Kilo TM: Hardware Transactional Memory for GPU Architectures
Wilson Wai Lun Fung, I. Singh, Andrew Brownsword, Tor M. Aamodt · IEEE Micro · 2012
Programming GPUs is challenging for applications with irregular fine-grained communication between threads. To improve GPUs' programmability and thus extend their usage to a wider range of applications, the authors propose to enable transactional memory (TM) on GPUs via Kilo TM, a novel hardware TM system that scales to thousands of concurrent transactions.