Monitoring and Prediction of Thermal Emergencies in High Performance Computing Systems

Modern scientific discoveries are driven by an unsatisfiable demand for computational resources. High-Performance Computing (HPC) systems are an aggregation of computing power to deliver considerably higher performance than one typical desktop computer can provide, to solve large problems in science, engineering, or business. An HPC room in the datacenter is a complex controlled environment that hosts thousands of computing nodes that consume electrical power in the range of megawatts, which gets completely transformed into heat. Although a datacenter contains sophisticated cooling systems, our studies indicate quantitative evidence of thermal bottlenecks in real-life production workload, showing the presence of significant spatial and temporal thermal and power heterogeneity. Therefore minor thermal issues/anomalies can potentially start a chain of events that leads to an unbalance between the amount of heat generated by the computing nodes and the heat removed by the cooling system originating thermal hazards. Although thermal anomalies are rare events, anomaly detection/prediction in time is vital to avoid IT and facility equipment damage and outage of the datacenter, with severe societal and business losses. For this reason, automated approaches to detect thermal anomalies in datacenters have considerable potential. This thesis analyzed and characterized the power and thermal characteristics of a Tier0 datacenter (CINECA) during production and under abnormal thermal conditions. Then, a Deep Learning (DL)-powered thermal hazard prediction framework is proposed. The proposed models are validated against real thermal hazard events reported for the studied HPC cluster while in production. This thesis is the first empirical study of thermal anomaly detection and prediction techniques of a real large-scale HPC system to the best of my knowledge. For this thesis, I used a large-scale dataset, monitoring data of tens of thousands of sensors for around 24 months with a data collection rate of around 20 seconds.

Flexible Computing Systems For AI Acceleration At The Extreme Edge Of The IoT

Embedding intelligence in extreme edge devices allows distilling raw data acquired from sensors into actionable information, directly on IoT end-nodes. This computing paradigm, in which end-nodes no longer depend entirely on the Cloud, offers undeniable benefits, driving a large research area (TinyML) to deploy leading Machine Learning (ML) algorithms on micro-controller class of devices. To fit the limited memory storage capability of these tiny platforms, full-precision Deep Neural Networks (DNNs) are compressed by representing their data down to byte and sub-byte formats, in the integer domain. However, the current generation of micro-controller systems can barely cope with the computing requirements of QNNs. This thesis tackles the challenge from many perspectives, presenting solutions both at software and hardware levels, exploiting parallelism, heterogeneity and software programmability to guarantee high flexibility and high energy-performance proportionality. The first contribution, PULP-NN, is an optimized software computing library for QNN inference on parallel ultra-low-power (PULP) clusters of RISC-V processors, showing one order of magnitude improvements in performance and energy efficiency, compared to current State-of-the-Art (SoA) STM32 micro-controller systems (MCUs) based on ARM Cortex-M cores. The second contribution is XpulpNN, a set of RISC-V domain specific instruction set architecture (ISA) extensions to deal with sub-byte integer arithmetic computation. The solution, including the ISA extensions and the micro-architecture to support them, achieves energy efficiency comparable with dedicated DNN accelerators and surpasses the efficiency of SoA ARM Cortex-M based MCUs, such as the low-end STM32M4 and the high-end STM32H7 devices, by up to three orders of magnitude. To overcome the Von Neumann bottleneck while guaranteeing the highest flexibility, the final contribution integrates an Analog In-Memory Computing accelerator into the PULP cluster, creating a fully programmable heterogeneous fabric that demonstrates end-to-end inference capabilities of SoA MobileNetV2 models, showing two orders of magnitude performance improvements over current SoA analog/digital solutions.

PCM-based in-memory computing: architectures, circuits and applications

Analog In-memory Computing (AIMC) has been proposed in the context of Beyond Von Neumann architectures as a valid strategy to reduce internal data transfers energy consumption and latency, and to improve compute efficiency. The aim of AIMC is to perform computations within the memory unit, typically leveraging the physical features of memory devices. Among resistive Non-volatile Memories (NVMs), Phase-change Memory (PCM) has become a promising technology due to its intrinsic capability to store multilevel data. Hence, PCM technology is currently investigated to enhance the possibilities and the applications of AIMC. This thesis aims at exploring the potential of new PCM-based architectures as in-memory computational accelerators. In a first step, a preliminar experimental characterization of PCM devices has been carried out in an AIMC perspective. PCM cells non-idealities, such as time-drift, noise, and non-linearity have been studied to develop a dedicated multilevel programming algorithm. Measurement-based simulations have been then employed to evaluate the feasibility of PCM-based operations in the fields of Deep Neural Networks (DNNs) and Structural Health Monitoring (SHM). Moreover, a first testchip has been designed and tested to evaluate the hardware implementation of Multiply-and-Accumulate (MAC) operations employing PCM cells. This prototype experimentally demonstrates the possibility to reach a 95% MAC accuracy with a circuit-level compensation of cells time drift and non-linearity. Finally, empirical circuit behavior models have been included in simulations to assess the use of this technology in specific DNN applications, and to enhance the potentiality of this innovative computation approach.

Quantum computing and post-quantum cryptography

One of the main practical implications of quantum mechanical theory is quantum computing, and therefore the quantum computer. Quantum computing (for example, with Shor’s algorithm) challenges the computational hardness assumptions, such as the factoring problem and the discrete logarithm problem, that anchor the safety of cryptosystems. So the scientific community is studying how to defend cryptography; there are two defense strategies: the quantum cryptography (which involves the use of quantum cryptographic algorithms on quantum computers) and the post-quantum cryptography (based on classical cryptographic algorithms, but resistant to quantum computers). For example, National Institute of Standards and Technology (NIST) is collecting and standardizing the post-quantum ciphers, as it established DES and AES as symmetric cipher standards, in the past. In this thesis an introduction on quantum mechanics was given, in order to be able to talk about quantum computing and to analyze Shor’s algorithm. The differences between quantum and post-quantum cryptography were then analyzed. Subsequently the focus was given to the mathematical problems assumed to be resistant to quantum computers. To conclude, post-quantum digital signature cryptographic algorithms selected by NIST were studied and compared in order to apply them in today’s life.

analisi delle prestazioni del cluster risc-v: monte cimone

Gli sforzi di ricerca relativi all'High Performance Computing, nel corso degli anni, hanno prodotto risultati importanti inerenti all'incremento delle prestazioni sia in termini di numero di operazioni effettuate per periodo temporale, sia introducendo o migliorando algoritmi paralleli presenti in letteratura. Tali traguardi hanno comportato cambiamenti alla struttura interna delle macchine; si è assistito infatti ad un'evoluzione delle architetture dei processori utilizzati e all'impiego di GPU come risorse di calcolo aggiuntive. La conseguenza di un continuo incremento di prestazioni è quella di dover far fronte ad un grosso dispendio energetico, in quanto le macchine impiegate nell'HPC sono ideate per effettuare un'intensa attività di calcolo in un periodo di tempo molto prolungato; l'energia necessaria per alimentare ciascun nodo e dissipare il calore generato comporta costi elevati. Tra le varie soluzioni proposte per limitare il consumo di energia, quella che ha riscosso maggior interesse, sia a livello di studio che di mercato, è stata l'integrazione di CPU di tipologia RISC (Reduced Instruction Set Computer), in quanto capaci di ottenere prestazioni soddisfacenti con un impiego energetico inferiore rispetto alle CPU CISC (Complex Instruction Set Computer). In questa tesi è presentata l'analisi delle prestazioni di Monte Cimone, un cluster composto da 8 nodi di calcolo basati su architettura RISC-V e distribuiti in 4 piattaforme (\emph{blade}) dual-board. Verranno eseguiti dei benchmark che ci permetteranno di valutare: le prestazioni dello scambio di dati a lunga e corta distanza; le prestazioni nella risoluzione di problemi che presentano un principio di località spaziale ridotto; le prestazioni nella risoluzione di problemi su grafi e, nello specifico, ricerca in ampiezza e cammini minimi da sorgente singola.

Development of a heterogeneous clustering algorithm for particle shower reconstruction in high energy physics using the SYCL abstraction layer

Nei prossimi anni è atteso un aggiornamento sostanziale di LHC, che prevede di aumentare la luminosità integrata di un fattore 10 rispetto a quella attuale. Tale parametro è proporzionale al numero di collisioni per unità di tempo. Per questo, le risorse computazionali necessarie a tutti i livelli della ricostruzione cresceranno notevolmente. Dunque, la collaborazione CMS ha cominciato già da alcuni anni ad esplorare le possibilità offerte dal calcolo eterogeneo, ovvero la pratica di distribuire la computazione tra CPU e altri acceleratori dedicati, come ad esempio schede grafiche (GPU). Una delle difficoltà di questo approccio è la necessità di scrivere, validare e mantenere codice diverso per ogni dispositivo su cui dovrà essere eseguito. Questa tesi presenta la possibilità di usare SYCL per tradurre codice per la ricostruzione di eventi in modo che sia eseguibile ed efficiente su diversi dispositivi senza modifiche sostanziali. SYCL è un livello di astrazione per il calcolo eterogeneo, che rispetta lo standard ISO C++. Questo studio si concentra sul porting di un algoritmo di clustering dei depositi di energia calorimetrici, CLUE, usando oneAPI, l'implementazione SYCL supportata da Intel. Inizialmente, è stato tradotto l'algoritmo nella sua versione standalone, principalmente per prendere familiarità con SYCL e per la comodità di confronto delle performance con le versioni già esistenti. In questo caso, le prestazioni sono molto simili a quelle di codice CUDA nativo, a parità di hardware. Per validare la fisica, l'algoritmo è stato integrato all'interno di una versione ridotta del framework usato da CMS per la ricostruzione. I risultati fisici sono identici alle altre implementazioni mentre, dal punto di vista delle prestazioni computazionali, in alcuni casi, SYCL produce codice più veloce di altri livelli di astrazione adottati da CMS, presentandosi dunque come una possibilità interessante per il futuro del calcolo eterogeneo nella fisica delle alte energie.

Serverless Computing il caso Google Cloud Functions

Il serverless é un paradigma del cloud computing al giorno d’oggi sempre più diffuso; si basa sulla scrittura di funzioni stateless in quanto le attività relative alla loro manutenzione e scalabilità fanno capo al provider dei servizi cloud. Lo sviluppatore deve quindi solamente concentrarsi sulla creazione del prodotto. Questo lavoro si apre con un’analisi del cloud computing introducendo i principali modelli di applicazione, passando dal parlare di servizi cloud, con le varie sottocategorie e i relativi utilizzi per poi arrivare a parlare di serverless. Si é scelto di focalizzarsi sulla piattaforma Google con la suite: Google Cloud Platform. In particolare, si parlerà di Google Cloud Functions, una nuova offerta serverless della compagnia, di recente sviluppo e in continuo aggiornamento. Partiremo dalle prime release, analizzeremo l’ambiente di sviluppo, i casi d’uso, vantaggi, svantaggi, parleremo poi di portabilità e verranno mostrati alcuni esempi del loro utilizzo.

Stereo matching su GPU

La Stereo Vision è un popolare argomento di ricerca nel campo della Visione Artificiale; esso consiste nell’usare due immagini di una stessa scena,prodotte da due fotocamere diverse, per estrarre informazioni in 3D. L’idea di base della Stereo Vision è la simulazione della visione binoculare umana:le due fotocamere sono disposte in orizzontale per fungere da “occhi” che guardano la scena in 3D. Confrontando le due immagini ottenute, si possono ottenere informazioni riguardo alle posizioni degli oggetti della scena.In questa relazione presenteremo un algoritmo di Stereo Vision: si tratta di un algoritmo parallelo che ha come obiettivo di tracciare le linee di livello di un area geografica. L’algoritmo in origine era stato implementato per la Connection Machine CM-2, un supercomputer sviluppato negli anni 80, ed era espresso in *Lisp, un linguaggio derivato dal Lisp e ideato per la macchina stessa. Questa relazione tratta anche la traduzione e l’implementazione dell’algoritmo in CUDA, ovvero un’architettura hardware per l’elaborazione pa- rallela sviluppata da NVIDIA, che consente di eseguire codice parallelo su GPU. Si darà inoltre uno sguardo alle difficoltà che sono state riscontrate nella traduzione da *Lisp a CUDA.

Integrating mobile devices into a scalable Grid Computing architecture designed to perform distributed computations

The idea of Grid Computing originated in the nineties and found its concrete applications in contexts like the SETI@home project where a lot of computers (offered by volunteers) cooperated, performing distributed computations, inside the Grid environment analyzing radio signals trying to find extraterrestrial life. The Grid was composed of traditional personal computers but, with the emergence of the first mobile devices like Personal Digital Assistants (PDAs), researchers started theorizing the inclusion of mobile devices into Grid Computing; although impressive theoretical work was done, the idea was discarded due to the limitations (mainly technological) of mobile devices available at the time. Decades have passed, and now mobile devices are extremely more performant and numerous than before, leaving a great amount of resources available on mobile devices, such as smartphones and tablets, untapped. Here we propose a solution for performing distributed computations over a Grid Computing environment that utilizes both desktop and mobile devices, exploiting the resources from day-to-day mobile users that alternatively would end up unused. The work starts with an introduction on what Grid Computing is, the evolution of mobile devices, the idea of integrating such devices into the Grid and how to convince device owners to participate in the Grid. Then, the tone becomes more technical, starting with an explanation on how Grid Computing actually works, followed by the technical challenges of integrating mobile devices into the Grid. Next, the model, which constitutes the solution offered by this study, is explained, followed by a chapter regarding the realization of a prototype that proves the feasibility of distributed computations over a Grid composed by both mobile and desktop devices. To conclude future developments and ideas to improve this project are presented.

Censored Linear Regression Models For Irregularly Observed Longitudinal Data Using The Multivariate-t Distribution.

In acquired immunodeficiency syndrome (AIDS) studies it is quite common to observe viral load measurements collected irregularly over time. Moreover, these measurements can be subjected to some upper and/or lower detection limits depending on the quantification assays. A complication arises when these continuous repeated measures have a heavy-tailed behavior. For such data structures, we propose a robust structure for a censored linear model based on the multivariate Student's t-distribution. To compensate for the autocorrelation existing among irregularly observed measures, a damped exponential correlation structure is employed. An efficient expectation maximization type algorithm is developed for computing the maximum likelihood estimates, obtaining as a by-product the standard errors of the fixed effects and the log-likelihood function. The proposed algorithm uses closed-form expressions at the E-step that rely on formulas for the mean and variance of a truncated multivariate Student's t-distribution. The methodology is illustrated through an application to an Human Immunodeficiency Virus-AIDS (HIV-AIDS) study and several simulation studies.

Experimento didático de quimiometria para análise exploratória de óleos vegetais comestíveis por espectroscopia no infravermelho médio e análise de componentes principais: um tutorial, parte I

This manuscript aims to show the basic concepts and practical application of Principal Component Analysis (PCA) as a tutorial, using Matlab or Octave computing environment for beginners, undergraduate and graduate students. As a practical example it is shown the exploratory analysis of edible vegetable oils by mid infrared spectroscopy.

Desenvolvimento de ferramentas de análise de desempenho de atletas em 3 diferentes modalidades : Handebol, Provas Combinadas e Salto com Vara

Universidade Estadual de Campinas. Faculdade de Educação Física

Saude coletiva e aptdão fisica de escolares de segundo grau : estudo a partis do Colegio Tecnico Industrial - UNESP, Bauru

Universidade Estadual de Campinas . Faculdade de Educação Física

Stress distribution in the cervical region of an upper central incisor in a 3D finite element model

The aim of this study was to evaluate the stress distribution in the cervical region of a sound upper central incisor in two clinical situations, standard and maximum masticatory forces, by means of a 3D model with the highest possible level of fidelity to the anatomic dimensions. Two models with 331,887 linear tetrahedral elements that represent a sound upper central incisor with periodontal ligament, cortical and trabecular bones were loaded at 45º in relation to the tooth's long axis. All structures were considered to be homogeneous and isotropic, with the exception of the enamel (anisotropic). A standard masticatory force (100 N) was simulated on one of the models, while on the other one a maximum masticatory force was simulated (235.9 N). The software used were: PATRAN for pre- and post-processing and Nastran for processing. In the cementoenamel junction area, tensile forces reached 14.7 MPa in the 100 N model, and 40.2 MPa in the 235.9 N model, exceeding the enamel's tensile strength (16.7 MPa). The fact that the stress concentration in the amelodentinal junction exceeded the enamel's tensile strength under simulated conditions of maximum masticatory force suggests the possibility of the occurrence of non-carious cervical lesions such as abfractions.