How rstarzl/cmu-lib works
Project type and purpose
• This is an open-source, multi-module “CMU-Lib” toolkit for high-performance big-data processing. It bundles:
• MATLAB data-generation and SVD prototypes (Data_Generator, SVD_matlab)
• A core Java matrix library (CoolMatrixUtility) built atop MTJ
• A Java communication middleware (communication) for Master–Slave messaging
• A Java data-transmit middleware (data_transmit_midware) for buffering, caching and request handling
• A dummy distributed system simulator (DummySystem) to drive and test the above libraries
• A simple static website for demos
File structure
• Top-level modules correspond to functional layers.
• Data_Generator, SVD_matlab: MATLAB scripts for input data and prototyping
• communication: low-level packet/Message/MiddleWare API
• data_transmit_midware: higher-level controller/model/utils on top of communication
• CoolMatrixUtility: core matrix classes (Mat, MatOp) and SVD decomposition distributed across Master/Slave
• DummySystem: sample Master/Slave applications using both middlewares and CoolMatrixUtility
README
• Confirms positioning as “Next Generation Big Data Platform,” focusing on performance and ease-of-use.
Final advice
• Interactions: method calls, network packets, file I/O.
• Build artifacts (JARs, .m scripts) as endpoints.
• Keep it high-level but include all essential layers to reflect the toolkit’s modular design.