Search

Fil: Hirsch Jofre, Matias Eberardo. Consejo Nacional de Investigaciones Cientificas y Tecnicas. Centro Cientifico Tecnologico Conicet - Tandil. Instituto Superior de Ingenieria del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingenieria del Software; Argentina

The computing resources of today’s smartphones are underutilized most of the time. Using these resources could be highly beneficial in edge computing and fog computing contexts, for example, to support urban services for citizens. However, new challenges, especially regarding job scheduling, arise. Smartphones may form ad hoc networks, but individual devices highly differ in computational capabilities and (tolerable) energy usage. We take into account these particularities to validate a task execution scheme that relies on the computing power that clusters of mobile devices could provide. In this paper, we expand the study of several practical heuristics for job scheduling including execution scenarios with state-of-the-art smartphones. With the results of new simulated scenarios, we confirm previous findings and better comprehend the baseline approaches already proposed for the problem. This study also sheds some light on the capabilities of small-sized clusters comprising mid-range and low-end smartphones when the objective is to achieve real-time stream processing using Tensorflow object recognition models as edge jobs. Ultimately, we strive for industry applications to improve task scheduling for dew computing contexts. Heuristics such as ours plus supporting dew middleware could improve citizen participation by allowing a much wider use of dew computing resources, especially in urban contexts in order to help build smart cities.

Within the main goals of Dew Computing is improving users’ experience via on-premise nearby computers while reducing to the minimum the dependency on remote infrastructures. Moreover, smartphones are one of the commonest, most affordable and versatile personal devices exploited by end users everyday, which can in turn be employed as on-premise computers themselves. Nowadays the market offers smartphones with varied capabilities. Hence, studying their software and hardware properties is crucial to Dew computing since they help Dew researchers and app developers to get the most out of these devices. Studies frequently involve the configuration of real smartphones to obtain profile data from the execution of custom codes (benchmarks or profiling routines) under certain conditions (device charging or discharging, initial battery level, baseline CPU usage, display state, etc.). The purposes of such data capturing procedures range from describing component behavior (e.g., battery depletion rate) to determining computing power, in order to build holistic device capabilities rankings. However, few to none supports exist to automate this tedious procedure. For this reason, we present Motrol 2.0, a hardware/software platform that allow Dew researchers and developers to automate benchmarking and profiling tests on Android-based smartphones.

Mobile grid computing has been a popular topic for researchers due to mobile and IoT devices’ ubiquity and their evergrowing processing potential. While many scheduling algorithms for harnessing these resources exist in the literature for standard grid computing scenarios, surprisingly, there is little insight into this matter in the context of hybrid-powered computing resources, typically found in Dew and Edge computing environments. This paper proposes new algorithms aware of devices’ power source for scheduling tasks in hybrid environments, i.e., where the battery- and non-battery-powered devices cooperate. We simulated hybrid Dew/Edge environments by extending DewSim, a simulator that models battery-driven devices’ battery behavior using battery traces profiled from real mobile devices. We compared the throughput and job completion achieved by algorithms proposed in this paper using as a baseline a previously developed algorithm that considers computing resources but only from battery-dependent devices called Enhanced Simple Energy-Aware Schedule (E-SEAS). The obtained results in the simulation reveal that our proposed algorithms can obtain up to a 90% increment in overall throughput and around 95% of completed jobs in hybrid environments compared to E-SEAS. Finally, we show that incorporating these characteristics gives more awareness of the type of resources present and can enable the algorithms to manage resources more efficiently in more hybrid environments than other algorithms found in the literature.

Hardware improvements of mobile devices have allowed them to be considered as first-class resources at the edge, which has led to the need to propose new techniques and scheduling heuristics to make efficient usage of these devices. Current scheduling heuristics available in the literature do not take into account mobile grid nodes that do not rely on batteries to work, a factor that is of great importance in the advent of mobile-edge computing. In this paper we propose a simple algorithm that incorporates a more advanced awareness of a mobile device's power supply and we show that incorporating this characteristic can enable these heuristics to manage edge resources in a more efficient way.

In-laboratory mobile device data gathering is useful to support fields of study that rely on data derived from mobile devices as elementary research input. Particularly, Dew Computing, a sub-area of mobile distributed computing, aims at scavenging idle computing resources from mobile devices at the edge. To produce repeatable experiments for developed Dew approaches, simulation of relevant mobile device aspects is an acceptable practice, being battery behavior one of such aspects. Our recently-proposed DewSim simulation toolkit uses a trace-based approach to model battery behavior realistically. However, to generically characterize the impact of different device components – e.g., CPU at different usages – on battery behavior, it is necessary to easily capture battery traces, and run benchmarks to quantify computing capabilities. Considering that traces are captured during long charging or discharging cycles, such data gathering duty is tedious and time-consuming and no tool has been proposed yet to automate it. To fill this gap, we propose a platform that leverages common IoT hardware to control battery state of devices subject to pre-configured profiling/benchmarking plans. The platform has a server-side component to manage benchmark/profiling executions using one out of two possible operation modes (exclusive or shared), and an extensible Android application that implements the benchmark and profiling logic to be run on devices. We conclude that the operation modes represent a clear trade-off between benchmark/profile execution time and IoT hardware cost. From validation experiments, we also conclude that using our platform to run a benchmark does not introduce a considerable performance and energy footprint compared to running the same benchmark as a plain Android application.

The ever-growing adoption of smart mobile devices is a worldwide phenomenon that positions smartphones and tablets as primary devices, i.e., that people mostly use. In addition to this, the computing capabilities of such devices, often under-utilized by their owners, are in continuous improvement. Today, smart mobile devices have multi-core CPUs, several gigabytes of RAM, and the ability to communicate through several wireless networking technologies. These facts caught the attention of researchers who propose to leverage smart mobile devices aggregated computing capabilities for running resource intensive software at the edge of the network. Such idea is conditioned by key features, named singularities in the context of this work, that makes smart mobile devices resource exploitation a difficult problem. These are the ability of devices to change location (user mobility), the shared condition -i.e., non-dedicated nature- of resources provided (lack of ownership) and the limited operation time given by the finite energy source (exhaustible resources). In this paper, we provide an in-depth analysis of proposals materializing this idea. We show that (a) existing approaches differ in the singularities combinations they target and the way they address each singularity through novel taxonomies, and (b) this fact makes them suitable for distinct goals and resource exploitation opportunities also schematized in this paper. The latter are represented by real life situations where resources provided by groups of smart mobile devices can be exploited, which in turn, are characterized by a social context and a networking support used to link and coordinate devices. The behavior of people in a given social context configure a special availability level of resources, while the networking support imposes restrictions on how data/computational tasks are distributed and results are collected. We conclude our analysis by discussing strong/weak points of the approaches and by identifying prospective future lines in the area. Fil: Hirsch Jofré, Matías Eberardo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Mateos Diaz, Cristian Maximiliano. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Zunino Suarez, Alejandro Octavio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina

To expose software as Web-accesible services, Web Service technologies demand developers to implement certain sofware artifacts, such as the service description using WSDL. Therefore, developers usually use automatic tools to perform this task, which take as input a code written in a programming language –e.g. Java– and generate the necessary artifacts for invoking it remotely. However, as a result of tool flaws and some bad coding practices, the description of the resulting Web Services might contain anti-patterns that difficult their discovery and use. In earlier work we proposed a tool-supported, code-first approach named Gapidt to develop Web Services in Java while early reducing the presence of anti-patterns in their descriptions through code refactorings. Bad coding practices, which potentially reduce the textual and structural information of generated WSDL documents, are automatically detected and informed to the developer by means of the GAnalyzer module so he/she can fix the service code. Moreover, developer provided information, such as service parameter names and operation comments, as well as re-utilization of data-type definitions, are exploited by the GMapper module upon generating WSDL documents. This paper focuses on a comprehensive experimental evaluation of the approach oriented at prospective users to assess expected discoverability gains and usage considerations taking into account various relevant service publishing technologies. In addition, we introduce a detailed comparison of Gapidt with a similar approach from the literature. The results show that Gapidt outperforms its competitor in terms of discoverability while improves Web Service description quality (better documentation and data-models). The Web Service discoverability levels of Gapidt outperforms that of third-party tools, either when using the GAnalyzer plus the GMapper, or only the GMapper. Fil: Hirsch Jofré, Matías Eberardo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Rodriguez, Juan Manuel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Rodriguez, Juan Manuel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Mateos Diaz, Cristian Maximiliano. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Zunino Suarez, Alejandro Octavio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina

Given mobile devices ubiquity and capabilities, some researchers now consider them as resource providers of distributed environments called mobile Grids for running resource intensive software. Therefore, job scheduling has to deal with device singularities, such as energy constraints, mobility and unstable connectivity. Many existing schedulers consider at least one of these aspects, but their applicability strongly depends on information that is unavailable or difficult to estimate accurately, like job execution time. Other efforts do not assume knowing job CPU requirements but ignore energy consumption due to data transfer operations, which is not realistic for data-intensive applications. This work, on the contrary, considers the last as non negligible and known by the scheduler. Under these assumptions, we conduct a performance study of several traditional scheduling heuristics adapted to this environment, which are applied with the known information of jobs but evaluated along with job information unknown to the scheduler. Experiments are performed via a simulation software that employs hardware profiles derived from real mobile devices. Our goal is to contribute to better understand both the capabilities and limitations of this kind of schedulers in the incipient area of mobile Grids.

The profusion of mobile devices over the world and their evolved computational capabilities promote their inclusion as resource providers in traditional Grid environments. However, their efficient exploitation requires adapting current schedulers to operate with computing capabilities limited by energy supply and mobile devices that cannot be assumed to be dedicated, among other concerns. We propose a two-phase scheduling approach for running CPU-intensive jobs on mobile devices that combines novel energy-aware criteria with job stealing techniques. The approach was evaluated through an event-based simulator that uses battery consumption profiles extracted from real mobile devices. CPU usage derived from non-Grid processes was also modelled. For evaluating the first phase we compared the number of finalized jobs by all energy-aware criteria, while for the second phase we analyzed the performance boost introduced by job stealing. While the best first phase criteria finalized up to 90 % of submitted jobs, job stealing increased this percentage by up to 9 %. Fil: Hirsch Jofré, Matías Eberardo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Rodriguez, Juan Manuel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Mateos Diaz, Cristian Maximiliano. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Zunino Suarez, Alejandro Octavio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina

With self-provisioning of resources as premise, dew computing aims at providing computing services by minimizing the dependency over existing internetwork back-haul. Mobile devices have a huge potential to contribute to this emerging paradigm, not only due to their proximity to the end user, ever growing computing/storage features and pervasiveness, but also due to their capability to render services for several hours, even days, without being plugged to the electricity grid. Nonetheless, misusing the energy of their batteries can discourage owners to offer devices as resource providers in dew computing environments. Arguably, having accurate estimations of remaining battery would help to take better advantage of a device&rsquo, s computing capabilities. In this paper, we propose a model to estimate mobile devices battery availability by inspecting traces of real mobile device owner&rsquo, s activity and relevant device state variables. The model includes a feature extraction approach to obtain representative features/variables, and a prediction approach, based on regression models and machine learning classifiers. On average, the accuracy of our approach, measured with the mean squared error metric, overpasses the one obtained by a related work. Prediction experiments at five hours ahead are performed over activity logs of 23 mobile users across several months.

Fil: Mateos Diaz, Cristian Maximiliano. Consejo Nacional de Investigaciones Cientificas y Tecnicas. Centro Cientifico Tecnologico - CONICET - Tandil. Instituto Superior de Ingenieria del Software; Argentina

Grid Computing allows scientists and engineers to run compute intensive experiments that were unfeasible not so long ago. On the downside, for users not proficient in distributed technologies, programming for Grids is difficult, tedious, time-consuming and error-prone. Then, disciplinary users typically waste precious time that could be instead invested into analyzing results. In a previous paper, we introduced BYG (Mateos et al., 2011) [28], a Java-based software that automatically parallelizes sequential applications by directly modifying their compiled codes. In addition, BYG is designed to harness Grid resources by reusing existing Grid platforms and schedulers. In its current shape, however, BYG lacks support for some state-of-the-art Grid schedulers and mechanisms for introducing application-dependent optimizations to parallelized codes. In this paper, we present several extensions to BYG aimed at overcoming these problems and thus improving its applicability and delivered efficiency. We also report experiments by using traditional computational kernels and real-life applications to show the positive practical implications of the proposed extensions. Fil: Mateos Diaz, Cristian Maximiliano. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Zunino Suarez, Alejandro Octavio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Hirsch Jofré, Matías Eberardo. Universidad Nacional del Centro de la Provincia de Buenos Aires. Facultad de Ciencias Exactas; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina Fil: Fernández, Mariano. Universidad Nacional del Centro de la Provincia de Buenos Aires. Facultad de Ciencias Exactas; Argentina

Computational Grids deliver the necessary computational infrastructure to perform resource-intensive computations such as the ones that solve the problems scientists are facing today. Exploiting Computational Grids comes at the expense of explicitly adapting the ordinary software implementing scientific problems to take advantage of Grid resources, which unavoidably requires knowledge on Grid programming. The recent notion of "gridifying" ordinary applications, which is based on semi-automatically deriving a Grid-aware version from the compiled code of a sequential application, promises users to be relieved from the requirement of manual usage of Grid APIs within their source codes. In this paper, we describe a novel gridification tool that allows users to easily parallelize Java applications on Grids. Extensive experiments with two real-world applications - ray tracing and sequence alignment - suggest that our approach provides a convenient balance between ease of gridification and Grid resource exploitation compared to manually using Grid APIs for gridifying ordinary applications. Fil: Mateos Diaz, Cristian Maximiliano. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Zunino Suarez, Alejandro Octavio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina Fil: Hirsch Jofré, Matías Eberardo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil; Argentina. Universidad Nacional del Centro de la Provincia de Buenos Aires; Argentina Fil: Fernández, Mariano. Universidad Nacional del Centro de la Provincia de Buenos Aires; Argentina Fil: Campo, Marcelo Ricardo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tandil. Instituto Superior de Ingeniería del Software. Universidad Nacional del Centro de la Provincia de Buenos Aires. Instituto Superior de Ingeniería del Software; Argentina

Although producing high quality WSDL documents is a key factor in a Web Service system success, previous works have shown that WSDL document quality, in terms of readability and discoverability, is often disregarded by service developers. One of the main causes of this is that developers do not write the WSDL documents directly. Instead, these documents are automatically generated from service implementations, which are called code-first Web Services. This work presents a novel tool for assisting Java code-first Web Service development that spots potential issues that might lead to low quality WSDL documents. The approach was empirically evaluated using 81 open-source Web Service implementations. The evaluation shows that the WSDL documents generated using our tool can be more easily retrieved from Web Service registries by users. This indirectly implies that these WSDL documents have a better quality than the ones generated using alternative tools.

We are facing a hardware revolution given by the increasing availability of multicore computers, clusters, Grids, and combinations of these. Consequently, there is plenty of computational power, but today's programmers are not fully prepared to exploit parallelism and distribution. Particularly, Java has helped in handling the heterogeneity of such environments, but there is a lack of facilities to easily and elegantly parallelizing applications. One path to this end seems to be the synthesis of semi-automatic parallelism and Parallelism as a Concern (PaaC). We briefly survey relevant Java-based parallel development tools, identify their drawbacks, and discuss some insights of an ongoing approach for mitigating them.