A Big-Data Centric Framework for Smart Systems in the World of Internet of Everything Pedro A. Pena Computer Science Department Univ. of Miami, Coral Gables, USA Email: p.pena3@umiami.edu Dilip Sarkar Computer Science Department Univ. of Miami, Coral Gables, USA Email: sarkar@cs.miami.edu Parul Maheshwari Computer Science Department Univ. of Miami, Coral Gables, USA Email: p.maheshwari@umiami.edu Abstract—Current research and development efforts for integrating Everything — heterogeneous objects, devices, and people — in the Internet have advanced network technologies and protocols to develop the Internet of Everything (IoE). The emergence of smart systems such as Smart Cities and Smart Traffic Management Systems have been proposed to gather data from relevant sources via IoE. Recently, cloud computing has been introduced into the IoE paradigm to provide a multitude of services necessary to compose a smart system that utilizes big data. In this paper, we propose a big-data centric framework for smart systems that encompasses four phases: data collection, processing, management, and interpretation. The proposed framework facilitates a modular architecture for smart systems where security and cognition are interactive modules. Keywords-Cloud Computing; Cognition with Knowledge; Internet of Everything; Security; Smart Systems; I. I NTRODUCTION Currently, research on networking tools and protocols for creating Internet of Everything (IoE) have gained significant attention [1], [2], [3]. The growth of data from IoE is exponential, and it is putting a strain on IT infrastructure. Architectures for smart cities have been proposed using IoTs in [4], [5]. These smart systems will add big data to today’s 2.7 zetabytes of data 1 . As smart systems gather data from all sources, unauthorized access to data is a major concern and precautions are needed to secure data to preserve privacy of smart systems and individuals [6]. In this paper, we propose a generic framework for smart systems that includes cognition for data collection, pro- cessing, management, interpretation, and security. This is a generalization over the system proposed in [7] that has cognition for data management and interpretation only. Trust is a major concern because of all the security breaches in the recent decade, and experts predict an increasing trend. In our proposed framework, we include a module for security to increase safety and trust. We will discuss different methodologies used to protect privacy and sensitivity of data. Cloud computing has been a favorite platform for IoEs [8], because it delivers i) Software as a Service (SaaS), 1 http://wikibon.org/blog/big-data-statistics/ ii) Platform as a Service (PaaS), iii) Data Storage as a Service (DSaaS) iv) Infrastructure as a Service (IaaS) [9]. Unlike past proposals, the building blocks of the proposed framework are cloud services. Utilizing the concept from [7], we propose to organize data into virtual objects for easy storage and management. We introduce a cognition module that gathers information from the overall system to create knowledge for intelligent services, such as detection and prevention of intrusions or malicious attacks. Our aim is to have a generic framework for all smart systems, including Smart City, Smart Hospital, Smart Education, etc. A system derived from the framework is expected to support horizontal integration with other smart systems to understand the world around us, and find interrelationships among them. The rest of the paper is organized as follow: description of IoE enabling technologies is presented in Section II. An overview of a novel framework for smart systems is presented in III. The security framework is presented in Section IV, and then cognition framework is discussed in Section V. Finally we discuss the use of the framework for smart cities in Section VI. II. I OEENABLING TECHNOLOGIES In this section, we provide a very brief overview of the network and the cloud, two central components of IoE. A. Network Networking has received most attention on discussions of IoEs because they enable and provide communication among components, and supports interoperability of smart systems within IoEs. Since the network protocols handle the communication between heterogeneous objects, significant efforts have been dedicated to standardize these protocols. The following section lists the most important protocols by the type of service they provide. An extensive and comprehensive survey on this subject can found in [1]. 1) Network Protocols: • Application Protocols: Data Distribution Service (DDS), CoAP, AMQP, MQTT / MQTT-NS, XMPP, HTTP/REpresentational State Transfer (REST). • Service Discovery: Multicast DNS (mDNS), DNS-SD. 2015 International Conference on Computational Science and Computational Intelligence 978-1-4673-9795-7/15 $31.00 © 2015 IEEE DOI 10.1109/CSCI.2015.62 307 2015 International Conference on Computational Science and Computational Intelligence 978-1-4673-9795-7/15 $31.00 © 2015 IEEE DOI 10.1109/CSCI.2015.62 306