The server section of the laboratory for big data processing is equipped with a nine-node Hadoop cluster on the Hortonworks (HDP) platform. Here, computations can be performed on large amounts of data ranging from tens to hundreds of TB.
In addition to the distributed HDFS storage, tools such as Apache Spark for ultra-fast in-memory computations, the columnar database HBase, the computation engine Hive, and other tools (e.g., Sqoop, Kafka, Zeppelin, Shark) are available.
In the laboratory, it is possible to conduct analyses of large datasets from IoT devices or precision agriculture technologies, research satellite data from remote sensing, research machine learning and deep learning algorithms for big data, as well as a variety of other computations.