Runtime Infrastructure Research Articles

Adapting HLA-based co-simulation for interdependent infrastructure resilience management

Since infrastructure systems are threatened by natural or man-made hazards with unprecedented frequency and intensity, decision-makers spare no efforts in proposing methods and tools to improve the resilience of interdependent infrastructures. However, these approaches and toolkits are flawed in characterizing the heterogeneous operating regimes of different infrastructure systems. Moreover, worst-case scenario is commonly harnessed when generating the hazard information maps and assessing the induced adverse consequences on infrastructures, which is exaggerated as all the identified vulnerable spots in infrastructures rarely suffer from the worst situation simultaneously. Besides, resilience assessment methods are mostly static and not competent in capturing the emergence and development of specified hazard event and revealing the dynamic resilience behavior of interdependent infrastructure systems. To address these gaps, this paper proposes an overarching High Level Architecture (HLA)-based co-simulation framework which is composed of the HLA-based service adaptation supported by the Run Time Infrastructure (RTI) libraries, simulators and interface design between the federates, and typical scenarios to facilitate interdependent infrastructure resilience management. It further adapts and orchestrates the generic HLA-based co-simulation framework to suit the interdependent infrastructure resilience management scenarios, while accommodating the physics-based domain-specific simulators; supports seamless and interoperable message exchange between interdependent infrastructures along with the development of specific hazards; and provides interface design between the infrastructure-specific simulators and the RTI. A typical flooding case is adopted to demonstrate the validity and transferability of our proposed framework. The overarching HLA-based co-simulation framework aims to formalize a modeling paradigm to investigate the time-varying resilience behavior of interdependent infrastructure systems, in order to contribute to the collaborative and joint decision making efforts between stakeholders of infrastructure regulatory agencies.

Curvetime: A blockchain framework for Artificial Intelligence computation

<h2>Abstract</h2> Curvetime is a Blockchain framework that organically orchestrates proof-of-work and AI model training on one platform, optimizing resource usages for intense computation. In this framework proof-of-work is represented as a reinforcement-learning problem, in which a learning agent makes an optimal decision over the environment's states, whereas a new block is added and verified. It has been a backbone for blockchain based industrial applications, and a platform incubating diversified industrial intelligence models. As a two-in-one runtime infrastructure of blockchain and AI, curvetime improves the effectiveness of computing resources and accelerates AI model training at the extents proportional to the number of computing nodes.

Elastic-DF: Scaling Performance of DNN Inference in FPGA Clouds through Automatic Partitioning

Customized compute acceleration in the datacenter is key to the wider roll-out of applications based on deep neural network (DNN) inference. In this article, we investigate how to maximize the performance and scalability of field-programmable gate array (FPGA)-based pipeline dataflow DNN inference accelerators (DFAs) automatically on computing infrastructures consisting of multi-die, network-connected FPGAs. We present Elastic-DF, a novel resource partitioning tool and associated FPGA runtime infrastructure that integrates with the DNN compiler FINN. Elastic-DF allocates FPGA resources to DNN layers and layers to individual FPGA dies to maximize the total performance of the multi-FPGA system. In the resulting Elastic-DF mapping, the accelerator may be instantiated multiple times, and each instance may be segmented across multiple FPGAs transparently, whereby the segments communicate peer-to-peer through 100 Gbps Ethernet FPGA infrastructure, without host involvement. When applied to ResNet-50, Elastic-DF provides a 44% latency decrease on Alveo U280. For MobileNetV1 on Alveo U200 and U280, Elastic-DF enables a 78% throughput increase, eliminating the performance difference between these cards and the larger Alveo U250. Elastic-DF also increases operating frequency in all our experiments, on average by over 20%. Elastic-DF therefore increases performance portability between different sizes of FPGA and increases the critical throughput per cost metric of datacenter inference.

Simple Run-Time Infrastructure (SRTI): An accessible distributed computing platform for interdisciplinary simulation

Distributed computing is a necessity for conducting cross-disciplinary research where field-specific computational models (simulators) are available, but have not been designed to work together. An example of this is natural hazards research. Simulators abound in the disparate fields that fall under this area, e.g. social science, engineering, economics, and health, but little progress has been made to integrate the simulators to study overarching and cross-disciplinary disaster scenarios. The reason for slow penetration of this technology is the high barrier to entry, which requires extensive knowledge of computer science and programming. Building upon an existing platform named Simple Run Time Infrastructure (SRTI v1), a new, fundamentally different version (SRTI v2) is developed to address the issues mentioned above. Designed to provide a low barrier to use, SRTI v2 is developed for users with limited programming experience and designed to simplify and streamline a user’s effort to compose a distributed simulation and handle time management. To achieve this primary objective, pre-compiled components are provided including the RTI Management Server, RTI Wrapper, and a GUI. By exploiting these pre-compiled components, users can compose a scalable distributed simulation with heterogeneous computational models. To demonstrate the concepts behind SRTI, a cross-language simulation, modified and extended from a time-dependent resilience analysis of an electric power system in the literature, is presented to show the scalability and usability of SRTI. Features of the different versions of SRTI are discussed and useful features to develop in the future are outlined.

MockFog 2.0: Automated Execution of Fog Application Experiments in the Cloud

Fog computing is an emerging computing paradigm that uses processing and storage capabilities located at the edge, in the cloud, and possibly in between. Testing and benchmarking fog applications, however, is hard since runtime infrastructure will typically be in use or may not exist, yet. While approaches for the emulation of infrastructure testbeds do exist, their focus is typically the emulation of edge devices. Other approaches also emulate infrastructure within the core network or the cloud, but they miss support for automated experiment orchestration. In this paper, we propose to evaluate fog applications on an emulated infrastructure testbed created in the cloud which can be manipulated based on a pre-defined orchestration schedule. Developers can freely design the infrastructure, configure performance characteristics, manage application components, and orchestrate their experiments. We also present our proof-of-concept implementation MockFog 2.0. We use MockFog 2.0 to evaluate a fog-based smart factory application and showcase how its features can be used to study the impact of infrastructure changes and workload variations. With these experiments, we also show that MockFog can achieve good experiment reproducibility, even in a public cloud environment.

HLA Run Time Infrastructure: A Comparative Study

Distributed computer simulation systems use a general-purpose architecture known as HLA (High Level Architecture). HLA aims to provide common architecture for all types of distributed modeling and simulations by providing interoperability and reusability among simulations. A middleware known as RTI (Run Time Infrastructure) provides distributed simulation services under HLA. RTI offers a communication framework which is necessary for federates to interact during simulation process. Many RTI implementations are available commercially and are open source to achieve reusability and interoperability by implementing HLA. However, functionality and performance of RTI is based on its structure. This review presents the performance analysis of multiple RTI implementations, so that the best suitable RTI can be selected for a given simulation scenario and provides the structural comparison of central and service distributed RTI. It also presents the performance analysis of multiple RTI implementations and provides the structural comparison of CeRTI (Central Run Time Infrastructure) and SDRTI (Service Distributed Run Time Infrastructure). Many simulations use HLA, and RTI as basic part of these simulations. This comparative study describes characteristics of different RTIs, their comparisons and implementations that will help the reader to select suitable RTI according to environment and requirements.

A novel multi-level evaluation approach for human-coupled IoT applications

The recent emerging trend to promote the harmonious interactions between human and smart objects in Internet of Things (IoT) applications has led to the rising demand for the comprehensive exploration of associated IoT design spaces, considering system-, network-, and human-level perspectives. The traditional design approach of networked smart objects tends to ignore the effects caused by human interactions so that the existing approach has the limited capability in joint system/network/human multi-level exploration. In this paper, we propose a high-level system modeling and multi-level simulation approach for microsystem models to interoperate with high-level models in the human-level perspective and to provide comprehensive co-exploration. The high-level system modeling enables one to abstract the detailed operations of hardware platforms using power and timing data obtained by the proposed high-level state-based profiling method and using the event-driven modeling concept. For the event-driven system modeling, we utilize the discrete-event system specification (DEVS) to support scalable model-driven prototyping. To represent the coupled relationship of network-level system activity and human-level interactions between node systems and humans, we employ a general-purpose network simulator to model the operation and communication of network modules in smart objects and developed agent-based human behavioral models. For the co-simulation of multi-level models, we designed a distributed simulation platform to enable the interoperation between the DEVS simulator and the network simulator using a runtime infrastructure. The proposed modeling and simulation approach is applied for the multi-level evaluation of a smart museum application to estimate the effect of energy-efficiency policies.

Infrastructure System Simulation Interoperability Using the High-Level Architecture

Infrastructure planners need new methods to exchange technical data in the pursuit of long-term and cross-sector objectives including sustainability. This paper demonstrates a method for infrastructure system simulation interoperability by applying the high-level architecture (HLA) to the infrastructure system of systems (ISoS) modeling framework. The ISoS considers time-aggregated simulations with iterative data exchange periods to resolve cyclic dependencies. Applying the HLA defines object models, operational agreements, federate implementations, and physical infrastructure required for a federated simulation. An example demonstrates the method on a scenario with 18 infrastructure elements across water, electricity, petroleum, and social sectors as individual federates. The cost of interoperability arises from network latency, runtime infrastructure (RTI) overhead, and additional development effort. Results across several RTIs show that a distributed simulation requires 10-100 times longer executions than noninteroperable configurations. Analysis of source code shows that the HLA contributes about 1500 (56%) more software lines of code to implement federation object models and agreements. The value of interoperability arises from new capabilities provided by the HLA: physical distribution and independent control of constituent models. Conclusions describe a role for the HLA in model-based concept exploration during collaborative and interorganizational design activities with live human participation similar to its use for wargames.

Co-Simulation of Distributed Smart Grid Software Using Direct-Execution Simulation

The use of distributed computation and control is pervasive for a wide range of smart grid research topics. However, recent developments of smart grid co-simulation platforms have not been able to provide effective support for the modeling and simulation of distributed software systems. In particular, co-simulation literatures only focused on the integration of electrical and communication network simulators, and the responsibility for modeling distributed software is often delegated to one of these two simulators. Since these domain specific simulators are not designed for this purpose, such delegation incurs many limitations that prevent convenient, effective, and accurate modeling of software behaviors. To mitigate the problem, this paper presents, to our knowledge, the first co-simulation integration of direct-execution simulators to provide dedicated support for distributed smart grid software. We first present the development of the novel DecompositionJ framework (DEterministic, COncurrent Multi-PrOcessing SImulaTION for Java programs), which is a compiler-based code analyzer and transformer to automatically convert multi-thread Java programs into direct-execution simulators, eliminating the need for manual code or model development. Next, we apply DecompositionJ framework to generate simulators for a popular multi-agent platform JADE. The JADE simulators are then integrated with electromagnetic transient simulator (PSCAD) and packet-level network simulator (OPNET) using standardized co-simulation runtime infrastructure. At last, we conduct a case study on agent-based smart grid restoration using this novel co-simulation platform. Through the analysis of simulation results, it is shown that the proposed direct-execution simulation framework is able to facilitate the understanding, evaluation, and debugging of distributed smart grid software.

국방 훈련체계용 가상물리시스템 시간 동기화 기법

LVC(Live Virtual Constructive) 통합 훈련 체계는 대표적인 가상물리시스템 중 하나이다. LVC 훈련 체계를 구성하는 각 L,V,C 체계들은 서로 다른 시간 도메인, 해상도 및 동작 방식을 갖는다. 이에 각 체계의 이종 미들웨어들을 공통의 통신 미들웨어로서 통합하고, 연동 gateway (GW)를 이용하여 이종 체계를 연동하는 방안 연구는 매우 중요하다. 특히, LVC체계는 서로 다른 시간 도메인을 사용하고 있기 때문에 다른 체계 간 이벤트들의 인과성 보장과 시간 동기화에 대한 연구가 필요하다. 본 연구에서는 LVC 훈련체계의 통합 미들웨어로서 L 체계의 실시간 통신 미들웨어인 OMG DDS(Data Distribution Service)를 기반으로 HLA(High Level Architecture)/RTI(Run Time Infrastructure)기반의 시뮬레이션 시간을 사용하는 V, C 체계와 시간 차원에서 연동하는 방안을 제시한다. 본 논문에서는 HLA의 Time management 기능과 Clock Federate를 기반으로 DDS와 HLA 각각의 통신 개체인 파티서펀트(Participant)와 페더레이트(Federate)간 인과성을 유지하는 정밀한 시간 동기화 기법을 제안한다. 더불어, 각 미들웨어 상에서 실제 응용을 개발하여 제안한 방식에 의해 이종 체계 간 시간 제어 및 동기화가 이루어짐을 검증한다. LVC(live-virtual-constructive) integrated training system is a representative cyber-physical system. Each systems in a LVC system has different time domain, resolution and operation methods. So, it is very important to integrate different middlewares as a common middleware for heterogeneous systems using inter-working GWs. Especially, since the LVC system uses different time, it is necessary to study the method for guaranteeing causality and time synchronization among the events from different systems. In this study, we propose an time synchronization scheme to integrate the virtual and constructive system which use the simulation time of HLA (High Level Architecture)/ RTI (Run Time Infrastructure) into the live system based on the OMG DDS (Data Distribution Service). We propose a precise time synchronization scheme based on HLA time management and clock federate between participants and federates which are the communication objects of DDS and HLA/RTI respectively. In addition, we verified that time is well-synchronized among heterogeneous systems using the suggested scheme by implementing and demonstrating simulation applications on each middleware.

Technical Note on Secure Service Access in VANET using PBKDF Algorithm

Vehicular networks and their applications have gained great attention to the research community and vehicle industry in past few years. The applications in Vehicular Ad-hoc Networks (VANETs) are growing rapidly. The two main application classes have lately gained popularity such as secured and non secured applications. In this paper, we focus on security in VANETs using Packet Bit Key Data Function (PBKDF) Algorithm. The performance of proposed algorithm is assessed using Simulation of Urban Mobility (SUMO) coupled with V2X simulation runtime infrastructure.

A Scalable Modeling and Simulation Environment for Chemical Gas Emergencies

To reflect an evacuation process using a conventional agent-based approach to model human movement under chemical gas exposure, a scalable and hybrid agent-based simulation (ABS) model incorporates an interactive computational fluid dynamics (CFD) gas flow model. To embrace the hybrid ABS model, CFD model, and models for countermeasure in various domains, a scalable runtime infrastructure (RTI)-based simulation environment is also proposed. This environment provides a simulator-level interface to integrate a continuous and discrete-event simulator into the RTI by resolving data/event interaction and time synchronization among heterogeneous simulation models. The authors successfully interoperated the hybrid ABS model, interactive CFD model, control center model, and gas sensor model to evaluate the countermeasures in the proposed environment. As a case study, they applied a 3D-based virtual training engine as a standalone modeling and simulation element to show the scalability of the proposed environment.

Achieving resilience in distributed software systems via self-reconfiguration

Improvements in mobile networking combined with the ubiquitous availability and adoption of low-cost development boards have enabled the vision of mobile platforms of Cyber-Physical Systems (CPS), such as fractionated spacecraft and UAV swarms. Computation and communication resources, sensors, and actuators that are shared among different applications characterize these systems. The cyber-physical nature of these systems means that physical environments can affect both the resource availability and software applications that depend on resource availability. While many application development and management challenges associated with such systems have been described in existing literature, resilient operation and execution have received less attention. This paper describes our work on improving runtime support for resilience in mobile CPS, with a special focus on our runtime infrastructure that provides autonomous resilience via self-reconfiguration. We also describe the interplay between this runtime infrastructure and our design-time tools, as the later is used to statically determine the resilience properties of the former. Finally, we present a use case study to demonstrate and evaluate our design-time resilience analysis and runtime self-reconfiguration infrastructure.

Integrated coupling of road traffic and network simulation for realistic emulation of connected vehicle applications

To meet the requirements of connected vehicle (CV) system simulation evaluations, different simulators (i.e., traffic flow simulation, network communication simulation and CV application simulation) must be combined and interaction must be enabled among them during the run-time of the simulation. This paper proposes a CV simulation prototype system based on integration of three components in one coupling. Considering the characteristics of the distributed interactive simulation of multi-emulators, a high-level architecture simulation and modeling concept is used to construct the simulation framework, and the three types of simulation elements are defined as the primary federates. The integration simulation system is constructed by defining the federation object model and developing the support environment of the run-time infrastructure during simulation. In the simulation implementation procedure, a commercial-off-the-shelf time-stepped VISSIM simulator and an open-source event-driven NS2 simulator are selected as the basic simulation tools to develop the system. In addition, selected related technologies are investigated, including the NS2 simulation model transformation, time synchronization management strategy, track consistency strategy, and integration of application members. Finally, two case studies of intersection speed guidance and collision warning are used to verify the validity of the simulation prototype system.

High-Level Architecture service management for the interoperation of federations

In High-Level Architecture, federates are integrated into a federation in single Run-Time Infrastructure middleware, which handles High-Level Architecture services for interoperation. The interoperation of federations is required in order to combine systems simulated in existing federations. Unlike a single federation, the interoperation of federations provides the data security and makes it possible to interoperate federates developed in different Run-Time Infrastructures. Previous research has used a proxy method to interoperate federations without modifying the Run-Time Infrastructure. Based on the proxy method, this paper proposes a form of High-Level Architecture service management for the interoperation of federations. Federations construct a twotwo–level hierarchical federation that uses proxies to interoperate different levels and that defines the data structure for interoperation and data security. For each High-Level Architecture service protocol, we propose the algorithms of the proxy and show how the service protocol works in the two-level hierarchical federations. In addition, the proxy handles the race condition, which may occur in interoperation of federations, and describes how to solve it. The algorithms are verified theoretically, and a case study shows the practical application for the interoperation of federations.

Quantitative performance metrics for evaluation and comparison of middleware interoperability products

Interoperability in modeling and simulation is understood as the ability of simulations to exchange information and use the information exchanged. Several alternative models have been applied to support interoperability. The traditional approach has been to apply standards, such as the High-level Architecture, to create a federation and use a runtime infrastructure to physically connect simulations. Recently, there has been a move towards web-based standards to loosely couple simulations, and the future points to cloud-enabled interoperability services. Despite the existence of implementations of these models from industry, academia and government, very few performance metrics have been formulated to evaluate them. In this paper, we propose quantitative performance metrics that includes a set of dependent and independent variables for implementations of interoperability models. We apply the metrics to a web-based infrastructure that uses web standards and a traditional runtime infrastructure. We analyze the results and discuss the tradeoffs that federation developers have to consider when selecting an implementation of an interoperability model.

HLA/RTI를 이용한 워게임 시뮬레이터와 통신 효과 시뮬레이터의 연동 시뮬레이션 연구

Wargame simulators are widely used in the field of defence modeling and simulation. Because of increasing importance of communication effects on the warfare, the war game simulator is also required to reflect communication effects. One way to satisfy the requirement is the interoperation simulation between war game simulator and communication effect simulators. This paper shows the application of interoperation simulation between war game and communication effect simulators using HLA(High-Level Architecture)/RTI(RunTime Infrastructure). The war game simulator mainly deal with the engagement of troops and the troops communicate each other at the mission execution level. In the other hand, The communication effect simulator perform communication actions between the troops in the engineering level. Using the interoperation simulation, we can reflect the communication effects on the war game simulation. We show various applications of the interoperation simulation with the point of the war game and communication effect simulator. with a case study, we explain the interoperation simulation improves the reality and fidelity of the war game simulator and how the interoperation simulation can be applied to developing doctrines and real communication system.

Telemetry data processing flow model: a case study

Purpose – The purpose of this paper is to develop the model of telemetry data processing flow (TDPF) for TDPF development and the TDPF run-time infrastructure to improve the spacecraft health monitoring capability. Design/methodology/approach – This research tries to develop the TDPF by flow-based programming (FBP) method and the component-based telemetry data processing software. Findings – The result from the case study is positive, thus reflecting the appropriateness of the suggested method. Practical implications – Application of the proposed TDPF model and the component-based telemetry data processing software may result in improved development efficiency and less development costs. Originality/value – This paper provides an effective way to develop TDPF without recompiling the software. It greatly facilitates the TDPF development that hopefully will save the TDPF development cost.

MESCOS—A Multienergy System Cosimulator for City District Energy Systems

This work introduces a multidomain simulation platform that enables a holistic analysis of city district scale energy systems. The objective for the development of the simulation platform is to provide a tool that supports the design of control and energy management algorithms for those systems. The platform allows long-term simulations of a large number of buildings, including internal energy supply or energy conversion systems, in combination with external energy supply systems like the electrical grid. The simulation of those physical systems represents the environment for sophisticated control and energy management algorithms that can be tested on the platform. The concept of this work is to combine commercial-off-the-shelf software packages, here simulators and runtime infrastructure (RTI), to a high performance multidomain cosimulation platform. The high performance of the platform regarding computation time has been achieved by exploiting the parallel computing capabilities of modern simulation servers. Especially the computation time of large numbers of instances of Modelica-based models has been reduced significantly by the development of the parallel execution framework (PEF). The implementation of the PEF, including the interface to the individual models and to the RTI, is described in detail. The partitioning of the simulated system among different simulators does not influence the simulation results, as shown on the basis of a small-scale simulation scenario. The performance regarding the computation time is demonstrated on several example simulation scenarios showing the scalability of the platform.

Development of Energy-aware Mobile Applications Based on Resource Outsourcing

Smart connected devices such as smartphones and tablets are battery-operated to facilitate their mobility. Therefore, low power consumption is a critical requirement for mobile hardware and for the software designed for such devices. In addition to efficient power management techniques and new battery technologies based on nanomaterials, cloud computing has emerged as a promising technique for reducing energy consumption as well as augmenting the computational and memory capabilities of mobile devices. In this study, we designed and implemented a framework that allows for the energy-efficient execution of mobile applications by partially offloading the workload of a mobile device onto a resourceful cloud. This framework comprises a development toolkit, which facilitates the development of mobile applications capable of supporting computation offloading, and a runtime infrastructure for deployment in the cloud. Using this framework, we implemented three different mobile applications and demonstrated that considerable energy savings can be achieved compared with local processing for both resource-intensive and lightweight applications, especially when using high-speed networks such as Wi-Fi and Long-Term Evolution.