Zoran Babovic

Web Performance Evaluation for Internet of Things Applications

An area of intensive research under the umbrella of the Internet of Things (IoT) has resulted in intensive proliferation of globally deployed sensor devices that provide a basis for the development of different use-case applications working with real-time data and demanding a rich user interface. Overcoming the lack of the standard HTML platform, HTML5 specifications WebSocket and Canvas graphics strongly supported the development of rich real-time applications. Such support has been offered by browser plug-ins such as Adobe Flash and Microsoft Silverlight for years. In order to provide a deep insight into IoT Web application performance, we implemented two test applications. In the first application, we measured latencies induced by different communication protocols and message encodings, as well as graphics rendering performance, while comparing the performance of different Web platform implementations. In the second application, we compared Web performance of IoT messaging protocols such as MQTT, AMQP, XMPP, and DDS by measuring the latency of sensor data message delivery and the message throughput rate. Our tests have shown that although Adobe Flash has the best performance at the moment, HTML5 platform is also very capable of running real-time IoT Web applications, whereas Microsoft Silverlight is noticeably behind both platforms. On the other hand, MQTT is the most appropriate messaging protocol for a wide set of IoT Web applications. However, IoT application developers should be aware of certain MQTT message broker implementation shortcomings that could prevent the usage of this protocol.

Novel System Architectures for Semantic-Based Integration of Sensor Networks

Abstract There are many on-going projects and research initiatives that are proposing new semantic-oriented services for obtaining, delivery, and processing sensor data gathered from integrated sensor networks. Rich information models based on the ontologies for heterogeneous sensor data description are necessary for achieving interoperability among various deployed sensor networks while providing context-related information with raw sensor data. As expected by the Sensor Web vision and the Future Internet initiatives, certain architecture is faced with performance requirements while providing complex services. This study identifies main challenges and design issues of sensor networks integration platforms, and gives a survey of existing approaches specifically emphasizing semantic-oriented approaches. The survey includes non-semantic approaches that are improved by employing semantics on certain levels, as well as approaches fully based on semantic technologies. As their original contribution, the authors propose a new architecture designed as an infrastructural platform for enabling semantic-based sensor networks integration. The key idea behind the proposed innovation is to utilize a flexible distributed repository called column store for keeping semantically modeled sensor data providing a scalable platform capable of supporting huge amounts of sensor data and large numbers of users. Moreover, column store is employed for push-based data propagation and support for including more complex processing elements.

A survey on the use of Mobile Agents in Wireless Sensor Networks

This survey paper describes and classifies solutions/approaches that use Mobile Agents in Wireless Sensor Networks. There are eight identified criteria which are used for classification of available solutions/approaches and include: autonomicity, security, adaptability, availability, routing, filtering, recovery, and configuration. For each criterion, a selected solution/approach is described so that it illustrates principles corresponding to that criterion. Description of each solution is given from the viewpoint of the role which a Mobile Agent has in the Wireless Sensor Network, also shown in a figure per surveyed paper. Every described solution has been discussed in respect of improvements that are implemented and potential drawbacks.

Concept modeling: From origins to multimedia

The origins of concept modeling are in the field of artificial intelligence. This is where the initial algorithms were introduced first. With the emerging developments in the field of multimedia systems, a strong need is generated to examine and implement concepts-based retrieval of multimedia-contents, from large data bases or from the Internet. The early works were based on appropriate modifications of classical approaches. The latest developments utilize the algorithms that make sense only in the case of multimedia systems. This paper presents a number of classical approaches to concept modeling and their applicability to multimedia. Then it discusses a number of approaches introduced specifically for multimedia. Finally it presents an approach which was fully implemented and tested in an academic environment for industry needs.

Chapter Five - A Novel Infrastructure for Synergistic Dataflow Research, Development, Education, and Deployment: The Maxeler AppGallery Project.

Abstract This chapter presents the essence and the details of a novel infrastructure that synergizes research, development, education, and deployment in the context of dataflow research. To make it clearer to fundamental scientists, the essence of the approach is explained by referencing the results of the work of four different Nobel laureates. To make it clearer to research community, crucial details are presented in the form of a manual. Till this point, the development of dataflow applications was based on tools inherited from the controlflow environment. We here describe a set of tools developed from scratch with dataflow specifics in mind. These tools are not only tuned to the dataflow environment, but they are also tuned to synergize with each other, for the best possible performance in minimal time, counting from the moment when new researchers enter the dataflow arena, until the moment when they are able to deliver a quality code for maximal speed performance and minimal energy consumption. The effectiveness of the presented synergetic approach was measured empirically, using a group of students in a dataflow course. The measured results clearly indicate the superiority of the proposed approach in the following five domains: time to design, time to program, time to build, time to test, and speedup ratio.

DataFlow Systems: From Their Origins to Future Applications in Data Analytics, Deep Learning, and the Internet of Things

With the slowdowns in Dennard scaling and limited performance gain in multi-core scaling, we are witnesses of the high-performance computing shift to domain-specific hardware systems which empower big data and high-performance applications. Likewise, dataflow systems are experiencing a revival with both hardware and software approaches widely exploited. In our work, we give an overview of dataflow system origins and similar technologies such as systolic architecture whose principles are applied by some of today’s leading high-performance systems such as Multiscale dataflow Computing (MDC). In the second part, we highlight certain applications that could benefit from delegating critical processing to a MDC system. We emphasize algorithms and applications from data analytics, deep learning, and the Internet of Things (IoT), with a special focus on their execution within the cloud environment. We discuss the integration of software distributed dataflow systems such as Apache Spark with MDC systems, analyze design issues and challenges for implementation of deep neural networks using MDC, and how semantic-enabled IoT platforms and services could be improved by using MDC systems in order to become more effective. We expect that these selected case studies would motivate researchers to investigate engagement of hardware dataflow systems to support applications from other areas with similarly rigid requirements.

Introductory Overview on Implementation Tools

When programming the dataflow engines, developers need to change the way of thinking from the way of thinking used when programming CPUs where a program that gets executed in time is written to thinking about how to write a spatial recipe that will best configure the dataflow engines so that the data gets processed in space flowing through the configured devices.

Binary Search in the DataFlow Paradigm

This chapter represents a study about the binary search algorithm implementation and its usage implications in dataflow paradigm on Maxeler technology. In the chapter you will see the Binary Search algorithm explained and the differences between its implementations on two different architectures. It will be shown that the difference in the amount of data needed to be processed is in connection with the resulting speedup achieved on a Maxeler machine.

Implementing Neural Networks by Using the DataFlow Paradigm

In this chapter we will present one implementation of Neural Networks using dataflow paradigm. The dataflow paradigm presents a new approach to BigData applications. Existence of BigData is one of biggest application problems in many fields: financial engineering, geophysics, medical analysis, air flow simulations, data mining, and many others. Most of these applications are based on Neural Networks, and that being said, the way a network is implemented is crucial for the application performance. Such applications pay more attention to data than to the process itself. In order to be able to perform correct predictions, Neural Networks should be trained first. In some cases, they spend a lot of execution time on training process. The main challenge is finding a way to process such big quantities of data. Regardless of which level of parallelism is achieved, the execution process is essentially slow. In this chapter, the dataflow paradigm is presented as an alternative paradigm in solving this problem.

Solving the Poisson Equation by Using DataFlow Technology

In this chapter we will discuss the problem of solving the three-dimensional Poisson equation. We are going to do it using the Fourier method and dataflow technology. Poisson equation represents a widely used partial differential equation. With usage in physics for computing potential, analysis of team behavior patterns in sport and lifesaving applications like tsunami wave modeling, this equation proved its significance. It represents maybe the most significant member of one big family of partial differential equations. Because of its enormous usage, efficient programs for finding the equation solution should be developed. First of all, the conventional solutions in ControlFlow world will be presented. We will discuss the basic idea and implement it programmatically. For great number of computations, the time needed to find out the equation solution is crucial. Those computations demand efficient methods with strong mathematical base and fast numerical computer solutions. One of those efficient methods will be presented and chosen for implementation. In spite of this method efficiency, when it comes to computations of large data size or enormous number of datasets, ControlFlow solutions are unacceptable to use. Their low performances are caused by classical computer architecture, which only allows programming in time. To overcome this problem, we call dataflow technology for help. Thanks to its ability for efficient big data computations, it provides us a good base for developing our solution. After deeply analyzing ControlFlow solution’s advantages and disadvantages, we will start our journey of chosen algorithm realization in dataflow world. We’re going to move through the crossroads of project decisions as we vigilantly follow the procedure of building the goal solution. Finally, at the end of this chapter, we are giving advices and ideas for improving the solution.