Memory Nodes Research Articles

AEROSPACE APPLICATIONS IN NANO SCALE CMOS TECHNOLOGY USING 12T MEMORY CELL

Semiconductor sizes and the gaps between them are decreasing with the advancement of technology. This means that SRAM cells used in aerospace applications are more vulnerable to soft-error when the fundamental charge of the fragile nodes decreases. A single-event upset (SEU) caused by a radiation particle striking a sensitive node in a conventional 6T SRAM cell could lead to data inversion. To lessen the consequences of SEUs, this study proposes the use of a Soft-Error-Aware Read-StabilityEnhanced Low-Power 12T (SARP12T) SRAM cell. The performance of SARP12T is evaluated in relation to other recently released soft-error-aware SRAM cells, such as QUCCE12T, QUATRO12T, RHD12T, RHPD12T, and RSP14T. In the event that a radiation attack flips the values of the sensitive nodes in SARP12T, the data may still be recoverable. When a storage node pair initiates a singleevent multi-node upset (SEMNU), SARP12T is resilient enough to endure it. During read operations, the bitline makes it simple to reach the '0' storing memory nodes in the suggested cell, and it also makes them very resilient to interruptions. In terms of energy usage, SARP12T is the most effective holding strategy. SARP12T outperforms competing cells in terms of write performance, and its write latency is much decreased. To achieve all of these advantages, the suggested cell very slightly increases read latency and read/write energy.

Induction of Progenitor Exhausted Tissue-Resident Memory CD8+ T Cells Upon Salmonella Typhi Porins Adjuvant Immunization Correlates With Melanoma Control and Anti-PD-1 Immunotherapy Cooperation.

Immunotherapy has improved the clinical response in melanoma patients, although a relevant percentage of patients still cannot be salvaged. The search for the immune populations that provide the best tumor control and that can be coaxed by immunotherapy strategies is a hot topic in cancer research nowadays. Tumor-infiltrating TCF-1+ progenitor exhausted CD8+ T cells seem to grant the best melanoma prognosis and also efficiently respond to anti-PD-1 immunotherapy, giving rise to a TIM-3+ terminally exhausted population with heightened effector activity. We tested Porins from Salmonella Typhi as a pathogen associated molecular pattern adjuvant of natural or model antigen in prophylactic and therapeutic immunization approaches against murine melanoma. Porins induced protection against melanomas, even upon re-challenging of tumor-free mice. Porins efficiently expanded IFN-γ-producing CD8+ T cells and induced central and effector memory in lymph nodes and tissue-resident (Trm) T cells in the skin and tumors. Porins induced TCF-1+ PD-1+ CD8+ Trm T cells in the tumor stroma and the presence of this population correlated with melanoma growth protection in mice. Porins immunization also cooperated with anti-PD-1 immunotherapy to hamper melanoma growth. Importantly, the potentially protective Trm populations induced by Porins in the murine model were also observed in melanoma patients in which their presence also correlated with disease control. Our data support the use of cancer vaccination to sculpt the tumor stroma with efficient and lasting Trm T cells with effector activities, highlighting the use of Porins as an adjuvant. Furthermore, our data place CD8+ Trm T cells with a progenitor exhausted phenotype as an important population for melanoma control, either independently or in cooperation with anti-PD-1 immunotherapy.

An Analysis on Task Migration Strategy of Big Data Streaming Storm Computing Framework for Distributed Processing

In this modern era, a large volume of data is generated regularly, which needs to be processed for gaining profits of latent information. The processing of big data is composed of a challenge termed as communication overhead. In order to minimize communication overhead on the premise of various resource constraints, a task migration strategy under heterogeneous storm environment is proposed. The proposed strategy is based on the establishment and demonstration of storm resource constraint model, optimal communication overhead model, and task migration model. Where the source node selection algorithm adds the nodes beyond the threshold to the source node set according to the load of CPU, memory, and network bandwidth of each working node in the cluster and the priority order of various resources. The experiment shows that compared with the existing research, computing storm can effectively reduce the delay and communication overhead between nodes, and the execution overhead is small.

Memory Performance Optimization of DTN Relay Node Based on M/G/1

Delay/Disruption Tolerant Networking is an internetworking architecture for reliable data delivery. It combined with the actual application scenarios of M/G/1 queuing model and based on the data delivery mechanism of store-and-forward and retransmission under Bundle Protocol/Licklider Transmission Protocol in DTN. The data delivery time can be divided into two parts by the last round of LTP data segments transmission process, the custody queue length model of BP data unit (i.e. bundle) in relay nodes memory space is determined by calculating the retransmission time of LTP data segments spent in different delivery parts, which can directly measure the consumption and use of relay nodes memory resources. In order to obtain a more perfect memory resources allocation strategy, the optimized length of LTP data segment and the optimized number of bundles aggregated by per LTP block are proposed as the joint optimization scheme. The simulation results show that in deep-space channel environment where bundle arrival rate and bit error rate constantly change, the joint optimization scheme proposed in this paper can always maintain the shortest average queue length, which means, occupy less memory space. Compared with the queue without optimization, the queue length of the joint optimization scheme is reduced by 76.65%, which demonstrates the superior performance of this scheme, and realizes the reasonable optimization and utilization of the memory resources of the relay nodes in DTN.

Improving dynamic service function chaining classification in NFV/SDN networks through the offloading concept

Service Function Chaining (SFC) paradigm improves network capabilities thanks to the support of application-driven-networking, which is realized through the invocation of an ordered set of Service Functions (SFs). The programmability and flexibility provided by emerging technologies such as Software-Defined Networking (SDN) and Network Function Virtualization (NFV) are perfect features for efficiently managing the lifecycle of SFCs. However, the limitation of Ternary Content Address Memories (TCAMs) of the SDN nodes in SDN-based SFC scenarios can lead to network performance degradation when the SFC classifier is not able to install new classification rules. To tackle the Dynamic Chain Request Classification Offloading (D-CRCO) problem presented in this paper, a hybrid eviction and split-and-distribute approach is proposed, where i) the dynamic behavior of SFC requests is exploited by removing the corresponding idle rules from the flow tables if necessary; and ii) SFC classification is not forced to be carried out by the ingress node, but by any transit node in the domain. An Integer Linear Programming (ILP) formulation and an heuristic are provided to solve D-CRCO, with the goal of maximizing the number of SFC requests that can be served, respecting TCAM size, link capacity, and SF availability constraints. Simulation and emulation results over two real topologies show that the proposed solution is able to significantly increase the number of served SFC requests with a negligible impact on the network performance.

RDON: a rack-scale disaggregated data center network based on a distributed fast optical switch

Aiming to solve the low utilization and high operational cost in current data centers and the diversified resource requirements for different application types, we propose and investigate what we believe to be a novel rack-scale disaggregated optical data center network (RDON). Different types of hardware resources (i.e., CPU, memory, and storage) are deployed in standalone resource pools of the RDON. The RDON architecture encompasses nanosecond distributed fast optical switches for the low latency and high bandwidth interconnects between the computing and memory nodes, while electrical packet switches are utilized to support the traffic of latency insensitivity and low bandwidth between memory nodes and storage nodes. By replacing the onboard data bus with a network interconnection, RDON architecture can provide a fine granularity hardware resource and relieve the constraints of server-centric architecture. To assess the performance of the RDON under realistic scenarios, statistics of diverse applications are captured from the data center lab. The RDON is modeled based on the OMNeT++ discrete event simulator and compared with the server-centric architecture. Numerical results show that the performance of the RDON is closely coupled with the type of deployed application. Compared to the server-centric architecture, the RDON can achieve up to 2.44 × faster running time when running a single application, whereas at most 2.25 × is the fastest running time under coexisting diverse types of applications. Finally, the analysis of costs and power consumption show that, with respect to server-centric architecture, the RDON requires 21.5% extra capital cost but 46.7% less power consumption.

Nanophotonic quantum network node with neutral atoms and an integrated telecom interface

The realization of a long-distance, distributed quantum network based on quantum memory nodes that are linked by photonic channels remains an outstanding challenge. We propose a quantum network node based on neutral alkali atoms coupled to nanophotonic crystal cavities that combines a long-lived memory qubit with a photonic interface at the telecom range, thereby enabling the long-distance distribution of entanglement over low loss optical fibers. We present a novel protocol for the generation of an atom–photon entangled state which uses telecom transitions between excited states of the alkali atoms. We analyze the realistic implementation of this protocol using rubidium and cesium atoms taking into account the full atomic level structure and properties of the nanophotonic crystal cavity. We find that a high fidelity entangled state can be generated with current technologies.

Key pre-distribution scheme based on transversal design in large mobile fog networks with multi-clouds

Assuring the security of the communication channel is a significant issue in the fog networks. Designing an efficient and secure key management scheme with a common key among nodes is a critical factor. The available key distribution schemes of hierarchical networks are not scalable and impose high communication and computation overhead to networks. These schemes are impractical on limited resources’ environments like the internet of things (IoT) networks. In this paper, a hierarchical key pre-distribution scheme based on the transversal design and residual theorem (TD-R) is proposed for large mobile fog networks with multi-clouds. The objective of this paper to reveal that TD-R is the key pre-distribution scheme for establishing secure intra and inter-fog nodes’ communication channels and intra and inter-end nodes with moderate network connectivity. The obtained results indicate that the TD-R scheme increases network scalability in comparison with available schemes, while it decreases memory consumption and node storage overhead.

History of Alcohol Consumption and HIV Status Related to Functional Connectivity Differences in the Brain During Working Memory Performance.

Poorer working memory function has previously been associated with alcohol misuse, Human Immunodeficiency Virus (HIV) positive status, and risky behavior. Poorer working memory performance relates to alterations in specific brain networks. The current study examined if there was a relationship between brain networks involved in working memory and reported level of alcohol consumption during an individual's period of heaviest use. Furthermore, we examined whether HIV status and the interaction between HIV and alcohol consumption was associated with differences in these brain networks. Fifty adults, 26 of whom were HIV positive, engaged in an n-back working memory task (0-back and 2-back trials) administered in a magnetic resonance imaging (MRI) scanner. The Kreek- McHugh-Schluger-Kellogg (KMSK) scale of alcohol consumption was used to characterize an individual's period of heaviest use and correlates well with their risk for alcohol dependence. Connectivity analyses were conducted using data collected during n-back task. Functional connectivity differences associated with greater alcohol consumption included negative connectivity, primarily from parietal attention networks to frontal networks. Greater alcohol consumption was also associated with positive connectivity from working memory nodes to the precuneus and paracingulate. HIV positive status was associated with more nodes of negative functional connectivity relative to alcohol consumption history alone, particularly in the frontoparietal networks. The HIV positive individuals with heavier drinking history related to negative fronto-parietal connectivity, along with positive connectivity from working memory nodes to mesolimbic regions. Findings allow for a better understanding of brain networks affected by HIV and alcohol and may provide avenues for interventions.

A novel approach for energy- and memory-efficient data loss prevention to support Internet of Things networks

Internet of Things integrates various technologies, including wireless sensor networks, edge computing, and cloud computing, to support a wide range of applications such as environmental monitoring and disaster surveillance. In these types of applications, IoT devices operate using limited resources in terms of battery, communication bandwidth, processing, and memory capacities. In this context, load balancing, fault tolerance, and energy and memory efficiency are among the most important issues related to data dissemination in IoT networks. In order to successfully cope with the abovementioned issues, two main approaches—data-centric storage and distributed data storage—have been proposed in the literature. Both approaches suffer from data loss due to memory and/or energy depletion in the storage nodes. Even though several techniques have been proposed so far to overcome the abovementioned problems, the proposed solutions typically focus on one issue at a time. In this article, we propose a cross-layer optimization approach to increase memory and energy efficiency as well as support load balancing. The optimization problem is a mixed-integer nonlinear programming problem, and we solve it using a genetic algorithm. Moreover, we integrate the data-centric storage features into distributed data storage mechanisms and present a novel heuristic approach, denoted as Collaborative Memory and Energy Management, to solve the underlying optimization problem. We also propose analytical and simulation frameworks for performance evaluation. Our results show that the proposed method outperforms the existing approaches in various IoT scenarios.

Low-Complexity Security Algorithm Again Black-hole Attacking Node Considering Energy Balance for WSNs

Wireless sensor network (WSN) is a network through radio waves to link a large number of unevenly distributed nodes on a wide range of areas for the purpose of sensing, processing and collecting data.The data transmitted in WSNs is usually sensitive that needs to be protected. WSNs are more vulnerable to attacks due to network characteristics such as wireless transmission, changing network topologies, computing power, limited memory and power of nodes comparing to wired networks. There are many types of attacks on WSNs such as sinkhole attacks, data integrity attacks, wormhole attacks, Black hole attacks, etc. In this article, we will evaluate the impact of Black-hole attacks in the WSN network when using the AODV routing protocol (Ad-hoc On-demand Distance Vector routing). The Black hole attack model will be simulated by Network Simulator 2 (NS-2) by the number of lost packets by counting the number of packets sent by the sending node, receiving packets to destination, and rest energy of the nodes in two cases.

Development of a method for selecting the approximatimg functions for the observable processes of cloud infrastructure

This paper considers the techniques for improving the effectiveness of monitoring the cloud infrastructure processes implying the reduction of a computational burden while maintaining the required level of measurement accuracy. A technique for organizing the monitoring of cloud infrastructure processes, based on the approximation of accumulated measurements, has been further developed in this study. The necessary and sufficient set of approximating functions has been built, corresponding to the key properties of the observable processes. A method for selecting the approximating functions for the observable cloud infrastructure processes has been constructed. The method implies the assessment of properties of an observable process and the selection of its approximating function. The practical value of this research relates to the ability to reduce the computational burden by reducing the number of planned measurements at an acceptable level of the decrease in their accuracy. The originality of the approach is the use of the a priori data about the observable processes aimed to obtain more accurate estimates of their properties. The practical implementation of the proposed method shows a 20–40 % decrease in the number of planned measurements at the level of monitoring accuracy not lower than 95 %. The proposed method makes it possible to reduce the load on cloud infrastructure components, to decrease the use of processor time, as well as the disk and random-access memories of physical and virtual nodes. The results of the study can be used for the software implementation of the system of cloud infrastructure monitoring.

EARSHOT: A Minimal Neural Network Model of Incremental Human Speech Recognition.

Despite the lack of invariance problem (the many-to-many mapping between acoustics and percepts), human listeners experience phonetic constancy and typically perceive what a speaker intends. Most models of human speech recognition (HSR) have side-stepped this problem, working with abstract, idealized inputs and deferring the challenge of working with real speech. In contrast, carefully engineered deep learning networks allow robust, real-world automatic speech recognition (ASR). However, the complexities of deep learning architectures and training regimens make it difficult to use them to provide direct insights into mechanisms that may support HSR. In this brief article, we report preliminary results from a two-layer network that borrows one element from ASR, long short-term memory nodes, which provide dynamic memory for a range of temporal spans. This allows the model to learn to map real speech from multiple talkers to semantic targets with high accuracy, with human-like timecourse of lexical access and phonological competition. Internal representations emerge that resemble phonetically organized responses in human superior temporal gyrus, suggesting that the model develops a distributed phonological code despite no explicit training on phonetic or phonemic targets. The ability to work with real speech is a major advance for cognitive models of HSR.

Experimental demonstration of memory-enhanced quantum communication.

The ability to communicate quantum information over long distances is of central importance in quantum science and engineering1. Although some applications of quantum communication such as secure quantum key distribution2,3 are already being successfully deployed4-7, their range is currently limited by photon losses and cannot be extended using straightforward measure-and-repeat strategies without compromising unconditional security8. Alternatively, quantum repeaters9, which utilize intermediate quantum memory nodes and error correction techniques, can extend the range of quantum channels. However, their implementation remains an outstanding challenge10-16, requiring a combination of efficient and high-fidelity quantum memories, gate operations, and measurements. Here we use a single solid-state spin memory integrated in a nanophotonic diamond resonator17-19 to implement asynchronous photonic Bell-state measurements, which are a key component of quantum repeaters. In a proof-of-principle experiment, we demonstrate high-fidelity operation that effectively enables quantum communication at a rate that surpasses the ideal loss-equivalent direct-transmission method while operating at megahertz clock speeds. These results represent a crucial step towards practical quantum repeaters and large-scale quantum networks20,21.

Cybersecurity Issues in Wireless Sensor Networks: Current Challenges and Solutions

Wireless sensor networks are deployed without predefined infrastructure and are generally left unattended. Indeed, the vulnerability of the wireless sensor networks to attacks comes principally from their inherent characteristics. As the data are transmitted over the air, it is very easy for an adversary to spy on traffic. Also, to meet the strict budgetary requirements, the sensor nodes tend to not be tamperproof and thus offer no protection against security attacks. Alongside with these vulnerabilities, the human intervention is always not allowed to deal with adversaries who attempt to compromise the network. Therefore, security systems are mainly needed to secure the network and ensure the protection against security threats. Indeed, cryptographic based systems are generally used to ensure security. However, due to the lack of memory and power (low computing, limited energy reserves) of the sensor nodes, most of these approaches are not suitable. Therefore, providing security while respecting the specific constraints of the sensors, represents one of the most important research issue in wireless sensor networks. Indeed, several studies have been conducted these last decades to propose lightweight and efficient security protocols for wireless sensor networks. In this paper, we review the most leading protocols and classify them based the addressed security issue. Also, we outline the main security constraints and challenges and present the future research directions based on the emerged application fields.

Hotness- and Lifetime-Aware Data Placement and Migration for High-Performance Deep Learning on Heterogeneous Memory Systems

Heterogeneous memory systems that comprise memory nodes with disparate architectural characteristics (e.g., DRAM and high-bandwidth memory (HBM)) have surfaced as a promising solution in a variety of computing domains ranging from embedded to high-performance computing. Since deep learning (DL) is one of the most widely-used workloads in various computing domains, it is crucial to explore efficient memory management techniques for DL applications that execute on heterogeneous memory systems. Despite extensive prior works on system software and architectural support for efficient DL, it still remains unexplored to investigate heterogeneity-aware memory management techniques for high-performance DL on heterogeneous memory systems. To bridge this gap, we analyze the characteristics of representative DL workloads on a real heterogeneous memory system. Guided by the characterization results, we propose HALO, hotness- and lifetime-aware data placement and migration for high-performance DL on heterogeneous memory systems. Through quantitative evaluation, we demonstrate the effectiveness of HALO in that it significantly outperforms various memory management policies (e.g., 28.2 percent higher performance than the HBM-Preferred policy) supported by the underlying system software and hardware, achieves the performance comparable to the ideal case with infinite HBM, incurs small performance overheads, and delivers high performance across a wide range of application working-set sizes.

An Enhanced Hybrid Routing and Clustering Technique for Wireless Sensor Network

Wireless Sensor Networks (WSN) have extensively deployed in a wide range of applications. However, WSN still faces several limitations in processing capabilities, memory, and power supply of sensor nodes. It is required to extend the lifetime of WSN. Mainly this is achieved by routing protocols choosing the best transmission path in-network with desired power conservation.This cause is developing a generic protocol framework for WSNa big challenge. This work proposed a new routing technique, described as Hybrid Routing-Clustering (HRC) model. This new approach takes advantage of clustering and routing procedures defined in K-Mean clustering and AODV routing, which constituted of three phases. This development aims to achieve enhanced power conservation rate in consequence network lifetime. An extensive evaluation methodology utilized to measure the performance of the proposed model in simulated scenarios.The results categorized in terms of the average amount of packet received and power conservation rate. The Hybrid Routing-Clustering (HRC) model was determined, showed enhanced results regarding both parameters. In the end, they are comparing these results with well-known routing and well-known clustering algorithms.

Photonic Switched Optically Connected Memory: An Approach to Address Memory Challenges in Deep Learning

Deep learning has been revolutionizing many aspects of our society, powering various fields including computer vision, natural language processing, and activity recognition. However, the scaling trends for both datasets and model size are constraining system performance. Variability of memory requirements can lead to poor resource utilization. Reconfigurable photonic interconnects provide scalable solutions and enable efficient use of disaggregated memory resources. We propose a photonic switched optically connected memory system architecture that tackles the memory challenges while showing the functionality of optical switching for deep learning models. Our proposed system architecture utilizes a “lite” (de)serialization scheme for memory transfers via optical links to avoid network overheads and supports the dynamic allocation of remote memories to local processing systems. In order to test the feasibility of our proposal, we built an experimental testbed with a processing system and two remote memory nodes using silicon photonic switch fabrics and evaluated the system performance. The optical switching time is measured to be 119 μs and an overall 2.78 ms latency is achieved for the end-to-end reconfiguration. The collective results and existing high-bandwidth optical I/Os show the potential of integrating the photonic switched optically connected memory to state-of-the-art processing systems.

Hierarchical routing protocols for wireless sensor network: a compressive survey

Wireless sensor networks (WSNs) are one of the key enabling technologies for the internet of things (IoT). WSNs play a major role in data communications in applications such as home, health care, environmental monitoring, smart grids, and transportation. WSNs are used in IoT applications and should be secured and energy efficient in order to provide highly reliable data communications. Because of the constraints of energy, memory and computational power of the WSN nodes, clustering algorithms are considered as energy efficient approaches for resource-constrained WSNs. In this paper, we present a survey of the state-of-the-art routing techniques in WSNs. We first present the most relevant previous work in routing protocols surveys then highlight our contribution. Next, we outline the background, robustness criteria, and constraints of WSNs. This is followed by a survey of different WSN routing techniques. Routing techniques are generally classified as flat, hierarchical, and location-based routing. This survey focuses on the deep analysis of WSN hierarchical routing protocols. We further classify hierarchical protocols based on their routing techniques. We carefully choose the most relevant state-of-the-art protocols in order to compare and highlight the advantages, disadvantage and performance issues of each routing technique. Finally, we conclude this survey by presenting a comprehensive survey of the recent improvements of low-energy adaptive clustering hierarchy routing protocols and a comparison of the different versions presented in the literature.

A Novel B-Tree Index with Cascade Memory Nodes for Improving Sequential Write Performance on Flash Storage Devices

Flash storage devices such as solid-state drives and multimedia cards have been widely used in various applications because of their fast access speed, low power consumption, and high reliability. They consist of NAND flash memories that perform slow block erasures before overwriting data on a prewritten page. This characteristic can lead to performance degradation when applying the original B-tree on the flash storage device without any changes. Although various B-trees have been proposed for flash memory, they still require many flash operations that degrade overall performance. To address the problem, we propose a novel B-tree index structure that reduces the number of write operations and improves the sequential writes by employing cascade memory nodes. The proposed B-tree index structure delays the updates for the modified B-tree nodes and later performs batch writes in a cascade manner. Also, when records with continuous key values are sequentially inserted, the proposed B-tree index structure does not split the leaf node so that it improves write throughput and page utilization. Through mathematical analysis and experimental results, we show that the proposed B-tree index structure always yields better performance than existing techniques.