Stream Processing and Real-time analytics has become one of the most important topic in Big Data. The emergence of new business cases have created a need to develop more robust, more performant, more intelligent Stream Processing applications and analytics. Today, we can find use cases in various industries such as Bank with real-time fraud detection, real time credit risk scoring; Insurance with real-time claim quotation, telematics; Retails with real-time marketing; Telecom with adaptive SLA in network, real-time customer care and many other examples.

These last two years, we have seen arriving another usage of Stream & complex event processing: the data management. New architecture patterns have been proposed to resolve data pipeline and data management within enterprise. In [11,12], the authors describe a way to redesign ETL (Extract Transform and Load) using Stream processing.This opened the door to completely redesign the way the data are transported, stored and used within Big Data environment by breaking down silos between EDW and Big Data lake as shown by [13].

In the past years, researchers and practitioners in the area of data stream management [1, 2, 3] and Complex Event Processing (CEP) [4, 5, 6] have developed systems to process unbounded streams of data and quickly detect situations of interest.

Nowadays, big data technologies provide a new ecosystem to foster research in this area. Highly scalable distributed stream processors, the convergence of batch and stream engines and the emergence of state management & statull stream processing (such as Apache Spark [9] or Apache Flink [10]) open new doors for highly scalable and distributed real-time analytics. Going further, those technologies also provide a solid foundation for real-time analytics algorithms that are complementary to the CEP in the use cases required by the industry. Finally, the Stateful nature of Stream Processors [14] allows to apply Stream SQL statement directly in the streaming engine and creating Dynamic tables [16 ,12].

As a result, we encourage submissions studying scalable online learning and incremental learning on stream processing infrastructure. In addition, we also encourage submissions on Data Stream management, data architectures using Stream processing and Internet of Things data streaming, Finally, we also encourage submissions studying the usage of stream processing in new innovative architectures.

After the success of the first and the second edition, this workshop, co-located with the IEEE Big Data 2016 & 2017, this third edition is an excellent opportunity to gather together actors from academia and industry to discuss, to explore and to refine new opportunities and use cases in the area. The workshop will benefit to both researchers and practitioners interested in the latest researches in real-time and stream processing. The workshop will showcase prototypes or products leveraging big data technologies as well as models, efficient algorithms for scalable complex event processors and context detection engines, or new architecture leveraging stream processing.

Research Topics

The topics of interest include but are not limited to:


Fabian Hueske, Data Artisans

Unified Processing of Static and Streaming Data with SQL on Apache Flink.

SQL is the lingua franca of data processing and everybody working with data knows SQL. While in the past, most open-source stream processing frameworks only provided Java or Scala-based APIs, stream processing with SQL is recently gaining a lot of attention because is makes stream processing accessible to a wider audience and significantly reduces the effort to solve common use cases.

About three years ago, the Apache Flink community started working on adding support for SQL. Today, thousands of continuous SQL queries power production systems in Alibaba, Huawei, Lyft, and Uber. Flink follows the approach of leveraging ANSI SQL syntax and semantics for processing static and streaming data. Unified syntax and semantics are important for various reasons, including existing user expertise, query portability, and the ability to efficiently bootstrap query state or backfill results from recorded data in case of failures. In this talk, I will explain Flink's approach in detail and highlight its benefits. Moreover, I'll discuss the challenges that arise when queries are continuously evaluated on infinite input..


To be announced



October 10, 2018
November 1, 2018
November 15, 2018


Your paper should be written in English and formatted to IEEE Computer Society Proceedings Manuscript Formatting Guidelines (Templates). The length of the paper should not exceed 6 pages.

All accepted papers will be published in the Workshop Proceedings by the IEEE Computer Society Press



  • Sabri Skhiri
  • Albert Bifet
    Télécom Paris Tech, FR
  • Alessandro Margara
    Politecnico di Milano, IT


  • Amine Ghrab
  • Fabian Hüske
    Data Artisans, DE
  • Fabricio Enembreck
    Pontifícia Universidade
    Católica do Paraná, BR
  • Guido Salvaneschi
    TU Darmstadt, DE
  • Jian Chen
    University of North Alabama, US
  • José del Campo Ávila
    Universidad de Málaga, ES
  • Nam-Luc Tran
  • Oscar Romero
    UPC Barcelona, ES
  • Peter Beling
    University of Virginia, US
  • Raju Gottumukkala
    University of Louisiana,US
  • Thomas Peel
  • Vijay Raghavan
    University of Louisiana, US