Monday 16th of October 2017
 

Real-time Data Stream Processing:Challenges and Perspectives


Soumaya Ounacer, Mohamed Amine Talhaoui, Soufiane Ardchird, Abderrahmane Daif and Mohamed Azouazi

Nowadays, with the recent evolution of sensor technologies, wireless communications, powerful mobile devices and other real-time sources, the way to process the high-speed and real-time data stream brings new challenges. These challenges of big data systems are to detect, anticipate and predict information with the finest granularity possible. The problem is that the system relies on batch processing, which can give great insight into what has happened in the past; however, they do not have the capacity to deal with what is happening at the moment, and of course it is crucial to process events as they happen to a real-time preview. Many applications like fraud detection, data analytics, and production inspection need a fast response and processing time since big data system is based on the MapReduce framework that can only process a finite set of data, is not suitable for the processing of data stream and is inappropriate to satisfy the constraints in real time. Hence the need for a real-time data stream processing system, since it is fast and processes the data in a minimal time with low latency. This paper gives a clear comparison among the different systems that exist for real time data stream processing as well as a model that was based on the comparison that was conducted before.

Keywords: Real-time processing, MapReduce, Spark, Storm, Lambda architecture, kappa architecture.

Download Full-Text


ABOUT THE AUTHORS

Soumaya Ounacer
Université hassan II. Faculté des sciences ben msik casablanca.

Mohamed Amine Talhaoui
Université hassan II. Faculté des sciences ben msik casablanca.

Soufiane Ardchird
Université hassan II. Faculté des sciences ben msik casablanca.

Abderrahmane Daif
Université hassan II. Faculté des sciences ben msik casablanca.

Mohamed Azouazi
Université hassan II. Faculté des sciences ben msik casablanca


IJCSI Published Papers Indexed By:

 

 

 

 
About IJCSI

IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us
FAQs

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482
Email: info@ijcsi.org

More contact details »