Often resource requirements on Data Processing Clusters exhibit high variation. So in this case elastic deployments reduce total cost of ownership. In this talk we will introduce the basic concepts on container isolation exemplified on Docker and explain how Apache Flink is made elastic using Docker-Compose. Finally we push the cluster to the cloud exemplified on the IBM Docker Cloud.
About the speaker
Romeo Kienzler is a post-graduate computer scientist with a focus on information systems, bioinformatics and statistical data analysis, and since 2008 at IBM now working as a Data Scientist in the IBM Cloud Transformation Lab for Data Science Zurich
Simon Laws is a member of the IBM Hursley Emerging Technology team. During his 28 years with IBM Simon has undertaken a wide variety of technical roles from lab based product development through to customer focused bespoke solution development. He is currently focusing on building innovative Big Data solutions through the IBM Big Data Stampede team.
Big Data is widely applicable and Simon has been involved in exploiting the approach for many different customer problems including fraud detection, compliance assessment, device monitoring, complaints processing, profiling and several projects requiring the application of machine learning techniques to geospatial problems.
Simon is a member of the Apache Software Foundation