TeaFiles provide a very simple, yet highly efficient way to store time series data in a way that allows data exchange between programs written in C++, C# or in applications like R, Octave, Matlab or others, running on Linux, Unix, Mac OS X or Windows.
TeaFiles store binary data composed from elementary data types signed and unsigned integers, double and float in IEEE 754 format. Data can be directly accessed via memory mapping. Since their header stores a description of the item structure, they relieve the opaqueness of normal binary files and remain always accessible. In other words: Knowing that a file is a TeaFile is enough to access its data.