VectorPU is a light-weight vector abstraction on CUDA-based heterogeneous system,
it allows you to write code without worrying about data transfer,
and it does so without the need of compiler support.
The automatic data transfer feature guarantees no redundant data transfer,
also the overhead of VectorPU is very low and unnoticeable.

The software was tested under gcc 4.9.3, CUDA 7.5 and Ubuntu 16.04.2 LTS

This software is a prototype used in the following paper:

Lu Li, Christoph Kessler. VectorPU: A Generic and Efficient Data-container and
Component Model for Transparent Data Transfer on GPU-based Heterogeneous
Systems. In Proc. 8th Workshop on Parallel Programming and Run-Time Management
Techniques for Many-core Architectures and 6th Workshop on Design Tools and
Architectures for Multicore Embedded Computing Platforms (PARMA-DITAM'17). ACM,
2017.

The documentation of this software is very limited, more details are described
in the above paper.

The software is under LGPLv3 licence.

Contact: Lu Li, explore.leo@gmail.com or lu.li@liu.se