FusionSim: Characterizing the Performance Benefits of Fused CPU/GPU Systems.

摘要

We present FusionSim, a modeling framework capable of cycle-accurate simulation of a complete x86-based computer system with (a) a CPU and a GPU on the same die, and (b) a CPU and a GPU connected as separate components.;We use FusionSim to characterize the performance of the Rodinia benchmarks on fused and discrete systems. We demonstrate that the speed-up due to fusion is highly correlated with the input data size. We demonstrate that for benchmarks that benefit most from fusion, a 9.72x speed-up is possible for small problem sizes. This speedup reduces to 1.84x with medium problem sizes. We study a software-managed coherence solution for the fused system. We find that it imposes a minor performance overhead of 2% for most benchmarks. Finally, we develop an analytical model for the performance benefit that is to be expected from fusion and show that FusionSim follows the predicted performance trend.