North Carolina State University
Emerging accelerating architectures, such as GPUs, have proved successful in providing significant performance gains to various application domains. However, their viability to operate on general streaming data is still ambiguous. In this paper, the authors propose GStream, a general-purpose, scalable data streaming framework on GPUs. The contributions of GStream are as follows: They provide powerful, yet concise language abstractions suitable to describe conventional algorithms as streaming problems. They project these abstractions onto GPUs to fully exploit their inherent massive data parallelism. They demonstrate the viability of streaming on accelerators.