Abstract
Given the need to achieve maximum performance possible, offloading intensive computation workload to GPU is a key to achieve this goal. Offloading most of the workload to GPU may not results in desired performance, so a middle approach is more suitable such as splitting the workload between the CPU and the GPU can be considered as an optimized approach. In this study, we used a popular high performance computation workload which can also be implemented using a hybrid approach in which part of the workload is offloaded to the CPU. We also present a performance estimation method which is verified to estimate performance with in 5% error margin.
Highlights
The CPU was the focus when it comes to application with intensive computation requirements
We present a hybrid implementation for Monte Carlo (MC) workload to maximize performance
Hybrid performance depends on many factors, such as problem size, load balancing, hiding memory operations using parallelism, algorithm optimization and capabilities of devices which are best when the devices are well balanced
Summary
The CPU was the focus when it comes to application with intensive computation requirements. While most of the high performance computation algorithm focuses on GPU for data intensive computation, there is still a small window for the CPU to perform parallel task with the GPU which leads to hybrid implementation. The CPU is designed for different applications and can provide fast response times to a single task. Architectural features such as branch prediction, out-oforder execution and super-scalar are directly related to this performance improvement. CPUs on the other hand are designed to pack small number of processing cores while keeping within a given power and thermal limitations. This results in GPUs trading off single threads performance for increase parallel processing.
Sign-in/Register to view highlights & summary 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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
