In [Van Beeumen, et. al, HPC Asia 2020, https://www.doi.org/10.1145/3368474.3368497] a scalable and matrix-free eigensolver was proposed for studying the many-body localization (MBL) transition of two-level quantum spin chain models with nearest-neighbor $XX+YY$ interactions plus $Z$ terms. This type of problem is computationally challenging because the vector space dimension grows exponentially with the physical system size, and averaging over different configurations of the random disorder is needed to obtain relevant statistical behavior. For each eigenvalue problem, eigenvalues from different regions of the spectrum and their corresponding eigenvectors need to be computed. Traditionally, the interior eigenstates for a single eigenvalue problem are computed via the shift-and-invert Lanczos algorithm. Due to the extremely high memory footprint of the LU factorizations, this technique is not well suited for large number of spins $L$, e.g., one needs thousands of compute nodes on modern high performance computing infrastructures to go beyond $L = 24$. The matrix-free approach does not suffer from this memory bottleneck, however, its scalability is limited by a computation and communication imbalance. We present a few strategies to reduce this imbalance and to significantly enhance the scalability of the matrix-free eigensolver. To optimize the communication performance, we leverage the consistent space runtime, CSPACER, and show its efficiency in accelerating the MBL irregular communication patterns at scale compared to optimized MPI non-blocking two-sided and one-sided RMA implementation variants. The efficiency and effectiveness of the proposed algorithm is demonstrated by computing eigenstates on a massively parallel many-core high performance computer.

Many-body chaos has emerged as a powerful framework for understanding thermalization in strongly interacting quantum systems. While recent analytic advances have sharpened our intuition for many-body chaos in certain large N theories, it has proven challenging to develop precise numerical tools capable of exploring this phenomenon in generic Hamiltonians. To this end, we utilize massively parallel, matrix-free Krylov subspace methods to calculate dynamical correlators in the Sachdev-Ye-Kitaev model for up to N=60 Majorana fermions. We begin by showing that numerical results for two-point correlation functions agree at high temperatures with dynamical mean field solutions, while at low temperatures finite-size corrections are quantitatively reproduced by the exactly solvable dynamics of near extremal black holes. Motivated by these results, we develop a novel finite-size rescaling procedure for analyzing the growth of out-of-time-order correlators. Our procedure accurately determines the Lyapunov exponent, λ, across a wide range in temperatures, including in the regime where λ approaches the universal bound, λ=2π/β.

The ability to perform measurements in the middle of a quantum circuit is a powerful resource. It underlies a wide range of applications, from remote state preparation to quantum error correction. Here we apply mid-circuit measurements for a particular task: demonstrating quantum computational advantage. The goal of such a demonstration is for a quantum device to perform a computational task that is infeasible for a classical device with comparable resources. In contrast to existing demonstrations, the distinguishing feature of our approach is that the classical verification process is efficient, both in asymptotic complexity and in practice. Furthermore, the classical hardness of performing the task is based upon well-established cryptographic assumptions. Protocols with these features are known as cryptographic proofs of quantumness. Using a trapped-ion quantum computer, we perform mid-circuit measurements by spatially isolating portions of the ion chain via shuttling. This enables us to implement two interactive cryptographic proofs of quantumness, which when suitably scaled to larger systems, promise the efficient verification of quantum computational advantage. Our methods can be applied to a range of interactive quantum protocols.