Search

DistributedCL is a middleware which enables transparent parallel processing on distributed GPUs. With the support of the DistributedCL middleware an application designed to use the OpenCL API can run in a distributed manner and transparently use remote GPUs without having to change or rebuild the code. The proposed architecture for the DistributedCL middleware is modular, with well-defined layers. A prototype was built according to the architecture, which considered various optimization points, including sending data in batches, network asynchronous communication and asynchronous request to the OpenCL API. The prototype was evaluated using available benchmarks and a specific benchmark, the CLBench, was developed to facilitate the evaluations according to the amount of processed data. The prototype presented good performance, higher when compared to similar proposals, which also consider transparent use of remote GPUs. The data size to be transmitted over the network was the major limiting factor. [ABSTRACT FROM AUTHOR]

Industry 4.0 has increased the interest in employing Industrial Wireless Sensor Network (IWSN) technologies in industrial automation. The advantages range from ease of installation and maintenance to reduced deployment time and infrastructure costs. However, industrial automation has critical requirements regarding network infrastructure, such as reliability and failure tolerance. Therefore, it is imperative to have an adequate placement of sensor and router nodes, to obtain a network with multiple paths, allowing the data to reach management systems within a reasonable time, even in the event of failures. The placement of router nodes has to consider latency, network lifespan, connectivity, and failure tolerance aspects in a possibly hostile environment, with classified areas and obstacles such as silos, tanks and buildings. We present a new approach, called POSIMNET-R, to place IWSN routing nodes in an industrial configuration, which circumvents forbidden areas and obstacles, based on Artificial Immunological Networks. The resulting network offers low failure rates and path redundancy criteria. The results have shown that POSIMNET-R was capable of providing a reliable network with multiple paths and resilience of the used routers equal to 81.50% in the basic case study and 73.66% in the real case scenario. • POSIMNET-R produces reliable networks inspired by Artificial Immune Systems. • The approach considers industrial sites with obstacles and classified areas. • Produced topologies have k-paths, maintaining network resilience. • POSIMNET-R places routers to make latency and power consumption acceptable. [ABSTRACT FROM AUTHOR]

Background: The challenges faced by caregivers of the elderly with chronic diseases are always complex. In this context, mobile technologies have been used with promising results, but often have restricted functionality, or are either difficult to use or do not provide the necessary support to the caregiver - which leads to declining usage over time. Therefore, we developed the Mobile System for Elderly Monitoring, SMAI. The purpose of SMAI is to monitor patients with functional loss and to improve the support to caregivers' communication with the health team professionals, informing them the data related to the patients' daily lives, while providing the health team better tools.Method: SMAI is composed of mobile applications developed for the caregivers and health team, and a web portal that supports management activities. Caregivers use an Android application to send information and receive care advice and feedback from the health team. The system was constructed using a refinement stage approach. Each stage involved caregivers and the health team in prototype release-test-assessment-refinement cycles. SMAI was evaluated during 18 months. We studied which features were being used the most, and their use pattern throughout the week. We also studied the users' qualitative perceptions. Finally, the caregiver application was also evaluated for usability.Results: SMAI functionalities showed to be very useful or useful to caregivers and health professionals. The Focus Group interviews reveled that among caregivers the use of the application gave them the sensation of being connected to the health team. The usability evaluation identified that the interface design and associated tasks were easy to use and the System Usability Scale, SUS, presented very good results.Conclusions: In general, the use of SMAI represented a positive change for the family caregivers and for the NAI health team. The overall qualitative results indicate that the approach used to construct the system was appropriate to achieve the objectives. [ABSTRACT FROM AUTHOR]