Evaluation of HPC Acceleration and Interconnect Technologies for High-Throughput Data Acquisition

Efficient data movement in multi-node systems is a crucial issue at the crossroads of scientific computing, big data, and high-performance computing, impacting demanding data acquisition applications from high-energy physics to astronomy, where dedicated accelerators such as FPGA devices play a key role coupled with high-performance interconnect technologies. Building on the outcome of the RECIPE Horizon 2020 research project, this work evaluates the use of high-bandwidth interconnect standards, namely InfiniBand EDR and HDR, along with remote direct memory access functions for direct exposure of FPGA accelerator memory across a multi-node system. The prototype we present aims at avoiding dedicated network interfaces built in the FPGA accelerator itself, leaving most of the resources for user acceleration and supporting state-of-the-art interconnect technologies. We present the detail of the proposed system and a quantitative evaluation in terms of end-to-end bandwidth as concretely measured with a real-world FPGA-based multi-node HPC workload.

Virtual reality as a tool for training specialists in the field of radiation non-destructive testing

Organizational, technical and methodological approaches to the creation and application of virtual reality in additional education are considered. Particularly for use and development of a digital radiography simulator in non-destructive testing of products and materials. It is noted that virtual reality technologies are most widely used in training and knowledge testing of engineering and technical personnel and workers in production, as well as in technological preparation during complex and exacting operations, including products and materials testing. The pilot solutions obtained and tested to date allow us to evaluate the results of complex scientific research. The prospects of expanding the applicability range of software and hardware virtual reality solutions, including those based on network interfaces, protocols and telecommunications solutions, are determined.

Investigating the Throughput Performance of the MPT-GRE Network Layer Multipath Library in Emulated WAN Environment

Nowadays there is a growing demand for a much faster and more secure communication without borders through the internet, which is provoking more and more both network designers and manufacturers of communication devices. Thanks to the BYOD trend, our communication devices can be easily carried anywhere in the world. They generally have several built-in network interfaces (e.g. Ethernet, Wi-Fi, 4G). Theoretically, using these cards in parallel, we could speed up data transmission, and thus communication, by aggregating the channel capabilities of the interfaces. On the other hand, we could make data transmission more secure by applying redundancy to the system. Unfortunately, traditional IP-based communications do not allow the use of parallel interfaces in a given communication session, leaving the hardware capabilities of our communications devices virtually untapped. To address this issue, we have developed a multipath communication solution called MPT-GRE, which we have already tested in several laboratory environments. The measurement results were published in our previous articles. In this paper we are going to test it in a much more realistic environment, using the Dummynet WAN emulation software. The measurement results confirmed that the MPT-GRE multipath solution is able to aggregate the performance of physical connections efficiently in the emulated Fast Ethernet IPv4 WAN environment as well.

Multimodal Network Architecture for Shared Situational Awareness Amongst Vessels

To shift the paradigm towards Industry 4.0, maritime domain aims to utilize shared situational awareness (SSA) amongst vessels. SSA entails sharing various heterogeneous information, depending on the context and use case at hand, and no single wireless technology is equally suitable for all uses. Moreover, different vessels are equipped with different hardware and have different communication capabilities, as well as communication needs. To enable SSA regardless of the vessel’s communication capabilities and context, we propose a multimodal network architecture that utilizes all of the network interfaces on a vessel, including multiple IEEE 802.11 interfaces, and automatically bootstraps the communication transparently to the applications, making the entire communication system environment-aware, service-driven, and technology-agnostic. This paper presents the design, implementation, and evaluation of the proposed network architecture which introduces virtually no additional delays as compared to the Linux communication stack, automates communication bootstrapping, and uses a novel application-network integration concept that enables application-aware networks, as well as network-aware applications. The evaluation was conducted for several IEEE 802.11 flavors. Although inspired by SSA for vessels, the proposed architecture incorporates several concepts applicable in other domains. It is modular enough to support existing, as well as emerging communication technologies.

S-CDCA: a semi-cluster directive-congestion protocol for priority-based data in WSNs

The internet of things (IoT) protocols and regulations are being developed forvarious applications includes: habitat monitoring, machinery control, general health-care, smart-homes and more. A great part of I0T comprised of sensors nodes in connected networks (i.e. sensor networks.). A sensor network is a group of nodes with sensory module and computational elements connected through network interfaces. The most interesting type of sensor networks are wireless sensor networks. The nodes here are connected through wirless interfaces. The shared medium between these nodes, creates different challenges. Congestion in such network is ineavitable. Different models andmethods were proposed to alleviate congestion in wireless sensor networks.This paper presents a semi-cluster directive congestion method that allivatenetwork congestion forpriority-baseddata transmission. The method simprove the network performance by implementing temporary cluster forlow level priority data packets while providing a clear link between highpriority data source node and the network base station. Simulation resultsshow that. The proposed method outperformes ad hocOn-demand distance vector (AODV) reactive procotol approach and priority-based congestion control dynamic clustering (PCCDC) a cluster-based methodin network energy consumption and control packets overhead during network operation.The proposed method also shows comparative improvments in end-to-enddelays versus PCCDC.

A Model-Driven and Business Approach to Autonomic Network Management

As corporate networks continue to expand, the technologies that underpin these enterprises must be capable of meeting the operational goals of the operators that own and manage them. Automation has enabled the impressive scaling of networks from the days of Strowger. The challenge now is not only to keep pace with the continuing huge expansion of capacity but at the same time to manage a huge increase in complexity – driven by the range of customer solutions and technologies. Recent advances in automation, programmable network interfaces, and model-driven networking will provide the possibility of closed-loop, self-optimizing, and self-healing networks. Collectively these support the goals of a truly automated network, commonly understood as “autonomic networking” even though this is a prospect yet to be achieved. This paper outlines the progress made towards autonomic networking and the framework and procedures developed during the UK Next Generation Converged Digital Infrastructure (NG-CDI) project. It outlines the operator-driven requirements and capabilities that have been identified, and proposes an autonomic management framework, and summarizes current art and the challenges that remain.

Wireless communication technologies for the Internet of Things

Internet of Things (IoT) is the inter-networking paradigm based on many processes such as identifying, sensing, networking and computation. An IoT technology stack provides seamless connectivity between various physical and virtual objects. The increasing number of IoT applications leads to the issue of transmitting, storing, and processing a large amount of data. Therefore, it is necessary to enable a system capable to handle the growing traffic requirements with the required level of QoS (Quality of Service). IoT devices become more complex due to the various components such as sensors and network interfaces. The IoT environment is often demanding for mobile power source, QoS, mobility, reliability, security, and other requirements. Therefore, new IoT technologies are required to overcome some of these issues. In recent years new wireless communication technologies are being developed to support the development of new IoT applications. This paper provides an overview of some of the most widely used wireless communication technologies used for IoT applications.

Programmability of Connectivity Control

Network programmability and edge computing as key features of next generation communications enable innovative services. While the programmability is focused on the core network of the fifth-generation system, the edge computing moves the network intelligence to the radio access network. This paper presents a study on the programmability of connectivity control as a function of radio access network using Multi-access Edge Computing. The capability of using more than one radio access technology simultaneously enhances reliability and increases the throughput, especially in dense networks. Opening the radio access network interfaces for programmability of multi-connectivity enables analytics applications to control the device connections to multiple radio links simultaneously based on information of radio conditions, user location or specific policies. The research novelty is in opening the radio access network interfaces for edge applications to access connectivity control.

Emergence of a smooth interface from growth of a dendritic network against a mechanosensitive contractile material

Structures and machines require smoothening of raw materials. Self-organized smoothening guides cell and tissue morphogenesis, and is relevant to advanced manufacturing. Across the syncytial Drosophila embryo surface, smooth interfaces form between expanding Arp2/3-based actin caps and surrounding actomyosin networks, demarcating the circumferences of nascent dome-like compartments used for pseudo-cleavage. We found that smoothening of the actomyosin interfaces requires Arp2/3 in vivo. To dissect the physical basis of this requirement, we reconstituted the interacting networks using node-based models. When actomyosin networks were simulated with clearances instead of Arp2/3 networks, rough boundaries persisted with low levels of myosin contractility. With addition of expanding Arp2/3 networks, network-network interfaces failed to smoothen, but accumulated myosin nodes and tension. After incorporating actomyosin mechanosensitivity, Arp2/3 network growth induced local contractility and smoothening of the interfaces, effects also evident in vivo. In this way, a smooth structure can emerge from the lateral interaction of irregular starting materials.