Energy consumption of parallel algorithms for solving linear systems on HPC architecture

Modern High-Performance Computing HPC systems are gradually increasing in size and complexity due to the correspondent demand of larger simulations requiring more complicated tasks and higher accuracy. However, as side effects of the Dennard’s scaling approaching its ultimate power limit, the efficiency of software plays also an important role in increasing the overall performance of a computation. Tools to measure application performance in these increasingly complex environments provide insights into the intricate ways in which software and hardware interact. The monitoring of the power consumption in order to save energy is possible through processors interfaces like Intel Running Average Power Limit RAPL. Given the low level of these interfaces, they are often paired with an application-level tool like Performance Application Programming Interface PAPI. Since several problems in many heterogeneous fields can be represented as a complex linear system, an optimized and scalable linear system solver algorithm can decrease significantly the time spent to compute its resolution. One of the most widely used algorithms deployed for the resolution of large simulation is the Gaussian Elimination, which has its most popular implementation for HPC systems in the Scalable Linear Algebra PACKage ScaLAPACK library. However, another relevant algorithm, which is increasing in popularity in the academic field, is the Inhibition Method. This thesis compares the energy consumption of the Inhibition Method and Gaussian Elimination from ScaLAPACK to profile their execution during the resolution of linear systems above the HPC architecture offered by CINECA. Moreover, it also collates the energy and power values for different ranks, nodes, and sockets configurations. The monitoring tools employed to track the energy consumption of these algorithms are PAPI and RAPL, that will be integrated with the parallel execution of the algorithms managed with the Message Passing Interface MPI.

Progetto e valutazione di una piattaforma sperimentale di Internet of Energy per veicoli elettrici

This dissertation deals with the development of a project concerning a demonstration in the scope of the Supply Chain 6 of the Internet of Energy (IoE) project: the Remote Monitoring Emulator, which bears my personal contribution in several sections. IoE is a project of international relevance, that means to establish an interoperability standard as regards the electric power production and utilization infrastructure, using Smart Space platforms. The future perspectives of IoE have to do with a platform for electrical power trade-of, the Smart Grid, whose energy is produced by decentralized renewable sources and whose services are exploited primarily according to the Internet of Things philosophy. The main consumers of this kind of smart technology will be Smart Houses (that is to say, buildings controlled by an autonomous system for electrical energy management that is interoperable with the Smart Grid) and Electric Mobility, that is a smart and automated management regarding movement and, overall, recharging of electrical vehicles. It is precisely in the latter case study that the project Remote Monitoring Emulator takes place. It consists in the development of a simulated platform for the management of an electrical vehicle recharging in a city. My personal contribution to this project lies in development and modeling of the simulation platform, of its counterpart in a mobile application and implementation of a city service prototype. This platform shall, ultimately, make up a demonstrator system exploiting the same device which a real user, inside his vehicle, would use. The main requirements that this platform shall satisfy will be interoperability, expandability and relevance to standards, as it needs to communicate with other development groups and to effectively respond to internal changes that can affect IoE.

From F energy rating to Nearly ZEB: project of energy retrofit and architectural upgrade of a brick terraced house in Co. Dublin, Ireland

Questo progetto è stato sviluppato durante un periodo di ricerca presso il Dipartimento di Ingegneria Civile del Trinity College e continuato presso l’Università di Bologna. Il progetto ha l’obiettivo di analizzare le soluzioni per l’ampliamento, la sostituzione degli impianti e l’ottimizzazione energetica di un tipico edificio residenziale Irlandese, una end of terrace in mattoni costruita negli anni ’20, collocata a Blackrock (Dublino). Diversi studi sostengono che lo stock abitativo irlandese è il peggiore del nord Europa per quanto riguarda la performance energetica. Questa tesi consta di una prima parte di studio del contesto e delle tecniche costruttive tradizionali irlandesi; è presente un capitolo di approfondimento sulle leggi riguardanti le costruzioni e gli incentivi forniti dal governo irlandese per interventi di retrofit energetico. Il terzo capitolo è un’analisi dell'esistente, con disegni del rilievo geometrico, immagini dell’edificio originale, termogrammi e dati riguardanti l’attuale performance energetica. Vengono poi mostrate diverse ipotesi di progetto e, una volta determinata la disposizione degli spazi interni, vengono considerate due soluzioni simili, ma costruite con pacchetti costruttivi diversi. Nel Progetto A l’involucro dell’addizione ha una struttura in muratura, nel Progetto B la struttura è in X-lam. Le performance energetiche delle due proposte vengono confrontate tramite una simulazione attuata grazie all'utilizzo del software dinamico IES-VE. Viene valutata l’applicazione di energie rinnovabili, quali l’energia solare e eolica e l’apporto che queste possono dare al bilancio energetico. Infine viene fatta un’analisi dei costi, valutando possibili suddivisioni dei lavori e ipotizzando un piano di ritorno dell’investimento, anche in combinazione con l’applicazione di energie rinnovabili. Alla fine del progetto si trova una valutazione quantitativa dei miglioramenti dell’edificio e un’analisi critica dei limiti del progetto.

Packet erasure correcting codes for wireless sensor networks: implementation and field trial measurements

This thesis regards the Wireless Sensor Network (WSN), as one of the most important technologies for the twenty-first century and the implementation of different packet correcting erasure codes to cope with the ”bursty” nature of the transmission channel and the possibility of packet losses during the transmission. The limited battery capacity of each sensor node makes the minimization of the power consumption one of the primary concerns in WSN. Considering also the fact that in each sensor node the communication is considerably more expensive than computation, this motivates the core idea to invest computation within the network whenever possible to safe on communication costs. The goal of the research was to evaluate a parameter, for example the Packet Erasure Ratio (PER), that permit to verify the functionality and the behavior of the created network, validate the theoretical expectations and evaluate the convenience of introducing the recovery packet techniques using different types of packet erasure codes in different types of networks. Thus, considering all the constrains of energy consumption in WSN, the topic of this thesis is to try to minimize it by introducing encoding/decoding algorithms in the transmission chain in order to prevent the retransmission of the erased packets through the Packet Erasure Channel and save the energy used for each retransmitted packet. In this way it is possible extend the lifetime of entire network.