Abstract
The article presents a comparative analysis of the implementation of parallel algorithms on the central processors of automation systems in agriculture. Modern automation systems impose increased requirements on the reliability of the implementation of parallel algorithms in real time. It is proposed to use models for the development, analysis and comparison of parallel algorithms on GPUs. The proposed model of parallel computing on GPUs is designed to simplify the development of parallel algorithms for a heterogeneous CPU-GPU environment. Those, with this model, you can: develop parallel algorithms that use data parallelism, while applying the existing experience in creating parallel algorithms for the PRAM machine; evaluate the running time of the parallel algorithm and analyze which part of it is the most resource-intensive and requires optimization; compare parallel algorithms according to some parameters of the model, and, if required, select the best one according to these parameters in real time for automation in agriculture.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
