EFEU is suitable for handling data cubes, which are especially useful for building data warehouses. It consists of a building system including mkmf (a xmkmf like Makefile generator), esh (a powerful C/C++ interpreter), efeudoc (a document generator with different output formats (LaTeX, HTML, roff, etc.)). EFEU has a huge set of C library core features, including robust memory allocation tools, functions to concatenate and copy strings with memory allocation, buffers for dynamically-growing strings and fields, data structures with reference counters and garbage collection, and file tools that allow you to specify pipes wherever a filename is expected and support automatic implementation of (de)compression filters (gzip) depending on filename extensions. It also includes a high-level interface to files, strings, and anything else you can read from or write to.
WWWcam is for grabbing, archiving, and Web publishing pictures from a video-device-based Webcam. With the CPiA webcam driver it supports even USB and parallel port webcams. Beside normal functions like connecting wwwcam directly with a browser and automatic picture reload from your Website, it has many special functions. It can write an adjustable title with timestamp in the picture, and it can perform automatic brightness correction. It can also compare the taken picture with last one and decide whether or not to save the picture for archiving. It tells you if there was activity in the camera's field of view. Finally, it does FTP upload and limits the number of pictures in a directory.
Beowulf includes an enhanced Linux kernel, libraries, and utilities specifically designed for clustering. Beowulf provides a single system image through BProc, the Beowulf cluster process management kernel enhancement. BProc makes the processes running on cluster "Computation Node" computers visible and manageable on a front-end "Master Node". Processes start on the front-end node and migrate to a cluster node. Process parent-child relationships and UNIX job control are maintained with migrated tasks. Cluster slave nodes are not required to contain resident applications. Their hard disks are used for application data and cache. This approach eliminates version skew common with previous generation clusters.
The Modular Application Engine (MAPE) allows programmers to write multi-tier applications in multiple programming languages, using different GUIs for the clients, and accessing multiple database systems. Whole applications are stored in the MAPE system catalog which is stored in a supported SQL database server. XML is used for application description. The applications are generally run on the server, but some parts can run on the client.