Abstract:Recently Graphics Processing Units (GPUs) have been used to speed up very CPU-intensive gravitational microlensing simulations. In this work, we use the Xeon Phi coprocessor to accelerate such simulations and compare its performance on a microlensing code with that of NVIDIA's GPUs. For the selected set of parameters evaluated in our experiment, we find that the speedup by Intel's Knights Corner coprocessor is comparable to that by NVIDIA's Fermi family of GPUs with compute capability 2.0, but less significant… Show more
“…Great effort has been expended to deal with this problem in two ways: (i) to perform the computation in parallel and (ii) to reduce the number of lenses (N * ) or light rays (N rays ) involved in the calculations. For the first one, different parallel strategies based on both CPUs (Chen et al 2017) and graphics processing units (GPUs) have been applied to the IRS method, between which GPUs have shown great computational potential (Thompson et al 2010). For the second, Wambsganss (1999) introduced a hierarchical tree method to reorder the lenses by their distance to individual light rays and group distant lenses into larger cells, allowing the N * directly involved in the calculations to be reduced.…”
We present an improved inverse-ray-shooting code based on graphics processing units (GPUs) to generate microlensing magnification maps. In addition to introducing GPUs to accelerate the calculations, we also invest effort into two aspects: (i) A standard circular lens plane is replaced by a rectangular one to reduce the number of unnecessary lenses as a result of an extremely prolate rectangular image plane. (ii) An interpolation method is applied in our implementation, achieving significant acceleration when dealing with the large number of lenses and light rays required by high-resolution maps. With these applications, we have greatly reduced the running time while maintaining high accuracy: The speed was increased by about 100 times compared with an ordinary GPU-based inverse-ray-shooting code and a GPU-D code when handling a large number of lenses. If a high-resolution situation with up to 10,0002 pixels, resulting in almost 1011 light rays, is encountered, the running time can also be reduced by two orders of magnitude.
“…Great effort has been expended to deal with this problem in two ways: (i) to perform the computation in parallel and (ii) to reduce the number of lenses (N * ) or light rays (N rays ) involved in the calculations. For the first one, different parallel strategies based on both CPUs (Chen et al 2017) and graphics processing units (GPUs) have been applied to the IRS method, between which GPUs have shown great computational potential (Thompson et al 2010). For the second, Wambsganss (1999) introduced a hierarchical tree method to reorder the lenses by their distance to individual light rays and group distant lenses into larger cells, allowing the N * directly involved in the calculations to be reduced.…”
We present an improved inverse-ray-shooting code based on graphics processing units (GPUs) to generate microlensing magnification maps. In addition to introducing GPUs to accelerate the calculations, we also invest effort into two aspects: (i) A standard circular lens plane is replaced by a rectangular one to reduce the number of unnecessary lenses as a result of an extremely prolate rectangular image plane. (ii) An interpolation method is applied in our implementation, achieving significant acceleration when dealing with the large number of lenses and light rays required by high-resolution maps. With these applications, we have greatly reduced the running time while maintaining high accuracy: The speed was increased by about 100 times compared with an ordinary GPU-based inverse-ray-shooting code and a GPU-D code when handling a large number of lenses. If a high-resolution situation with up to 10,0002 pixels, resulting in almost 1011 light rays, is encountered, the running time can also be reduced by two orders of magnitude.
“…(ii) To reduce the number of lenses (N * ) or light rays (N rays ) involved in the calculations. For the first direction, different parallel strategies based on both CPUs Chen et al (2017) and GPUs have been applied on IRS method, among which GPUs have shown great computational potential Thompson et al (2010). For the second direction, Wambsganss (1999) introduced hierarchical tree method to reorganize the lenses by their distance to individual light and group distant lenses into larger cells, allowing N * directly involved in calculations to be reduced.…”
We present an improved inverse ray-shooting code based on GPUs for generating microlensing magnification maps. In addition to introducing GPUs for acceleration, we put the efforts in two aspects: (i) A standard circular lens plane is replaced by a rectangular one to reduce the number of unnecessary lenses as a result of an extremely prolate rectangular image plane. (ii) Interpolation method is applied in our implementation which has achieved an significant acceleration when dealing with large number of lenses and light rays required by high resolution maps. With these applications, we have greatly reduced the running time while maintaining high accuracy: the speed has been increased by about 100 times compared with ordinary GPU based IRS code and GPU-D code when handling large number of lenses. If encountered the high resolution situation up to 10000 2 pixels, resulting in almost 10 11 light rays, the running time can also be reduced by two orders of magnitude.
Summary
Intel's Xeon Phi combines the parallel processing power of a many‐core accelerator with the programming ease of CPUs. In this paper, we present a survey of works that study the architecture of Phi and use it as an accelerator for a broad range of applications. We review performance optimization strategies as well as the factors that bottleneck the performance of Phi. We also review works that perform comparison or collaborative execution of Phi with CPUs and GPUs. This paper will be useful for researchers and developers in the area of computer‐architecture and high‐performance computing.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.