AMC: an adaptive multi-level cache algorithm in hybrid storage systems

Hybrid storage systems that consist of flash-based solid state drives (SSDs) and traditional disks are now widely used. In hybrid storage systems, there exists a two-level cache hierarchy that regard dynamic random access memory (DRAM) as the first level cache and SSD as the second level cache for disk storage. However, this two-level cache hierarchy typically uses independent cache replacement policies for each level, which makes cache resource management inefficient and reduces system performance. In this paper, we propose a novel adaptive multi-level cache (AMC) replacement algorithm in hybrid storage systems. The AMC algorithm adaptively adjusts cache blocks between DRAM and SSD cache levels using an integrated solution. AMC uses combined selective promote and demote operations to dynamically determine the level in which the blocks are to be cached. In this manner, the AMC algorithm achieves multi-level cache exclusiveness and makes cache resource management more efficient. By using real-life storage traces, our evaluation shows the proposed algorithm improves hybrid multi-level cache performance and also increases the SSD lifetime compared with traditional multi-level cache replacement algorithms.

Electric and dielectric behavior of CaCu3Ti4O12-based thin films obtained by soft chemical method

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Dielectric and fatigue properties of Pb(Zr0.53Ti0.47)O-3 thin films prepared from oxide precursors method

Fatigue is an important problem to be considered if a ferroelectric film is used for non-volatile memory devices. In this phenomena, the remanent polarization and coercive field properties degrades in cycles which increase in hysteresis loops. The reasons have been attributed to different mechanisms such as a large voltage applied on ferroelectric film in every reading process in Ferroelectric Random Access Memory (FeRAM) or memories for digital storage in computer, grain size effects and others. The aim of this work is to investigate the influence of the crystallization kinetics on dielectric and ferroelectric properties of the Pb(Zr0.53Ti0.47)O-3 thin films prepared by an alternative chemical method. Films were crystallized in air on Pt/Ti/SiO2/Si substrates at 700 degrees C for 1 hour, in conventional thermal annealing (CTA), and at 700 degrees C for 1 min and 700 degrees C 5 min, using a rapid thermal annealing (RTA) process. Final films were crack free and presented an average of 750 nm in thickness. Dielectric properties were studied in the frequency range of 100 Hz - 1 MHz. All films showed a dielectric dispersion at low frequency. Ferroelectric properties were measured from hysteresis loops at 10 kHz. The obtained remanent polarization (P-r) and coercive field (E-c) were 3.7 mu C/cm(2) and 71.9 kV/cm respectively for film crystallized by CTA while in films crystallized by RTA these parameters were essentially the same. In the fatigue process, the P, value decreased to 14% from the initial value after 1.3 x 10(9) switching cycles, for film by CTA, while for film crystallized by RTA for 5 min, P, decreased to 47% from initial value after 1.7 x 10(9) switching cycles.

An active leakage-injection scheme applied to low-voltage SRAMs

An active leakage-injection scheme (ALIS) for low-voltage (LV) high-density (HD) SRAMs is presented. By means of a feedback loop comprising a servo-amplifier and a common-drain MOSFET, a current matching the respective bit-line leakage is injected onto the line during precharge and sensing, preventing the respective capacitances from erroneous discharges. The technique is able to handle leakages up to hundreds of μA at high operating temperatures. Since no additional timing is required, read-out operations are performed at no speed penalty. A simplified 256×1bit array was designed in accordance with a 0.35 CMOS process and 1.2V-supply. A range of PSPICE simulation attests the efficacy of ALIS. With an extra power consumption of 242 μW, a 200 μA-leakage @125°C, corresponding to 13.6 times the cell current, is compensated.

Microscopic and dielectric analyses of vanadium and tungsten modified barium zirconium titanate ceramics

Dielectric spectroscopy was used in this study to examine polycrystalline vanadium and tungstendoped BaZr 0.1Ti 0.90O 3 (BZT10:2V and BZT10:2W) ceramics obtained by the mixed oxide method. According to X-ray diffraction analyses, addition of vanadium and tungsten lead to ceramics free of secondary phases. SEM analyses reveal that both dopants result in slower oxygen ion motion and consequently lower grain growth rate. Temperature dependence dielectric study showed normal ferroelectric to paraelectric transition well above the room temperature for the BZT10 and BZT10:2V ceramics. However, BZT10:2W ceramic showed a relaxor-like behavior near phase transition characterized by the empirical parameter γ. Piezoelectric force microscopy images reveals that the piezoelectric coefficient is strongly influenced by type of donor dopant suggesting promising applications for dynamic random access memories and data-storage media. Copyright © 2010 American Scientific Publishers All rights reserved.

Preparação e caracterização de filmes finos de 'BA''BI IND.2''TA IND.2''O IND.9(BBT)' e 'BA''BI IND.2''NB IND.2''O IND.9(BBN)'

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

A Novel Fiber Delay Line Buffering Architecture for Optical Packet Switching

Due to the lack of optical random access memory, optical fiber delay line (FDL) is currently the only way to implement optical buffering. Feed-forward and feedback are two kinds of FDL structures in optical buffering. Both have advantages and disadvantages. In this paper, we propose a more effective hybrid FDL architecture that combines the merits of both schemes. The core of this switch is the arrayed waveguide grating (AWG) and the tunable wavelength converter (TWC). It requires smaller optical device sizes and fewer wavelengths and has less noise than feedback architecture. At the same time, it can facilitate preemptive priority routing which feed-forward architecture cannot support. Our numerical results show that the new switch architecture significantly reduces packet loss probability.

Junction Field Effect on the Retention Time for One-Transistor Floating-Body RAM

One-transistor floating-body random access memory retention time distribution is investigated on silicon-on-insulator UTBOX devices. It is shown that the average retention time can be improved by two to three orders of magnitude by reducing the body-junction electric field. However, the retention time distribution, which is mainly caused by the generation-recombination center density variation, remains similar.

Dependence of Generation-Recombination Noise With Gate Voltage in FD SOI MOSFETs

A model for computing the generation-recombination noise due to traps within the semiconductor film of fully depleted silicon-on-insulator MOSFET transistors is presented. Dependence of the corner frequency of the Lorentzian spectra on the gate voltage is addressed in this paper, which is different to the constant behavior expected for bulk transistors. The shift in the corner frequency makes the characterization process easier. It helps to identify the energy position, capture cross sections, and densities of the traps. This characterization task is carried out considering noise measurements of two different candidate structures for single-transistor dynamic random access memory devices.

Magnetization dynamics in spin valves

Key technology applications like magnetoresistive sensors or the Magnetic Random Access Memory (MRAM) require reproducible magnetic switching mechanisms. i.e. predefined remanent states. At the same time advanced magnetic recording schemes push the magnetic switching time into the gyromagnetic regime. According to the Landau-Lifschitz-Gilbert formalism, relevant questions herein are associated with magnetic excitations (eigenmodes) and damping processes in confined magnetic thin film structures.rnObjects of study in this thesis are antiparallel pinned synthetic spin valves as they are extensively used as read heads in today’s magnetic storage devices. In such devices a ferromagnetic layer of high coercivity is stabilized via an exchange bias field by an antiferromagnet. A second hard magnetic layer, separated by a non-magnetic spacer of defined thickness, aligns antiparallel to the first. The orientation of the magnetization vector in the third ferromagnetic NiFe layer of low coercivity - the freelayer - is then sensed by the Giant MagnetoResistance (GMR) effect. This thesis reports results of element specific Time Resolved Photo-Emission Electron Microscopy (TR-PEEM) to image the magnetization dynamics of the free layer alone via X-ray Circular Dichroism (XMCD) at the Ni-L3 X-ray absorption edge.rnThe ferromagnetic systems, i.e. micron-sized spin valve stacks of typically deltaR/R = 15% and Permalloy single layers, were deposited onto the pulse leading centre stripe of coplanar wave guides, built in thin film wafer technology. The ferromagnetic platelets have been applied with varying geometry (rectangles, ellipses and squares), lateral dimension (in the range of several micrometers) and orientation to the magnetic field pulse to study the magnetization behaviour in dependence of these magnitudes. The observation of magnetic switching processes in the gigahertz range became only possible due to the joined effort of producing ultra-short X-ray pulses at the synchrotron source BESSY II (operated in the so-called low-alpha mode) and optimizing the wave guide design of the samples for high frequency electromagnetic excitation (FWHM typically several 100 ps). Space and time resolution of the experiment could be reduced to d = 100 nm and deltat = 15 ps, respectively.rnIn conclusion, it could be shown that the magnetization dynamics of the free layer of a synthetic GMR spin valve stack deviates significantly from a simple phase coherent rotation. In fact, the dynamic response of the free layer is a superposition of an averaged critically damped precessional motion and localized higher order spin wave modes. In a square platelet a standing spin wave with a period of 600 ps (1.7 GHz) was observed. At a first glance, the damping coefficient was found to be independent of the shape of the spin-valve element, thus favouring the model of homogeneous rotation and damping. Only by building the difference in the magnetic rotation between the central region and the outer rim of the platelet, the spin wave becomes visible. As they provide an additional efficient channel for energy dissipation, spin waves contribute to a higher effective damping coefficient (alpha = 0.01). Damping and magnetic switching behaviour in spin valves thus depend on the geometry of the element. Micromagnetic simulations reproduce the observed higher-order spin wave mode.rnBesides the short-run behaviour of the magnetization of spin valves Permalloy single layers with thicknesses ranging from 3 to 40 nm have been studied. The phase velocity of a spin wave in a 3 nm thick ellipse could be determined to 8.100 m/s. In a rectangular structure exhibiting a Landau-Lifschitz like domain pattern, the speed of the field pulse induced displacement of a 90°-Néel wall has been determined to 15.000 m/s.rn

Coded slotted aloha with multipacket reception over block fading channels (for satellite networks)

Random access (RA) protocols are normally used in a satellite networks for initial terminal access and are particularly effective since no coordination is required. On the other hand, contention resolution diversity slotted Aloha (CRDSA), irregular repetition slotted Aloha (IRSA) and coded slotted Aloha (CSA) has shown to be more efficient than classic RA schemes as slotted Aloha, and can be exploited also when short packets transmissions are done over a shared medium. In particular, they relies on burst repetition and on successive interference cancellation (SIC) applied at the receiver. The SIC process can be well described using a bipartite graph representation and exploiting tools used for analyze iterative decoding. The scope of my Master Thesis has been to described the performance of such RA protocols when the Rayleigh fading is taken into account. In this context, each user has the ability to correctly decode a packet also in presence of collision and when SIC is considered this may result in multi-packet reception. Analysis of the SIC procedure under Rayleigh fading has been analytically derived for the asymptotic case (infinite frame length), helping the analysis of both throughput and packet loss rates. An upper bound of the achievable performance has been analytically obtained. It can be show that in particular channel conditions the throughput of the system can be greater than one packets per slot which is the theoretical limit of the Collision Channel case.

Sophisticated security verification on routing repaired balanced cell-based dual-rail logic against side channel analysis

Conventional dual-rail precharge logic suffers from difficult implementations of dual-rail structure for obtaining strict compensation between the counterpart rails. As a light-weight and high-speed dual-rail style, balanced cell-based dual-rail logic (BCDL) uses synchronised compound gates with global precharge signal to provide high resistance against differential power or electromagnetic analyses. BCDL can be realised from generic field programmable gate array (FPGA) design flows with constraints. However, routings still exist as concerns because of the deficient flexibility on routing control, which unfavourably results in bias between complementary nets in security-sensitive parts. In this article, based on a routing repair technique, novel verifications towards routing effect are presented. An 8 bit simplified advanced encryption processing (AES)-co-processor is executed that is constructed on block random access memory (RAM)-based BCDL in Xilinx Virtex-5 FPGAs. Since imbalanced routing are major defects in BCDL, the authors can rule out other influences and fairly quantify the security variants. A series of asymptotic correlation electromagnetic (EM) analyses are launched towards a group of circuits with consecutive routing schemes to be able to verify routing impact on side channel analyses. After repairing the non-identical routings, Mutual information analyses are executed to further validate the concrete security increase obtained from identical routing pairs in BCDL.

Computación celular con membranas: un estudio de modelos de cómputo bioinspirados

Los resultados presentados en la memoria de esta tesis doctoral se enmarcan en la denominada computación celular con membranas una nueva rama de investigación dentro de la computación natural creada por Gh. Paun en 1998, de ahí que habitualmente reciba el nombre de sistemas P. Este nuevo modelo de cómputo distribuido está inspirado en la estructura y funcionamiento de la célula. El objetivo de esta tesis ha sido analizar el poder y la eficiencia computacional de estos sistemas de computación celular. En concreto, se han analizado dos tipos de sistemas P: por un lado los sistemas P de neuronas de impulsos, y por otro los sistemas P con proteínas en las membranas. Para el primer tipo, los resultados obtenidos demuestran que es posible que estos sistemas mantengan su universalidad aunque muchas de sus características se limiten o incluso se eliminen. Para el segundo tipo, se analiza la eficiencia computacional y se demuestra que son capaces de resolver problemas de la clase de complejidad ESPACIO-P (PSPACE) en tiempo polinómico. Análisis del poder computacional: Los sistemas P de neuronas de impulsos (en adelante SN P, acrónimo procedente del inglés «Spiking Neural P Systems») son sistemas inspirados en el funcionamiento neuronal y en la forma en la que los impulsos se propagan por las redes sinápticas. Los SN P bio-inpirados poseen un numeroso abanico de características que ha cen que dichos sistemas sean universales y por tanto equivalentes, en poder computacional, a una máquina de Turing. Estos sistemas son potentes a nivel computacional, pero tal y como se definen incorporan numerosas características, quizás demasiadas. En (Ibarra et al. 2007) se demostró que en estos sistemas sus funcionalidades podrían ser limitadas sin comprometer su universalidad. Los resultados presentados en esta memoria son continuistas con la línea de trabajo de (Ibarra et al. 2007) y aportan nuevas formas normales. Esto es, nuevas variantes simplificadas de los sistemas SN P con un conjunto mínimo de funcionalidades pero que mantienen su poder computacional universal. Análisis de la eficiencia computacional: En esta tesis se ha estudiado la eficiencia computacional de los denominados sistemas P con proteínas en las membranas. Se muestra que este modelo de cómputo es equivalente a las máquinas de acceso aleatorio paralelas (PRAM) o a las máquinas de Turing alterantes ya que se demuestra que un sistema P con proteínas, es capaz de resolver un problema ESPACIOP-Completo como el QSAT(problema de satisfacibilidad de fórmulas lógicas cuantificado) en tiempo polinómico. Esta variante de sistemas P con proteínas es muy eficiente gracias al poder de las proteínas a la hora de catalizar los procesos de comunicación intercelulares. ABSTRACT The results presented at this thesis belong to membrane computing a new research branch inside of Natural computing. This new branch was created by Gh. Paun on 1998, hence usually receives the name of P Systems. This new distributed computing model is inspired on structure and functioning of cell. The aim of this thesis is to analyze the efficiency and computational power of these computational cellular systems. Specifically there have been analyzed two different classes of P systems. On the one hand it has been analyzed the Neural Spiking P Systems, and on the other hand it has been analyzed the P systems with proteins on membranes. For the first class it is shown that it is possible to reduce or restrict the characteristics of these kind of systems without loss of computational power. For the second class it is analyzed the computational efficiency solving on polynomial time PSACE problems. Computational Power Analysis: The spiking neural P systems (SN P in short) are systems inspired by the way of neural cells operate sending spikes through the synaptic networks. The bio-inspired SN Ps possess a large range of features that make these systems to be universal and therefore equivalent in computational power to a Turing machine. Such systems are computationally powerful, but by definition they incorporate a lot of features, perhaps too much. In (Ibarra et al. in 2007) it was shown that their functionality may be limited without compromising its universality. The results presented herein continue the (Ibarra et al. 2007) line of work providing new formal forms. That is, new SN P simplified variants with a minimum set of functionalities but keeping the universal computational power. Computational Efficiency Analisys: In this thesis we study the computational efficiency of P systems with proteins on membranes. We show that this computational model is equivalent to parallel random access machine (PRAM) or alternating Turing machine because, we show P Systems with proteins can solve a PSPACE-Complete problem as QSAT (Quantified Propositional Satisfiability Problem) on polynomial time. This variant of P Systems with proteins is very efficient thanks to computational power of proteins to catalyze inter-cellular communication processes.

Improved Techniques for Video Compression and Communication

Thesis (Ph.D.)--University of Washington, 2016-06

Efficiency improvements to the Groupe Special Mobile (GSM) digital mobile radio system

Groupe Spécial Mobile (GSM) has been developed as the pan-European second generation of digital mobile systems. GSM operates in the 900 MHz frequency band and employs digital technology instead of the analogue technology of its predecessors. Digital technology enables the GSM system to operate in much smaller zones in comparison with the analogue systems. The GSM system will offer greater roaming facilities to its subscribers, extended throughout the countries that have installed the system. The GSM system could be seen as a further enhancement to European integration. GSM has adopted a contention-based protocol for multipoint-to-point transmission. In particular, the slotted-ALOHA medium access protocol is used to coordinate the transmission of the channel request messages between the scattered mobile stations. Collision still happens when more than one mobile station having the same random reference number attempts to transmit on the same time-slot. In this research, a modified version of this protocol has been developed in order to reduce the number of collisions and hence increase the random access channel throughput compared to the existing protocol. The performance evaluation of the protocol has been carried out using simulation methods. Due to the growing demand for mobile radio telephony as well as for data services, optimal usage of the scarce availability radio spectrum is becoming increasingly important. In this research, a protocol has been developed whereby the number of transmitted information packets over the GSM system is increased without any additional increase of the allocated radio spectrum. Simulation results are presented to show the improvements achieved by the proposed protocol. Cellular mobile radio networks commonly respond to an increase in the service demand by using smaller coverage areas. As a result, the volume of the signalling exchanges increases. In this research, a proposal for interconnecting the various entitles of the mobile radio network over the future broadband networks based on the IEEE 802.6 Metropolitan Area Network (MAN) is outlined. Simulation results are presented to show the benefits achieved by interconnecting these entities over the broadband Networks.