame} \begin{frame}\frametitle{Problem Statement II} \begin{tcolorbox}[title=What are predictive analytics? Why are they important?] Predictive analytics use statistics to predict events in advance. Almost any statistical technique can be used for predictive analytics, but nowadays it is mostly associated with Artificial Intelligence and Machine Learning (\emph{e.g.,} linear regression). Reducing \emph{e.g.,} job failures with timely predictions could potentially save millions of \$USD~\cite{DBLP:conf/acsos/TalluriOVTI21}. \end{tcolorbox} \begin{center} \includegraphics[width=0.9\linewidth]{images/predictive_analytics.pdf} \end{center} \tiny \textbf{Figure 1.2:} Where does our work fit within the field of datacenter digital twinning? There are 5 core elements to any Digital Twin: \myCircled{A} The Digital $\rightarrow$ Physical Twin link, \myCircled{B} the Physical Twin (\emph{e.g.,} the datacenter), \myCircled{C} the Physical $\rightarrow$ Digital Twin link, \myCircled{D} the Digital Twin, \myCircled{E} the features necessary to any Digital Twin. \textcolor{ForestGreen}{\faHighlighter~Highlighted areas are the contributions from this thesis, which include the autonomous actions resulting from predictive insights \myCircledGreen{A} and the predictive analysis itself within \myCircledGreen{E}.} \end{frame} \begin{frame}\frametitle{Research Questions} \begin{tcolorbox}[title=Main Research Question, colbacktitle=red!70!black,colback=red!20!white] How to enable predictive analytics for datacenters through digital twinning? \end{tcolorbox} \begin{tcolorbox}[title=Research Question 1] How to asses the current state-of-the-art of digital twinning for datacenters? \end{tcolorbox} \begin{tcolorbox}[title=Research Question 2] How to design a datacenter digital twin reference architecture using discrete-event simulation and predictive data analytics? \end{tcolorbox} \begin{tcolorbox}[title=Research Question 3] How to evaluate and validate a datacenter digital twin architecture in relation to system requirements? \end{tcolorbox} \end{frame} \begin{frame}\frametitle{\textbf{RQ1}: System Model} \begin{minipage}[b]{0.45\linewidth} \begin{center} \includegraphics[width=1.15\textwidth]{images/system_model.pdf} \end{center} \tiny \textbf{Figure 1.3:} To answer \textbf{RQ1} we designed a generic datacenter digital twin system model based on a comprehensive literature review and findings from \textbf{Table 1.1}. \end{minipage} % Consider splitting the figure into 2 a.k.a. top and bottom. % Data Lake -> Data Storage \hfill \begin{minipage}[b]{0.45\linewidth} \begin{tcolorbox}[title=Contribution 1] Duis non quam. Maecenas vi- tae dolor in ipsum auctor vehicula. Vivamus nec nibh eget wisi varius pulvinar. Cras a lacus. Etiam et massa. Donec in nisl sit amet dui imperdiet vestibulum. Duis porttitor nibh id eros. \end{tcolorbox} \begin{tcolorbox}[title=Contribution 2] \ipsum[1] \end{tcolorbox} % The ugly \vspace is a quick fix to align the observations. \vspace{0.2cm} \end{minipage} \end{frame} \begin{frame}\frametitle{\textbf{RQ2}: Reference Architecture} \end{frame} \begin{frame}\frametitle{\textbf{RQ3}: Experimental Results I} \begin{tcolorbox}[title=Main Finding I] Here explain what did you find. \end{tcolorbox} Here goes the figure that backs up claim in Main Finding I. Evidence for Main Finding I. % Explain what the axis are in the figure caption. % Talk about the experimental setup in the figure. % Give more reliable results than just numbers -- do statistical testing, i.e., standard deviation, confidence intervals. \end{frame} \begin{frame}\frametitle{\textbf{RQ3}: Experimental Results II} \begin{tcolorbox}[title=Main Finding II] Here explain what did you find. \end{tcolorbox} Here goes the figure that backs up claim in Main Finding II. \end{frame} \begin{frame}\frametitle{Key Takeaways} \begin{tcolorbox}[title=What is the societal context?] \end{tcolorbox} \begin{tcolorbox}[title=What problem did we solve?] \end{tcolorbox} \begin{tcolorbox}[title=How did we solve this problem?] \end{tcolorbox} \begin{tcolorbox}[colbacktitle=red!70!black, colback=red!20!white,title=What did we find?] \end{tcolorbox} \begin{tcolorbox}[title=What do we see in future work?] \end{tcolorbox} \end{frame} \setcounter{framenumber}{3} \setbeamertemplate{footline}[page number]{ % Unfortunately this must remain here. \setbeamercolor{frametitle}{fg=Brown,bg=Brown!20} \setbeamertemplate{frametitle}{ \vspace*{-0.1cm} \begin{beamercolorbox}[wd=\paperwidth, ht=0.75cm, dp=0.3cm,leftskip=10pt, rightskip=10pt]{frametitle} \usebeamerfont{frametitle}\insertframetitle\hfill \end{beamercolorbox} } \begin{frame}[allowframebreaks]\frametitle{Extra Slides: References} \tiny \bibliographystyle{is-plain} \bibliography{main.bib} \end{frame} \begin{frame}\frametitle{Extra Slides: Societal Impact} \begin{tcolorbox}[title=Why is this research important today?] Over 3 million jobs in the Netherlands directly depend on cloud services, which are hosted in datacenters~\cite{DBLP:journals/corr/IosupKLVG22}. Already the rapid expansion of datacenters has increased the presence of service failures across all cloud services~\cite{DBLP:conf/acsos/TalluriOVTI21}. We need to act now. \end{tcolorbox} \begin{center} \includegraphics[height=10em]{images/manageability.pdf} \end{center} \tiny \textbf{Figure E.1:} Horizontally: the most important research areas in computer science in Netherlands. Vertically: qualities we should ensure across all research areas with the most outstanding impact on society. Datacenter manageability is a top-priority~\cite{DBLP:journals/corr/IosupKLVG22}. \end{frame} \begin{frame}\frametitle{Extra Slides: Why Digital Twinning?} \begin{tcolorbox}[title=Definition] A DCDT mirrors the structure, context and behaviour of a datacenter~\cite{DBLP:journals/computer/AthavaleBBMMPS24}. The prerequisite to any digital twin is good monitoring and sensing capabilities in the physical entity. Datacenters meet this requirement easily because they already connect hundreds of monitoring sensors. \end{tcolorbox} \begin{center} \includegraphics[height=10em]{images/dt_timeline.pdf} \end{center} \tiny \textbf{Figure E.2:} Due to insufficient technological foundations, little work is available on DTs between 2003 and 2018, and it is only with the rapid growth of cloud computing, Internet-of-Things and Big Data analytics that DTs have reemerged~\cite{DBLP:conf/cirp/TAO2018169}. That is why nobody used digital twins to mirror datacenters earlier. \end{frame} \begin{frame}\frametitle{Extra Slides: Why not pure simulation?} \begin{tcolorbox}[title=Predicting job failures] Preventing failure-caused outages in advance can reduce huge operational costs, as over 20\% of all reported outages amount to more than 1 million US\$~\cite{DBLP:report/AnnualOutageAnalysis2025}. Only a constant bi-directional interaction (digital twin $\iff$ physical entity) can achieve this. \end{tcolorbox} \begin{center} \includegraphics[height=10em]{images/digital_twin_ms.pdf} \end{center} \tiny \textbf{Figure E.3:} Real-time control that is tightly-coupled with the IT equipment is a prerequisite for timely predictions within seconds/minutes~\cite{DBLP:journals/computer/AthavaleBBMMPS24}. \end{frame} % Computational Fluid Dynamics (CFD) have high computation overhead, unsuitable for real-time simulation of a dynamic datacenter. %Moreover oftentimes a poorly configured CFD model can lead to high error rates~\cite{DBLP:conf/sensys/WangZD0TCWZ20}. %Data-driven Machine Learning performs poorly by the cases not covered in the training data. \end{document}