The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The evolution of convolutional neural networks (CNNs) into more complex forms of organization, with additional layers, larger convolutions and increasing connections, established the state-of-the-art in terms of accuracy errors for detection and classification challenges in images. Moreover, as they evolved to a point where Gigabytes of memory are required for their operation, we have reached a stage...
DRAM is a crucial component in computing systems, and is expected to be even more important as data-intensive applications become more prominent. A key challenge in advancing DRAM technology is the growing cost of refresh operations, which can impose a large impact on the energy efficiency of DRAM modules. Existing refresh mitigation techniques all require hardware modifications, which may be undesirable...
Recently it was demonstrated that an asymmetric DRAM capacitor stack can introduce non-volatility and at the same time outperform ferroelectric HfO2 based FeRAM in terms of cycle endurance. With the present work, we provide an in-depth study of the underlying mechanisms and perform a comprehensive retention study that characterizes ferroelectric memories. Piezoelectric force microscopy is applied...
Social media networks as well as online graph analytics operate on large-scale graphs with millions of vertices, even billions in some cases. Low-latency access is essential, but caching suffers from the mostly irregular access patterns of the aforementioned application domains. Hence, distributed in-memory systems are proposed keeping all data always in memory. But, the sheer amount of small data...
Recent studies showed that DRAM restore time degrades as technology scales, which imposes large performance and energy overheads. This problem, prolonged restore time (PRT), has been identified by the DRAM industry as one of three major scaling challenges.This paper proposes DrMP, a novel fine-grained precision-aware DRAM restore scheduling approach, to mitigate PRT. The approach exploits process...
This work proposes a fault diagnosis methodology intended for resistive memories. Multiple cells of a block may suffer from stuck-at faults during repetitive write operations in resistive memory. It causes serious hindrance to large scale adoption of resistive memory. This issue is addressed through introduction of von Neumann's Cellular Automata (CA). The two single length cycle attractor CA (TACA)...
This paper introduces test solutions for Integrated Fan Out Wafer Level Chip Scale Packaging (InFO WLCSP) which has the promising of being a very cost effective solution to achieve “More than Moore's law” for mobile devices — more so than 3D integrated circuits (3DIC). InFO WLCSP has its own unique test challenges and it is important to have high-quality screen methodology for severe defects and potential...
This paper presents the state-of-the-art of integrated switched capacitor DC-DC converters in memory technology from NMOS to SOI DRAM and from 2D to 3D NAND focusing on capacitor structures with high capacitance density and small parasitic capacitance based on memory technology. The paper then proposes design for manufacturing to increase yield with redundant arrays of converter and design for reliability...
This paper explores the feasibility, in terms of performance and reliability, of gain-cell embedded DRAM (eDRAM) to be operative at sub-threshold range, when they are implemented with 10nm FinFET devices. The use of individual transistor resizing in order to achieve better cell performance (i.e. retention time, access time, and energy consumption) at the sub-VT operating level is studied. In this...
The spin transfer torque magnetic random access (STT-MRAM) is suitable for embedded memories and also for the second level cache memory in the mobile CPU's. The most capable NVM component is STT-MRAM, which enhances the performance by 3.3 nS access time. It has strong radiation hardness, higher integrity and maximum endurance compared to SRAM. The power consumption of STT-MRAM is decreased by an order...
Commodity HardWare (CHW) is currently used in the Internet to deploy large data centers or small computing nodes. Moreover, CHW will be also used to deploy future telecommunication networks, thanks to the adoption of the forthcoming network softwarization paradigm. In this context, CHW machines can be put in Active Mode (AM) or in Sleep Mode (SM) several times per day, based on the traffic requirements...
Technology scaling trends led to an aggressive increase of SRAM cell sensitivity to process variation. More accurate and non-invasive methodologies must be considered to provide detailed bit-cell characterization focusing on writability margin. This work proposes a new metric using regular read/write operations based on the estimation of the maximum word-line voltage drop during write operations....
The application of tool integration in an industrial use case in the context of the CRYSTAL European project is presented. In particular, the development of a connecting application conceived to provide interoperability between two tools used in the Model-Based design and analysis of aircraft systems is described. The main goal behind this activity is allowing different specialists to use several...
Spaceborne detection can obtain more direct, comprehensive and accurate astronomical data. Providing massive storage for spaceborne applications is the urgent requirement of applying Big Data technologies to advance astronomy research. Spacecraft storage must have the good balance between high performances, redundancy and low power consumption. Deepening inside the Existing research, it is difficult...
For many big data applications, non-volatile memory (NVM) can be utilized to store and process the data at faster rate due to its high-performance, scalable technology, DRAM-like interface and low energy requirement. NVMs such as phase change memory and memristor allow the applications to store persistent data directly in memory, and avoid data serialization and deserialization. However, NVM, like...
This paper presents a novel autonomous take-off and landing approach which uses a discrete-time finite-horizon suboptimal nonlinear control strategy to achieve a smooth landing on specific colored platform. By color-based tracking we can improve the performance of the position landing by correcting the position given by the GPS. The signal of the GPS is not always reliable since the accuracy strongly...
In cyber-physical systems, such as modern industrial plants, complex software is an essential part that enables cost-effective and flexible operation. However, this complexity increases the probability of problems that only reveal themselves after the deployment. This is even more important if security aspects are involved. Therefore, providing the possibility for software updates is an important...
Software aging is a cumulative process which can lead long-running systems to failures. Software rejuvenation is the countermeasure of software aging. Rejuvenation actions usually comprise system reboot or application restart. Software aging appears as a significant concern for the reliability of software systems. Therefore, techniques and methods to analyze software aging and rejuvenation are of...
Physical Unclonable Functions are promising candidates for lightweight authentication applications as they are hard to predict and clone. PUFs are dependent on process variations that occurs during silicon chip fabrication. As the CMOS technology scales down towards nanoscale dimensions, there are increasing transistor reliability challenges which impact the lifetime of integrated circuits. These...
Today, applications like data center/cloud, mobility and Internet of Things (IoT) are key market drivers for semiconductor industry. To meet the requirements of next generation Information and Communication Technology (ICT) systems, the packaging technology has to evolve along with the Integrated Circuit (IC) technology scaling. At the same time, design and development of packages have to meet the...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.