Generic Library for Asynchronous Data Operations and Streaming - GLADOS
Generic Library for Asynchronous Data Operations and Streaming - GLADOS
Generic Library for Asynchronous Data Operations and Streaming (GLADOS) provides a framework for data stream processing in a pipeline scheme. GLADOS provides a managed memory pool for device memory and host memory which allows high-throughput processing of streams without runtime memory allocations. GLADOS supports the fork-join paradigm to allow for parallel processing branches and the subject-observer pattern for asychnronous control of the process. Implementations of the processing steps use C++ templates and can therefore be adapted very flexibly.
Keywords: Pipeline; Data processing; C++; CUDA
Related publications
-
Real-time Data Processing for Ultrafast X-Ray Computed Tomography using …
ROBIS: 34743 is supplemented by this (Id 34605) publication
-
Software in external data repository
Publication year 2022
Programming language: C++, CUDA
System requirements: Linux-based operating system, CUDA-capable GPU
License: Apache-2.0 (Link to license text)
Hosted on https://gitlab.hzdr.de/risa/GLADOS: Link to location
Permalink: https://www.hzdr.de/publications/Publ-34605