We present a design for a hardware supported global synchronization unit that would be implemented on-chip and directly accessible by all processors in a multi-core architecture. This global synchronization unit will ...
详细信息
ISBN:
(纸本)9780769537139
We present a design for a hardware supported global synchronization unit that would be implemented on-chip and directly accessible by all processors in a multi-core architecture. This global synchronization unit will provide all processors with access to global state information from all other processors in just a few clock ticks, and can be used to perform highly efficient and scalable time synchronization for parallel simulations. Further, our design takes into account the possibility of transient messages, and allows for non-uniform lookahead between processors in conservative synchronization methods. Simulating this hardware in a system simulator, we demonstrate its ability to decrease the runtime of a low-lookahead network simulation by a factor of two over a shared-memory barrier synchronization.
Compared to the traditional SAR imaging algorithm, Back Projection(BP) algorithm is an accurate point-by-point imaging radar algorithm based on time-domain, with simple principle and without any approximation error in...
详细信息
ISBN:
(纸本)9781510822023
Compared to the traditional SAR imaging algorithm, Back Projection(BP) algorithm is an accurate point-by-point imaging radar algorithm based on time-domain, with simple principle and without any approximation error in the imaging process. However, because of intensive computation and low efficiency, it's a new challenge to storage to capacity, throughput and processing ability of DSPs, a single DSP is not enough to meet these demands. So a parallel implementation method of BP algorithm based on TMS320C6678 DSP is proposed in this *** put forward a large point FFT multi-core parallel processing method on 2/4/8 cores what is frequently used in BP algorithm, and a multi-core synchronization method based on distributed memory. Finally using the measured data, we verify the parallel method can greatly enhance the multi-core parallelism, and the real-time performance of BP algorithm has been significantly improved.
暂无评论