cuGimli: optimized implementation of the Gimli authenticated encryption and hash function on GPU for IoT applications

Recently, National Institute of Standards and Technology (NIST) in the U.S. had initiated a global-scale competition to standardize the lightweight authenticated encryption with associated data (AEAD) and hash function. Gimli is one of the Round 2 candidates that is designed to be efficiently implemented across various platforms, including hardware (VLSI and FPGA), microprocessors, and microcontrollers. However, the performance of Gimli in massively parallel architectures like Graphics Processing Units (GPU) is still unknown. A high performance Gimli implementation on GPU can be especially useful to Internet of Things (IoT) applications, wherein the gateway devices and cloud servers need to handle a massive number of communications protected by AEAD. In this paper, we show that with careful optimization, Gimli can be efficiently implemented in desktop and embedded GPU to achieve extremely high throughput. Our experiments show that the proposed Gimli implementation can achieve 661.44 KB/s (encryption), 892.24 KB/s (decryption), and 4344.46 KB/s (hashing) in state-of-the-art GPUs.

     

Synthetic, biofunctional nucleic acid-based molecular devices

Structural DNA nanotechnology seeks to create architectures of highly precise dimensions using the physical property that short lengths of DNA behave as rigid rods and the chemical property of Watson-Crick base-pairing that acts as a specific molecular glue with which such rigid rods may be joined. Thus DNA has been used as a molecular scale construction material to make molecular devices that can be broadly classified under two categories (i) rigid scaffolds and (ii) switchable architectures. This review details the growing impact of such synthetic nucleic acid based molecular devices in biology and biotechnology. Notably, a significant trend is emerging that integrates morphology-rich nucleic acid motifs and alternative molecular glues into DNA and RNA architectures to achieve biological functionality.     

On the parallelisation of bioinformatics applications

This paper surveys the computational strategies followed to parallelise the most used software in the bioinformatics arena. The studied algorithms are computationally expensive and their computational patterns range from regular, such as database-searching applications, to very irregularly structured patterns (phylogenetic trees). Fine- and coarse-grained parallel strategies are discussed for these very diverse sets of applications. This overview outlines computational issues related to parallelism, physical machine models, parallel programming approaches and scheduling strategies for a broad range of computer architectures. In particular, it deals with shared, distributed and shared/distributed memory architectures.     

FPNA: interaction between FPGA and neural computation

Neural networks are usually considered as naturally parallel computing models. But the number of operators and the complex connection graph of standard neural models can not be directly handled by digital hardware devices. More particularly, several works show that programmable digital hardware is a real opportunity for flexible hardware implementations of neural networks. And yet many area and topology problems arise when standard neural models are implemented onto programmable circuits such as FPGAs, so that the fast FPGA technology improvements can not be fully exploited. Therefore neural network hardware implementations need to reconcile simple hardware topologies with complex neural architectures. The theoretical and practical framework developed, allows this combination thanks to some principles of configurable hardware that are applied to neural computation: Field Programmable Neural Arrays (FPNA) lead to powerful neural architectures that are easy to map onto FPGAs, thanks to a simplified topology and an original data exchange scheme. This paper shows how FPGAs have led to the definition of the FPNA computation paradigm. Then it shows how FPNAs contribute to current and future FPGA-based neural implementations by solving the general problems that are raised by the implementation of complex neural networks onto FPGAs.     

Organic Field‐Effect Transistor for Energy‐Related Applications: Low‐Power‐Consumption Devices,Near‐Infrared Phototransistors,and Organic Thermoelectric Devices

The organic field‐effect transistor (OFET) is the basic building block of integrated circuits. The charge carrier mobility and operating frequency of OFETs have continued to increase; therefore, the power dissipation of OFETs can no longer be ignored. Many research efforts have been made to develop low‐power‐consumption OFETs and complementary circuits. Despite the switching function, OFETs can also be utilized in emerging energy‐related applications, such as near‐infrared (NIR) photodetectors and organic thermoelectric devices. Organic phototransistors show considerably higher photo responsivity than other photodetector architectures due to field‐effect charge modulation. The photoinduced gate modulating largely suppresses the dark current while simultaneously providing gain. These characteristics may favor NIR light detection and suggest that the organic phototransistor is a promising candidate for optoelectronic applications in the NIR regime. For organic thermoelectric applications, OFETs can work as a powerful tool for examining the charge and energy transport in the organic semiconductor, thus giving insight into organic thermoelectric studies. In this review, the authors highlight recent advances in OFET‐ related energy topics, including low‐power‐consumption OFETs, NIR photodetectors, and organic thermoelectric devices. The remaining challenges in the field will also be discussed.     

cuConv: CUDA implementation of convolution for CNN inference

Cluster Computing - Convolutions are the core operation of deep learning applications based on Convolutional Neural Networks (CNNs). Current GPU architectures are highly efficient for training and...     

Programmable Nanocarbon‐Based Architectures for Flexible Supercapacitors

Supercapacitors (SCs), also called electrochemical capacitors, often show high power density, excellent charge/discharge rates, and long cycle life. The recent development of flexible and wearable electronic devices requires that their power sources be sufficiently compact and flexible to match these electronic components. Therefore, flexible SCs have attracted much attention to power current advanced electronics that can be flexible and wearable. In the past several years, many different strategies have been developed to programmably construct different nanocarbon materials into bendable electrode architectures. Furthermore, flexible SC devices with simplified configurations have also been designed based on these nanocarbon‐based architectures. Here, recent developments in the programmable assembly of bendable architectures based on nanocarbon materials are presented. Additionally, the design of flexible nanocarbon‐based SC devices with various configurations is highlighted. The progress made recently paves the way for further development of nanocarbon architectures and corresponding flexible SC devices. Future development and prospects in this area are also analyzed.     

RNA and RNP as new molecular parts in synthetic biology

Synthetic biology has a promising outlook in biotechnology and for understanding the self-organizing principle of biological molecules in life. However, synthetic biologists have been looking for new molecular "parts" that function as modular units required in designing and constructing new "devices" and "systems" for regulating cell function because the number of such parts is strictly limited at present. In this review, we focus on RNA/ribonucleoprotein (RNP) architectures that hold promise as new "parts" for synthetic biology. They are constructed with molecular design and an experimental evolution technique. So far, designed self-folding RNAs, RNA (RNP) enzymes, and nanoscale RNA architectures have been successfully constructed by utilizing Watson-Crick base-pairs together with specific RNA-RNA or RNA-protein binding motifs of known defined 3D structures. Riboregulators for regulating targeted gene expression have also been designed and produced in vitro as well as in vivo. Lately, RNA and ribonucleoprotein complexes have been strongly attracting the attention of molecular biologists because a variety of noncoding RNAs discovered in nature perform spatiotemporal gene expressions. Thus we hope that newly accumulating knowledge on naturally occurring RNAs and RNP complexes will provide a variety of new parts, devices and systems for synthetic biology.     

Identification keys on mobile devices: The Dryades experience

Abstract

This article describes the experience of project Dryades, coordinated by the University of Trieste, in developing interactive identification keys in the form of applications for mobile devices (iPhone/iPad/iPodtouch). All of our keys were previously generated from a database of morpho-anatomical characters using software FRIDA. The applications for mobile devices, which were tested Europe-wide during the project KeyToNature, have proved to be useful in education, in the promotion of nature-aware tourism and in projects of citizen science. The first-generation apps were stand-alone packages which consisted into a sequence of stand-alone HTML pages, while those of the second generation do incorporate the digital key as a true database into the mobile device. All of the hitherto published applications are basically in the form of illustrated dichotomous keys. Future developments will also include a multi-entry query interface, and will be extended to devices which use the Android system.     

Large Neural Network Simulations on Multiple Hardware Platforms

To efficiently simulate very large networks of interconnected neurons, particular consideration has to be given to the computer architecture being used. This article presents techniques for implementing simulators for large neural networks on a number of different computer architectures. The neuronal simulation task and the computer architectures of interest are first characterized, and the potential bottlenecks are highlighted. Then we describe the experience gained from adapting an existing simulator, SWIM, to two very different architectures–vector computers and multiprocessor workstations. This work lead to the implementation of a new simulation library, SPLIT, designed to allow efficient simulation of large networks on several architectures. Different computer architectures put different demands on the organization of both data structures and computations. Strict separation of such architecture considerations from the neuronal models and other simulation aspects makes it possible to construct both portable and extendible code.     

Slowed diffusion in tumors revealed by microfiberoptic epifluorescence photobleaching

It has not been possible to measure diffusion deep in solid tissues such as tumors because of the limited light penetration of conventional optical techniques. Here we report a microfiberoptic epifluorescence photobleaching (MFEP) method in which photobleaching is done by laser epi-illumination through a multimode fiberoptic whose micron-sized tip can be introduced deep into tissues. We applied MFEP to measure the diffusion of fluorescent macromolecules in tumors in living mice, at depths well beyond those accessible by surface optical measurements. Macromolecule diffusion was slowed about twofold within 200 microm of the surface of a solid tumor, but was slowed greater than tenfold beyond 500 microm. Our results reveal a remarkable and previously unrecognized slowing of diffusion deep in tumors, which correlated with the differing tissue architectures of tumor periphery versus core, and with altered tumor vasculature produced by aquaporin-1 deletion. MFEP should have wide applications for measuring diffusion in organs, solid tumors and other light-inaccessible tissue masses.     

3D Printing Sulfur Copolymer‐Graphene Architectures for Li‐S Batteries

3D printing is becoming an efficient approach to facilely and accurately fabricate diverse complex architectures with broad applications. However, suitable inks and 3D print favorable architectures with high electrochemical performances for energy storage are still being explored. Here, sulfur copolymer‐graphene architectures with well‐designed periodic microlattices are 3D printed as a cathode for Li‐S batteries using a suitable ink composed of sulfur particles, 1,3‐diisopropenylbenzene (DIB), and condensed graphene oxide dispersion. Using thermal treatment, elemental sulfur can be reacted with DIB to produce sulfur copolymer, which can partially suppress the dissolution of polysulfides. Moreover, graphene in the architecture can provide high electrical conductivity for whole electrode. Hence, 3D printed sulfur copolymer‐graphene architecture exhibits a high reversible capacity of 812.8 mA h g^?1 and good cycle performance. Such a simple 3D printing approach can be further extended to construct many complex architectures for various energy storage devices.     

Middleware-based multi-agent development environment for building and testing distributed intelligent systems

The spread of the Internet of Things (IoT) is demanding new, powerful architectures for handling the huge amounts of data produced by the IoT devices. In many scenarios, many existing isolated solutions applied to IoT devices use a set of rules to detect, report and mitigate malware activities or threats. This paper describes a development environment that allows the programming and debugging of such rule-based multi-agent solutions. The solution consists of the integration of a rule engine into the agent, the use of a specialized, wrapping agent class with a graphical user interface for programming and testing purposes, and a mechanism for the incremental composition of behaviors. Finally, a set of examples and a comparative study were accomplished to test the suitability and validity of the approach. The JADE multi-agent middleware has been used for the practical implementation of the approach.

     

NSL a Neuro-Symbolic Language for a Neuro-Symbolic Processor (NSP)

A Neuro-Symbolic Language for monotonic and non-monotonic parallel logical inference by means of artificial neural networks (ANNs) is presented. Both the language and its compiler have been designed and implemented in order to translate the neural representation of a given problem into a VHDL software, which in turn can set devices such as FPGA. The result of this operation leads to an electronic circuit that we call NSP (Neuro-Symbolic Processor).     

Towards computing with proteins

Can proteins be used as computational devices to address difficult computational problems? In recent years there has been much interest in biological computing, that is, building a general purpose computer from biological molecules. Most of the current efforts are based on DNA because of its ability to self‐hybridize. The exquisite selectivity and specificity of complex protein‐based networks motivated us to suggest that similar principles can be used to devise biological systems that will be able to directly implement any logical circuit as a parallel asynchronous computation. Such devices, powered by ATP molecules, would be able to perform, for medical applications, digital computation with natural interface to biological input conditions. We discuss how to design protein molecules that would serve as the basic computational element by functioning as a NAND logical gate, utilizing DNA tags for recognition, and phosphorylation and exonuclease reactions for information processing. A solution of these elements could carry out effective computation. Finally, the model and its robustness to errors were tested in a computer simulation. Proteins 2006. © 2006 Wiley‐Liss, Inc.     

Light Harvesting Dendrimers

Tree-like dendrimers with decreasing number of chromophores from periphery to core is an attractive candidate for light-harvesting applications. Numerous dendritic designs with different kinds of light-collecting chromophores at periphery and an energy-sink at the core have been demonstrated with high energy transfer efficiency. These building blocks are now being developed for several applications such as light-emitting diodes, frequency converters and other photonic devices. This review outlines the efforts that are based on both conjugated and non-conjugated dendrimers.     

A phase-based stereo vision system-on-a-chip

A simple and fast technique for depth estimation based on phase measurement has been adopted for the implementation of a real-time stereo system with sub-pixel resolution on an FPGA device. The technique avoids the attendant problem of phase warping. The designed system takes full advantage of the inherent processing parallelism and segmentation capabilities of FPGA devices to achieve a computation speed of 65megapixels/s, which can be arranged with a customized frame-grabber module to process 211frames/s at a size of 640x480 pixels. The processing speed achieved is higher than conventional camera frame rates, thus allowing the system to extract multiple estimations and be used as a platform to evaluate integration schemes of a population of neurons without increasing hardware resource demands.     

Nanopore sensors: From hybrid to abiotic systems

This paper provides an overview of different nanostructured architectures utilised in electrochemical devices and their application in biosensing and bioelectronics. Emphasis is placed on the fabrication of nanostructured films based on a layer-by-layer (LBL) films approach. We discuss the theory and the mechanism of charge transfer in polyelectrolyte multilayer films (PEM), as well as between biomolecules and redox centres, for the development of more sensitive and selective biosensors. Further, this paper presents an overview of topics involving the interaction between nanostructured materials, including metallic nanoparticles and carbon materials, and their effects on the preservation of the activity of biological molecules immobilised on electrode surfaces. This paper also presents examples of biological molecules utilised in film fabrication, such as DNA, several kinds of proteins, and oligonucleotides, and of the role of molecular interaction in biosensing performance. Towards the utilisation of LBL films, examples of several architectures and different electrochemical approaches demonstrate the potential of nanostructured LBL films for several applications that include the diagnosis and monitoring of diseases. Our main aim in this review is to survey what can assist researchers by presenting various approaches currently used in the field of bioelectrochemistry utilising supramolecular architectures based on an LBL approach for application in electrochemical biosensing.     

Open, reconfigurable cytometric acquisition system: ORCAS.

A digital signal processing (DSP)-based digital data acquisition system has been developed to support novel flow cytometry efforts. The system flexibility includes how it detects, captures, and processes event data. Custom data capture boards utilizing analog to digital converters (ADCs) and field programmable gate arrays (FPGA) detect events and capture correlated event data. A commercial DSP board processes the captured data and sends the results over the IEEE 1394 bus to the host computer that provides a user interface for acquisition, display, analysis, and storage. The system collects list mode data, correlated pulse shapes, or streaming data from a variety of detector types using Linux, Mac OS X, and Windows host computers. It extracts pulse features not found on commercial systems with excellent sensitivity and linearity over a wide dynamic range. List mode data are saved in FCS 3.0 formatted files while streaming or correlated waveform data are saved in custom format files for postprocessing. Open, reconfigurable cytometric acquisition system is compact, scaleable, flexible, and modular. Programmable feature extraction algorithms have exciting possibilities for both new and existing applications. The recent availability of a commercial data capture board will enable general availability of similar systems.     

Direct GPU/FPGA communication Via PCI express

We describe a mechanism for connecting GPU and FPGA devices directly via the PCI Express bus, enabling the transfer of data between these heterogeneous computing units without the intermediate use of system memory. We evaluate the performance benefits of this approach over a range of transfer sizes, and demonstrate its utility in a computer vision application. We find that bypassing system memory yields improvements as high as 2.2× in data transfer speed, and 1.9× in application performance.