Data transfer minimization for financial derivative pricing using Monte Carlo simulation with GPU in 5G
Citations Over TimeTop 10% of 2015 papers
Abstract
Summary Mobile devices with an enhanced capability are desired to achieve more efficient and energy‐aware wireless communications in the future 5G networks. However, the performance of the expected hardware and networks will face a great challenge in efficiently calculating large data transmissions during the heavy wireless communications with a great number of random data. This paper proposes a novel paradigm for preventing future mobile technology implementations from the restrictions of hardware by using graphics processing units (GPUs). Our proposed mechanism is the first attempt in using GPUs to strengthen the wireless communication capability, which is proved by the experimental research and contributes to mobile technologies in 5G. We apply the Monte Carlo simulation method that is for evaluating the results of algorithms whose output depends on random variables. The experimental results show that Monte Carlo simulation based on GPU can achieve up to 25.9× speed compared with CPU‐based implementation. Copyright © 2015 John Wiley & Sons, Ltd.
Related Papers
- → Experimental results about MPI collective communication operations(1999)10 cited
- → Improving the efficiency of automated protocol implementations using a configurable FDT compiler(2000)5 cited
- → Comparison of MPI Implementations on a Shared Memory Machine(2000)7 cited
- → Experiences with Estelle, LOTOS and SDL: a protocol implementation experiment(1991)7 cited
- → Towards the automatic generation of quality-of-service-preserving implementations from formal specifications(1999)