Fig. 3: Classical simulation cost and extrapolated running time of sampling from m-cycle Sycamore random circuits with low XEB fidelities. | Nature Computational Science

Fig. 3: Classical simulation cost and extrapolated running time of sampling from m-cycle Sycamore random circuits with low XEB fidelities.

From: Efficient parallelization of tensor network contraction for simulating quantum computation

Fig. 3: Classical simulation cost and extrapolated running time of sampling from m-cycle Sycamore random circuits with low XEB fidelities.The alternative text for this image may have been generated using AI.

The dashed lines represent the theoretical number of FLOPs and the solid lines represent extrapolated running times from the experiments on an Nvidia V100 graphics card. The two axes are aligned by the theoretical GPU efficiency of a Nvidia V100. Consequently, the dashed lines represent runtime lower bounds provided that GPU efficiency is fully saturated. The velvet line is reported in Arute and colleagues2 using the hybrid Schrödinger–Feynman algorithm.

Back to article page