Big Data Science Lab
Part of the TU Delft - IBM CIC on Big Data Science is the Big Data Science Lab with:
P7+ cluster for big data
An advanced big data cluster based on the IBM Power7+ processor chips, running a Linux operating system. The cluster consists of 5 nodes, each 2 Power7+ processors (8 cores per processor), sporting 128 GB of memory and 10 Gb Ethernet connectivity. This cluster is used for education as well as research purposes. Peter Hofstee, IBM and part time chair professor focusing on Big Data Systems, on Memory coherent FPGA extensions for power 7
IBM big data software (Infosphere BigInsights and Streams)
These tools are being used to teach students at TU Delft the basics of Hadoop, MapReduce, Hive, etc. The courses are hosted and distributed on an open platform at the bigdatauniversity.com. These courses include hands-on labs that can be performed on especially made VMWare images, or by locally installing the required software.
IBM® Bluemix™ is the IBM open cloud platform that provides developers access to IBM software for integration, security, transaction, and other key functions, as well as software from business partners. Built on the Cloud Foundry open source technology, Bluemix offers more control to application developers by using its Platform as a Service (PaaS) offering, and also provides pre-built Mobile Backend as a Service (MBaaS) capabilities. The goal is to simplify the delivery of an application by providing services that are ready for immediate use and hosting capabilities to enable internal scale development. With the broad set of services and runtimes in Bluemix, the developer gains control and flexibility, and has access to various data options, from predictive analytics to big data.
In teaching DDS uses tools for Big Data Processing and Web Data Management. IBM solutions provide a computational infrastructure for data gathering and analysis and a hosting platform for experimental systems and application.
Remotely accessible Power8 servers for next-generation big data research
IBM provides access to Power8 servers located in the US, remotely accessible to TU Delft. The servers are equipped with various new hardware technologies such accelerators, and host state-of-the-art big data systems (e.g. Spark), that can be used for instance to conduct research on effective in-memory big data analytics.