Performance-hungry data center applications demand increasingly higher performance from their storage in addition to larger capacity memory at lower cost. While the existing storage technologies (e.g., HDD and flash-based SSD) are limited in their performance, the most prevalent memory technology (DRAM) is unable to address the capacity and cost requirements of these applications. Emerging byte-addressable, non-volatile memory technologies (such as PCM and RRAM) offer performance within an order of magnitude of DRAM, prompting their inclusion in the processor memory subsystem. Such load/store accessible non-volatile or persistent memory (referred to as NVM or PM) introduces an interesting new tier that bridges the performance gap between DRAM and PM, and serves the role of fast storage or slower memory. However, PM has several implications on system design, both hardware and software: (i) the hardware caching mechanisms, while necessary for acceptable performance, complicate the ordering and durability of stores to PM, (ii) the high performance of PM (compared to NAND) and the fact that it is byte-addressable necessitate rethinking of the system software to manage PM and the interfaces to expose PM to the applications, and (iii) the future memory-based applications that will likely employ systems coupling PM with DRAM (for cost and capacity reasons) must be extremely conscious of the performance characteristics of PM and the challenges of using fast vs. slow memory in ways that best meet their performance demands. The key contribution of our research is a set of technologies that addresses these challenges in a bottom-up fashion. Since the real hardware is not yet available, we first implement a hardware emulator that can faithfully emulate the relative performance characteristics of DRAM and PM in a system with separate DRAM and emulated PM regions. We use this emulator to perform all of our evaluations. Next we explore system software support to enable low-overhead PM access by new and legacy applications. Towards this end, we implement PMFS, an optimized light-weight POSIX file system that exploits PM's byte-addressability to avoid overheads of block-oriented storage and enable direct PM access by applications (with memory-mapped I/O). To provide strong consistency guarantees, PMFS requires only a simple hardware primitive that provides software enforceable guarantees of durability and ordering of stores to PM. We demonstrate that PMFS achieves significant (up to an order of magnitude) gains over traditional file systems (such as ext4) on a RAMDISK-like PM block device. Finally, we address the problem of designing memory-based applications for systems with both DRAM and PM by extending our system software to manage both the tiers. We demonstrate for several representative large in-memory applications that it is possible to use a small amount of fast DRAM and large amounts of slower PM without a proportional impact to an application's performance, provided the placement of data structures is done in a careful fashion. To simplify the application programming, we implement a set of libraries and automatic tools (called X-Mem) that enables programmers to achieve optimal data placement with minimal effort on their part. Finally, we demonstrate the potentially large benefits of application-driven memory tiering with X-Mem across a range of applications.

Despite the association between hippocampal atrophy and a vast array of highly debilitating neurological diseases, such as Alzheimer’s disease and frontotemporal lobar degeneration, tools to accurately and robustly quantify the degeneration of this structure still largely elude us. In this thesis, we firstly evaluate previously-developed hippocampal segmentation methods (FMRIB’s Integrated Registration and Segmentation Tool (FIRST), Freesurfer (FS), and three versions of a Classifier Fusion (CF) technique) on two clinical MR datasets, to gain a better understanding of the modes of success and failure of these techniques, and to use this acquired knowledge for subsequent method improvement (e.g., FIRSTv3). Secondly, a fully automated, novel hippocampal segmentation method is developed, termed Fast Marching for Automated Segmentation of the Hippocampus (FMASH). This combined region-growing and atlas-based approach uses a 3D Sethian Fast Marching (FM) technique to propagate a hippocampal region from an automatically-defined seed point in the MR image. Region growth is dictated by both subject-specific intensity features and a probabilistic shape prior (or atlas). Following method development, FMASH is thoroughly validated on an independent clinical dataset from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), with an investigation of the dependency of such atlas-based approaches on their prior information. In response to our findings, we subsequently present a novel label-warping approach to effectively account for the detrimental effects of using cross-dataset priors in atlas-based segmentation. Finally, a clinical application of MR hippocampal segmentation is presented, with a combined MR-PET analysis of wholefield and subfield hippocampal changes in Alzheimer’s disease and frontotemporal lobar degeneration. This thesis therefore contributes both novel computational tools and valuable knowledge for further neurological investigations in both the academic and the clinical field.

The ferroelectricity in fluorite-structure oxides such as hafnia and zirconia has attracted increasing interest since 2011. They have various advantages such as Si-based complementary metal oxide semiconductor-compatibility, matured deposition techniques, a low dielectric constant and the resulting decreased depolarization field, and stronger resistance to hydrogen annealing. However, the wake-up effect, imprint, and insufficient endurance are remaining reliability issues. Therefore, this paper reviews two major aspects: the advantages of fluorite-structure ferroelectrics for memory applications are reviewed from a material’s point of view, and the critical issues of wake-up effect and insufficient endurance are examined, and potential solutions are subsequently discussed.

The Phase Change Memories (PCM) based on chalcogenide glasses are being considered recently as a possible replacement for conventional Non Volatile Random Access Memories (NVRAM). The main advantages of chalcogenide phase change memories are their direct write/overwrite capability, lower voltages of operation, large write/erase cycles, easiness to integrate with logic, etc. The phase change random access memories work on the principle of memory switching exhibited by chalcogenide glasses during which a local structural change (between amorphous and crystalline states) occurs due to an applied electric field. The development of newer phase change materials for NVRAM applications is based on synthesizing newer glass compositions and investigating their electrical switching characteristics by applying current/voltage pulses of different waveforms. The thermal studies on chalcogenide glasses which provide information about thermal stability, glass forming ability, etc., are also important while selecting a chalcogenide glass for PCM applications. The present thesis work deals with electrical switching and thermal studies on certain silicon based ternary telluride glasses (As-Te-Si, Ge-Te-Si and Al-Te-Si). The effect of network topological thresholds on the composition dependence of switching voltages and thermal parameters such as glass transition temperature, specific heat capacity, non-reversing enthalpy, etc., of these glasses has been investigated. The first chapter of the thesis provides an introduction to various properties of chalcogenide glasses, including their applications in phase change memories. The fundamental aspects of amorphous solids such as glass formation, glass transition, etc., are presented. Further, the concepts of rigidity percolation and self organization in glassy networks and the influence of local structural effects on the properties of glassy chalcogenides are discussed. Also, a brief history of evolution of phase change memories is presented. The second chapter deals with the experimental techniques employed in this thesis work; for sample preparation and for electrical switching studies, Alternating Differential Scanning Calorimetry (ADSC), Raman spectroscopy, NMR spectroscopy, etc. The third chapter discusses the electrical switching and thermal studies on As30Te70-xSix (2 ≤ x ≤ 22) and As40Te60-xSix (2 ≤ x ≤ 17) glasses. The composition dependence of electrical switching voltage (VT) and thermal parameters such as glass transition temperature (Tg), crystallization temperature (Tc), thermal stability (Tc-Tg), etc., reveals the occurrence of extended rigidity percolation and chemical thresholds in As30Te70-xSix and As40Te60-xSix glasses. Chapter 4 presents the electrical switching and thermal studies on Ge15Te85-xSix glasses (2 ≤ x ≤ 12). These glasses have been found to exhibit memory type electrical switching. While Ge15Te85-xSix glasses with x ≤ 5 exhibit a normal electrical switching, an unstable behavior is seen in the I-V characteristics of Ge15Te85-xSix glasses with x > 5 during the transition to ON state. Further, the switching voltage (VT) and initial resistance (R) are found to increase with addition of Si, exhibiting a change in slope at the rigidity percolation threshold of the Ge15Te85-xSix system. The ADSC studies on these glasses indicate the presence of an extended stiffness transition and a thermally reversing window in Ge15Te85-xSix in the composition range of 2 ≤ x ≤ 6. The fifth chapter deals with electrical switching investigations, thermal and structural studies on Al15Te85-xSix glasses (2 ≤ x ≤ 12). These glasses have been found to exhibit two crystallization reactions (Tc1 and Tc2) for compositions with x < 8 and a single stage crystallization is seen for compositions above x = 8. Also, a trough is seen in the composition dependence of non-reversing enthalpy (ΔHNR), based on which it is proposed that there is a thermally reversing window in Al15Te85-xSix glasses in the composition range 4 ≤ x ≤ 8. Further, Al15Te85-xSix glasses are found to exhibit a threshold type electrical switching at ON state currents less than 2 mA. The start and the end of the thermally reversing window seen in the thermal studies are exemplified by a kink and saturation in the composition dependence of switching voltages respectively. 27Al Solid State NMR measurements reveal that in Al15Te85-xSix glasses, Al atoms reside in 4-fold as well as 6-fold coordinated environments. Unlike in Al-As-Te glasses, there is no correlation seen between the composition dependence of the fraction of 4-fold and 6-fold coordinated aluminum atoms and the switching behavior of Al-Te-Si samples. Chapter 6 provides a comparison of the properties of the three glassy systems studied (As-Te-Si, Ge-Te-Si and Al-Te-Si), made to identify the system better suited for phase change memory applications. It is found that the Ge-Te-Si glassy system has better electrical/thermal properties for phase change memory applications. The seventh chapter describes easily reversible SET-RESET processes in Ge15Te83Si2 glass which is a promising candidate for phase change memory applications. This sample exhibits memory switching at a comparatively low threshold electric field (Eth) of 7.3 kV/cm. The SET and RESET processes have been achieved with 1 mA triangular current pulse for the SET process and 1 mA rectangle pulse (of 10 msec width) for RESET operation respectively. Further, a self-resetting effect is seen in this material upon excitation with a saw-tooth/square pulse. About 6.5x104 SET-RESET cycles have been achieved without any damage to the device. In chapter 8, results of in-situ Raman scattering studies on the structural changes occurring during the SET and RESET processes in Ge15Te83Si2 sample, are presented. It is found that the degree of disorder in the glass is reduced from OFF to SET state. The local structure of the sample under RESET condition is similar to that in the OFF state. The Raman results are found to be consistent with the switching results which indicate that the Ge15Te83Si2 glass can be SET and RESET easily. Further, Electron Microscopic studies on switched samples indicate the formation of nanometer sized particles of cSiTe2. A summary of the results obtained and the scope for future work are included in the chapter 9 of the thesis.