Nowadays, data centers are taking a central role in the world of ICT, due to the increase of the computational capabilities needed for modern applications, especially considering the growing importance of artificial intelligence and machine learning. In the context of Data centers, that can be defined in a high level manner as a set of servers providing computational capabilities to interested parties, alongside with the increasing demand and performances needed, there is the problem of the environmental impact, especially in terms of power consumption. A technology promising with regards to these problems is asymmetric multiprocessing. Asymmetric Multiprocessors, AMPs, are processors composed by core with different characteristics, so combining both performance oriented and power efficiency oriented capabilities. This thesis has the goal of exploring potential application of AMPs in the context of data centers, by implementing Petri Net models for theoretical data centers using AMPs, and analyzing their behaviour with respect to target performance indexes with JMT, an integrated framework of Java tools for the definition and performance evaluation of computing systems.
Al giorno d’oggi, i data center stanno assumendo un ruolo centrale nel mondo dell’ICT, a causa dell’aumento delle capacità computazionali necessarie per le applicazioni moderne, soprattutto considerando la crescente importanza dell’intelligenza artificiale e del machine learning. Nell’ambito dei data center, che possono essere definiti ad alto livello come un insieme di server che forniscono capacità computazionale a terzi interessati, accanto alla crescente domanda e prestazioni necessarie, si pone il problema dell’impatto ambientale, soprattutto in termini di consumo di energia. Una tecnologia promettente riguardo a questi problemi è il multiprocessing asimmetrico. I multiprocessori asimmetrici, AMP, sono processori composti da core con caratteristiche diverse, combinando così funzionalità sia orientate alle prestazioni che orientate all’efficienza energetica. Questa tesi ha l’obiettivo di esplorare la potenziale applicazione degli AMP nel contesto dei data center, implementando modelli di Reti di Petri per ipotetici data center utilizzanti AMP, e analizzando il loro comportamento grazie a indici di prestazione definiti tramite JMT, un framework integrato di applicazioni Java per la definizione e la valutazione delle prestazioni di sistemi informativi complessi.
Modelling asymmetric Data Centers with JMT
DE NOVELLIS, GIOVANNI
2023/2024
Abstract
Nowadays, data centers are taking a central role in the world of ICT, due to the increase of the computational capabilities needed for modern applications, especially considering the growing importance of artificial intelligence and machine learning. In the context of Data centers, that can be defined in a high level manner as a set of servers providing computational capabilities to interested parties, alongside with the increasing demand and performances needed, there is the problem of the environmental impact, especially in terms of power consumption. A technology promising with regards to these problems is asymmetric multiprocessing. Asymmetric Multiprocessors, AMPs, are processors composed by core with different characteristics, so combining both performance oriented and power efficiency oriented capabilities. This thesis has the goal of exploring potential application of AMPs in the context of data centers, by implementing Petri Net models for theoretical data centers using AMPs, and analyzing their behaviour with respect to target performance indexes with JMT, an integrated framework of Java tools for the definition and performance evaluation of computing systems.File | Dimensione | Formato | |
---|---|---|---|
2024_07_DeNovellis_Tesi.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Tesi
Dimensione
1.64 MB
Formato
Adobe PDF
|
1.64 MB | Adobe PDF | Visualizza/Apri |
2024_07_DeNovellis_ExecutiveSummary.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Executive Summary
Dimensione
518.48 kB
Formato
Adobe PDF
|
518.48 kB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/222723