The rapid growth of stream applications in financial markets, health care, education, social media, and sensor networks represents a remarkable milestone for data processing analytic recent years, leading to new challenges handle Big Data real-time. Traditionally, single cloud infrastructure often holds the deployment Stream Processing because it has extensive adaptative virtual computing resources. Hence, sources send from distant different locations infrastructure, increasing application...

The number of Internet Things applications is forecast to grow exponentially within the coming decade. Owners such strive make predictions from large streams complex input in near real time. Cloud-based architectures often centralize storage and processing, generating high data movement overheads that penalize real-time applications. Edge Cloud architecture pushes computation closer where generated, reducing cost movements improving application response heterogeneity among edge devices cloud...

DNN inference is time-consuming and resource hungry. Partitioning early exit are ways to run DNNs efficiently on the edge. balances computation load multiple servers, offers quit process sooner save time. Usually, these two considered separate steps with limited flexibility. This work combines partitioning proposes a performance model estimate both latency accuracy. We use this offer best partitioned/early based deployment information user preferences. Our experiments show that flexibility...

The Internet of Things has enabled many application scenarios where a large number connected devices generate unbounded streams data, often processed by data stream processing frameworks deployed in the cloud. Edge computing enables offloading from cloud and placing it close to is generated, thereby reducing time process events deployment costs. However, edge resources are more computationally constrained than their counterparts, raising two interrelated issues, namely deciding on...

Internet of Things (IoT) applications often require the processing data streams generated by devices dispersed over a large geographical area. Traditionally, these are forwarded to distant cloud for processing, thus resulting in high application end-to-end latency. Recent work explores combination resources located clouds and at edges Internet, called cloud-edge infrastructure, deploying Data Stream Processing (DSP) applications. Most previous work, however, fails scale very IoT settings....

There is increasing demand for handling massive amounts of data in a timely manner via Distributed Stream Processing (DSP). A DSP application often structured as directed graph whose vertices are operators that perform transformations over the incoming and edges representing streams between operators. applications traditionally deployed on Cloud order to explore virtually unlimited number resources. Edge computing has emerged suitable paradigm executing parts by offloading certain from...

This paper explores the CO2 footprint of IoT applications by using system dynamics modeling to estimate emissions over time from a wireless video analytics application. We model impact application design and mobile infrastructure on short long term produced running both cloud edge computing infrastructures. Our analysis shows that base station radio wide-area data network are major contributors emissions. find can be reduced 50% placing centers near stations, exploiting new features 5G...

Chronic pain is often an ongoing challenge for patients to track and collect data. Pain-O-Vision a smartwatch enabled management system that uses computer vision capture the details of painful events from user. A natural reaction clench ones fist. The embedded camera used different types fist clenching, represent levels pain. An initial prototype was built on Android cloud-based classification service detect gestures. Our results show it possible map which allows patient record intensity...

A huge volume of data is produced every day by social networks (e.g. Facebook, Instagram, Whatsapp, etc.), sensors, mobile devices and other applications. Although the Cloud computing scenario has grown rapidly in recent years, it still suffers from a lack kind standardization that involves resource management for Big Data applications, such as case MapReduce. In this context, users face big challenge attempting to understand requirements application how consolidate resources properly. This...

Stream Processing Engines (SPEs) have to support high data ingestion ensure the quality and efficiency for end-user or a system administrator. The flow processed by SPE fluctuates over time, requires real-time near resource pool adjustments (network, memory, CPU other). This scenario leads problem known as skewed production caused non-uniform incoming at specific points on environment, resulting in slow down of applications network bottlenecks inefficient load balance. work proposes Aten...

Today, BSP (Bulk-Synchronous Parallel) represents one of the most often used models for writing tightly-coupled parallel programs. As resource substrates, commonly clusters and eventually computational grids are to run applications. In this context, here we investigate use collaborative computing idle resources execute kind demand, so proposing a model named BSPonP2P answer following question: How can develop an efficient viable applications on P2P Desktop Grids? We it by providing both...

Next generation applications such as augmented/vir-tual reality, autonomous driving, and Industry 4.0, have tight latency constraints produce large amounts of data. To address the real-time nature high bandwidth usage new applications, edge computing provides an extension to cloud infrastructure through a hierarchy datacenters located between devices cloud. Outside closer edge, network becomes more dynamic requiring stream processing frameworks adapt frequently. Cloud based very slowly...

As IoT devices multiply and produce vast volumes of data, there is a heightened demand for instantaneous data processing. However, traditional cloud computing cannot adequately address these demands due to its latency bandwidth limitations. Edge has emerged as viable alternative with hierarchical deployment datacenters. this introduces additional layers infrastructure management that increase application development complexity. Using shared file system an attractive method enhancing...

Under several emerging application scenarios, such as in smart cities, operational monitoring of large infrastructure, wearable assistance, and Internet Things, continuous data streams must be processed under very short delays. Several solutions, including multiple software engines, have been developed for processing unbounded a scalable efficient manner. More recently, architecture has proposed to use edge computing stream processing. This paper surveys state the art on engines mechanisms...