An Open-Source Cloud Architecture for Big Stream IoT Applications - Interoperability and Open-Source Solutions for the Internet of Things - page 73

point manages the execution of sub-units in a certain order and each sub-unit
provides a specific processing output, which is created to be used only within
the scope of its own process without the possibility to be shared among different
processes. This approach represents a major deviation from traditional Service
Oriented Architectures (SOAs), where the sub-units are external web services
invoked by a coordinator process rather than internal services [ 6 ]. Traditional
Big Data approaches might cause higher processing latencies since they are not
optimized for real-time processing tasks.
Big Stream-oriented systems should react effectively to changes and provide
smart behavior for allocating resources, thus implementing scalable and cost-
effective Cloud services. Dynamism and real-time requirements are the reasons
why Big Data approaches, due to their intrinsic inertia (i.e., Big Data typically
works with batch-based processing), are not suitable for many IoT scenarios. The
Big Stream paradigm allows to perform real-time and ad-hoc processing in order
to link incoming streams of data to consumers, with a high degree of scalability,
fine-grained and dynamic configuration, and management of heterogeneous data
formats. In brief, while both Big Data and Big Stream deal with massive amounts
of data, the former focuses on the analysis of data, while the latter focuses on
the management of flows of data, as shown in Fig. 1 . The main difference resides
in the meaning of the term “Big”: for Big Data it refers to volume of data while
for Big Stream it refers to global information generation rate as generated by
data sources. Additionally, for Big Data applications it is important to keep a
history of sensed data in order to be able to perform any required computation,
Big Stream applications might decide to perform data aggregation or pruning
in order to minimize the latency in conveying the results of computation to
consumers, with no need for persistence. Note that, as a generalization, Big
Data applications might be consumers of Big Stream data flows.
In this paper, we present an implementation of a novel Cloud architecture for
Big Stream applications based on standard protocols and open-source compo-
nents, which provides a scalable and ecient processing platform for IoT appli-
cations. The architecture has been designed to be open and extensible and to
decrease the latency between data generation and consumption. In order to assess

(a) (b)

Fig. 1. (a) The volume of data analysis in Big Data systems. (b) The multiple data
sources and listeners management in Big Stream system.