Computing Graph Neural Networks: A Survey from Algorithms to Accelerators

Graph Neural Networks (GNNs) have exploded onto the machine learning scene in recent years owing to their capability to model and learn from graph-structured data. Such an ability has strong implications in a wide variety of fields whose data are inherently relational, for which conventional neural networks do not perform well. Indeed, as recent reviews can attest, research in the area of GNNs has grown rapidly and has lead to the development of a variety of GNN algorithm variants as well as to the exploration of ground-breaking applications in chemistry, neurology, electronics, or communication networks, among others. At the current stage research, however, the efficient processing of GNNs is still an open challenge for several reasons. Besides of their novelty, GNNs are hard to compute due to their dependence on the input graph, their combination of dense and very sparse operations, or the need to scale to huge graphs in some applications. In this context, this article aims to make two main contributions. On the one hand, a review of the field of GNNs is presented from the perspective of computing. This includes a brief tutorial on the GNN fundamentals, an overview of the evolution of the field in the last decade, and a summary of operations carried out in the multiple phases of different GNN algorithm variants. On the other hand, an in-depth analysis of current software and hardware acceleration schemes is provided, from which a hardware-software, graph-aware, and communication-centric vision for GNN accelerators is distilled.

Online Vegan Food Delivery System

The system proposed here signifies Vegan food delivery process. This system will allow restaurants to quickly and easily manage an online menu which customers can browse and use to place orders with just a few clicks. The system then relays these orders to restaurant’s employees through an easy to navigate graphical interface for efficient processing. Keywords: Vegan food delivery, customers, vegan vibes, food order etc.

Exploration of Intelligent Manufacturing Methods for Complex Products Driven by Multisource Data

In order to improve the multisource data-driven fusion effect in the intelligent manufacturing process of complex products, based on the proposed adaptive fog computing architecture, this paper takes into account the efficient processing of complex product intelligent manufacturing services within the framework and the rational utilization of fog computing layer resources to establish a fog computing resource scheduling model. Moreover, this paper proposes a fog computing architecture for intelligent manufacturing services for complex products. The architecture adopts a three-layer fog computing framework, which can reasonably provide three types of services in the field of intelligent manufacturing. In addition, this study combines experimental research to verify the intelligent model of this article and counts the experimental results. From the analysis of experimental data, it can be seen that the complex product intelligent manufacturing system based on multisource data driven proposed in this paper meets the data fusion requirements of complex product intelligent manufacturing.

Mechanical Alloying Process Applied for Obtaining a New Biodegradable Mg-xZn-Zr-Ca Alloy

The aim of the present paper is to apply the mechanical alloying process to obtain from powder components a new biodegradable Mg-based alloy powder from the system Mg-xZn-Zr-Ca, with high biomechanical and biochemical performance. Various processing parameters for mechanical alloying have been experimented with the ultimate goal to establish an efficient processing route for the production of small biodegradable parts for the medical domain. It has been observed that for the same milling parameters, the composition of the powders has influenced the powder size and shape. On the other hand, for the same composition, the highest experimented milling speed and time conduct to finer powder particles, almost round-shaped, without pores or various inclusions. The most uniform size has been obtained for the powder sample with 10 wt.%Zn. These powders were finally processed by selective laser melting, an additive manufacturing technology, to obtain a homogeneous experimental sample, without cracking, for future more systematical trials.

Deep Eutectic Solvents For Efficient Processing of Renewable Resources Into Sustainable Biofuels and High Value-Added Carbon-Based Fine Chemicals: Recent Developments, Challenges, and Future Perspectives

Green chemistry techniques for the exploitation of renewable resources have emerged as beneficial techniques for producing sustainable biofuels and high value-added carbon-based fine chemicals with the potential to decrease the impact of anthropogenic activities on the environment. Despite various green chemistry technologies for processing renewable resources into different valuable products, there are still several major issues concerning the pretreatment processes and techniques, such as high cost and high-energy consumption. Thankfully, deep eutectic solvents (DESs), a potentially attractive “green solvent” biodegradable substitute to environmentally harmful organic solvents, have been progressively exploited for renewable resources processing. Therefore, the central focus of this review is to present recent developments and challenges of DESs as processing green solvents for renewable resources. We believe this comprehensive review will provide new insights towards developing state-of-the-art sustainable and new green technologies for the efficient processing of renewable resources for sustainable biofuels and value-added carbon-based fine chemicals.

Investigating Constraints Along the Plant Secretory Pathway to Improve Production of a SARS-CoV-2 Spike Vaccine Candidate

Given the complex maturation requirements of viral glycoproteins and the challenge they often pose for expression in plants, the identification of host constraints precluding their efficient production is a priority for the molecular farming of vaccines. Building on previous work to improve viral glycoprotein production in plants, we investigated the production of a soluble SARS-CoV-2 spike comprising the ectopic portion of the glycoprotein. This was successfully transiently expressed in N. benthamiana by co-expressing the human lectin-binding chaperone calreticulin, which substantially increased the accumulation of the glycoprotein. The spike was mostly unprocessed unless the protease furin was co-expressed which resulted in highly efficient processing of the glycoprotein. Co-expression of several broad-spectrum protease inhibitors did not improve accumulation of the protein any further. The protein was successfully purified by affinity chromatography and gel filtration, although the purified product was heterogenous and the yields were low. Immunogenicity of the antigen was tested in BALB/c mice, and cellular and antibody responses were elicited after low dose inoculation with the adjuvanted protein. This work constitutes an important proof-of-concept for host plant engineering in the context of rapid vaccine development for SARS-CoV-2 and other emerging viruses.

Distributed Streaming Big Data Analytics for Internet of Things (IoT)

In this chapter, a discussion on the integration of distributed streaming Big Data Analytics with the Internet of Things is presented. The chapter begins with the introduction of these two technologies by discussing their features and characteristics. Discussion on how the integration of these two technologies benefit in efficient processing of IoT device generated sensor data follows next. Such data centric processing of IoT data powered by cloud, services and other enablers will be the architecture of most of the realtime systems involving sensors and real-time monitoring and actuation. The Volume, Variety and Velocity of sensor generated data make it a Big Data scenario. In addition, the data is real time and requires decisions or actuations immediately. This chapter discusses how IoT data can be processed using distributed, scalable stream processing systems. The chapter is concluded with future directions of such real time Big Data Analytics in IoT.

A Review on Big Data Stream Processing Applications: Contributions, Benefits, and Limitations

The amount of data in our world has been rapidly keep growing from time to time. In the era of big data, the efficient processing and analysis of big data using machine learning algorithm is highly required, especially when the data comes in form of streams. There is no doubt that big data has become an important source of information and knowledge in making decision process. Nevertheless, dealing with this kind of data comes with great difficulties; thus, several techniques have been used in analyzing the data in the form of streams. Many techniques have been proposed and studied to handle big data and give decisions based on off-line batch analysis. Today, we need to make a constructive decision based on online streaming data analysis. Many researchers in recent years proposed some different kind of frameworks for processing the big data streaming. In this work, we explore and present in detail some of the recent achievements in big data streaming in term of contributions, benefits, and limitations. As well as some of recent platforms suitable to be used for big data streaming analytics. Moreover, we also highlight several issues that will be faced in big data stream processing. In conclusion, it is hoped that this study will assist the researchers in choosing the best and suitable framework for big data streaming projects.

Composites from Recycled and Modified Woods—Technology, Properties, Application

The intention of efficient processing and use of less valuable wood species, bio-damaged logs, sawmill residues, cuttings, chips, sawdust, recycled wooden products, and other lignocellulosic raw materials in the production of wood composites is the focus of several scientific research institutes in the world [...]