Pseudorapidity distribution and decorrelation of anisotropic flow within CLVisc hydrodynamics
Studies of fluctuations and correlations of soft hadrons and hard and electromagnetic probes of the dense and strongly interacting medium require event-by-event hydrodynamic simulations of high-energy heavy-ion collisions that are computing intensive. We develop a (3+1)D viscous hydrodynamic model -- CLVisc that is parallelized on Graphics Processing Unit (GPU) using Open Computing Language (OpenCL) with 60 times performance increase for space-time evolution and more than 120 times for the Cooper-Frye particlization relative to that without GPU parallelization. The pseudo-rapidity dependence of anisotropic flow $v_n(η)$ are then computed in CLVisc with initial conditions given by the A Multi-Phase Transport (AMPT) model, with energy density fluctuations both in the transverse plane and along the longitudinal direction. Although the magnitude of $v_n(η)$ and the ratios between $v_2(η)$ and $v_3(η)$ are sensitive to the effective shear viscosity over entropy density ratio $η_v/s$, the shape of the $v_{n}(η)$ distributions in $η$ do not depend on the value of $η_v/s$. The decorrelation of $v_n$ along the pseudo-rapidity direction due to the twist and fluctuation of the event-planes in the initial parton density distributions is also studied. The decorrelation observable $r_n(η^a, η^b)$ between $v_n\{-η^a\}$ and $v_n\{η^a\}$ with the auxiliary reference window $η^b$ is found not sensitive to $η_v/s$ when there is no initial fluid velocity. For small $η_v/s$, the initial fluid velocity from mini-jet partons introduces sizable splitting of $r_n(η^a, η^b)$ between the two reference rapidity windows $η^b \in [3, 4]$ and $η^b \in [4.4, 5.0]$, as has been observed in experiment. The implementation of CLVisc and guidelines on how to efficiently parallelize scientific programs on GPUs are also provided.