The purpose of this discussion is to compare the current RAM memory technologies (in particular SRAM and DRAM) that use a technology based on CMOS transistors (SRAM) and capacitive (DRAM), with the most promising technologies in a near future, which base their functioning on the so-called spintronics (i.e. they exploit the spin of electrons to store data bits). In particular, these innovative memories will be analyzed starting from the one with the most limitations and then presenting the most efficient alternatives both in terms of energy and integration density, demonstrating how, thanks to the right improvements, they can compete and even replace SRAM ensuring better performances. In this discussion, various configurations and methodologies for applying spintronic technology to memory systems that are candidates to play an important role in the near future have been analyzed. Some types of memory are already in use at the time of writing this text (year 2021/2022) such as the STT-MRAM, others are under study and require further experimentation. This thesis has been structured to ensure a complete evaluation of the benefits of spintronics applied to storage systems, initially providing the general operating model for each category, designed to provide an initial understanding, and then moving on to the strengths and weaknesses, to draw a critical analysis; with regard to the latter, alternative solutions or modifications have been presented such that many problems that initially arose can be solved.
Lo scopo di questa trattazione è di porre a confronto le attuali tecnologie di memoria RAM (in particolare le SRAM e le DRAM) che utilizzano una tecnologia basata sui transistor CMOS (SRAM) e capacitiva (DRAM), con le tecnologie più promettenti in un prossimo futuro, che basano il loro funzionamento sulla cosiddetta spintronica (ossia sfruttano lo spin degli elettroni per immagazzinare i bit). In particolare, le memorie innovative verranno analizzate iniziando da quella con più limitazioni e presentando in seguito le alternative più efficienti sia a livello energetico che per densità di integrazione, dimostrando come, grazie alle giuste migliorie, esse possano competere e persino sostituire la SRAM garantendo prestazioni migliori. In questa trattazione sono state analizzate svariate configurazioni e metodologie di applicazione della tecnologia spintronica ai sistemi di memoria candidati a giocare un ruolo importante nel prossimo futuro. Alcuni tipi di memoria sono già in utilizzo al momento della scrittura di questo testo (anno 2021/2022) come per esempio le STT-MRAM, altri sono in fase di studio e richiedono ulteriori sperimentazioni. Questa tesi è stata strutturata per garantire una completa valutazione sui benefici della spintronica applicati ai sistemi di memorizzazione, fornendo per ogni categoria inizialmente il modello di funzionamento generale, atto a fornire un’iniziale comprensione, per poi passare ai punti di forza nonché di debolezza, per trarne un’analisi critica; in merito a questi ultimi, sono state presentate soluzioni alternative o modifiche tali per cui molti problemi sorti inizialmente possano essere risolti.
Analisi dello Stato dell'Arte di Memorie Magneto-Resistive per Memorizzazione non Volatile
BORTOLATO, GIOVANNI
2021/2022
Abstract
The purpose of this discussion is to compare the current RAM memory technologies (in particular SRAM and DRAM) that use a technology based on CMOS transistors (SRAM) and capacitive (DRAM), with the most promising technologies in a near future, which base their functioning on the so-called spintronics (i.e. they exploit the spin of electrons to store data bits). In particular, these innovative memories will be analyzed starting from the one with the most limitations and then presenting the most efficient alternatives both in terms of energy and integration density, demonstrating how, thanks to the right improvements, they can compete and even replace SRAM ensuring better performances. In this discussion, various configurations and methodologies for applying spintronic technology to memory systems that are candidates to play an important role in the near future have been analyzed. Some types of memory are already in use at the time of writing this text (year 2021/2022) such as the STT-MRAM, others are under study and require further experimentation. This thesis has been structured to ensure a complete evaluation of the benefits of spintronics applied to storage systems, initially providing the general operating model for each category, designed to provide an initial understanding, and then moving on to the strengths and weaknesses, to draw a critical analysis; with regard to the latter, alternative solutions or modifications have been presented such that many problems that initially arose can be solved.File | Dimensione | Formato | |
---|---|---|---|
Bortolato_Giovanni.pdf
accesso aperto
Dimensione
4.54 MB
Formato
Adobe PDF
|
4.54 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/10868