PRUN is a cluster job scheduler and parallel task executor system for Linux/Unix systems. Like other full-featured batch systems, it provides a job queueing mechanism, job scheduling, a priority scheme, resource monitoring, and resource management. PRUN is a simple, easy-to-use, full-featured, stable solution to run short-term or long-term jobs across the cluster.
io-util is a small, scalable Java library for slicing and dicing fixed width tables on disk. The objective is to provide reusable blocks of code for building efficient, custom binary data stores. It allows you to build, search, and maintain a large, externally stored, fixed width, sorted table. The library user specifies the row width (in bytes), a row comparison function (which implicitly defines any given row's key), and an optional delete codec.
Erudite is an application for training and testing back propogation neural networks using the ANNeML (Artifical Neural Network Markup Language) XML format. It supports testing and training neural nets with CSV files and has support for randomized training sets, optional adapting learning rate, sigmoid or hyperbolic tangent transfer functions, optional bias and weight adjustment locking, and more.
KeyBox is a Web-based SSH console for executing commands and managing multiple systems simultaneously. It allows you to share terminal commands and upload files to all your systems. Once the sessions have been opened you can select a single system or any combination on which to run your commands. Also, additional system administrators can be added and their terminal sessions and history can be audited.
Weed-FS is a simple and highly scalable distributed file system. There are two objectives: to store billions of files, and to serve the files fast! Instead of supporting full POSIX file system semantics, it implements only a key-file mapping. Instead of managing all file metadata in a central master, it manages file volumes in the central master and lets volume servers manage files and the metadata. This relieves concurrency pressure from the central master and spreads file metadata into volume servers' memories, allowing faster file access with just one disk read operation. It is modelled on Facebook's Haystack design paper. Only 40 bytes of disk storage are required for each file's metadata, and disk reads are O(1).
uma::bson is a DOM-style C++ API for reading/writing BSON data. Unlike the MongoDB C++ API, which exposes a read-only interface with a separate interface for creating a BSON representation, this API allows reading/writing on the existing data. The API is designed primarily for serialising/deserialising BSON data to/from streams (files, socket connections, etc.).