The course starts with basic concepts and components of big data systems. Requirements for big data systems are motivated by the CAP theorem and the lambda architecture. The map-reduce paradigm is used to illustrate batch-processing. Stream processing is explained using Storm. Spark is presented for both processing modes.