Enabling Multi-programming Mechanism for Quantum Computing in the NISQ Era

Abstract

NISQ devices have several physical limitations and unavoidable noisy quantum operations, and only small circuits can be executed on a quantum machine to get reliable results. This leads to the quantum hardware under-utilization issue. Here, we address this problem and improve the quantum hardware throughput by proposing a Quantum Multi-programming Compiler (QuMC) to execute multiple quantum circuits on quantum hardware simultaneously. This approach can also reduce the total runtime of circuits. We first introduce a parallelism manager to select an appropriate number of circuits to be executed at the same time. Second, we present two different qubit partitioning algorithms to allocate reliable partitions to multiple circuits – a greedy and a heuristic. Third, we use the Simultaneous Randomized Benchmarking protocol to characterize the crosstalk properties and consider them in the qubit partition process to avoid the crosstalk effect during simultaneous executions. Finally, we enhance the mapping transition algorithm to make circuits executable on hardware using a decreased number of inserted gates. We demonstrate the performance of our QuMC approach by executing circuits of different sizes on IBM quantum hardware simultaneously. We also investigate this method on VQE algorithm to reduce its overhead.

Accelerating the variational quantum eigensolver using parallelism

“…The speedup here will depend on how many groupings of Pauli terms there are, but it could also be combined with any of the methods above. This is what was done in [4,6], combined with using different parameters. However, the available speedup for the Hubbard model from parallelising measurements alone would be relatively limited, as the energy can be measured using five measurements for an arbitrary lattice [12].…”

“…This behaviour seems to be due to the fact that we are adding pairs with worse fidelities rather than other factors such as crosstalk-see figure 9 in the appendix for a more detailed analysis. For larger and longer circuits, crosstalk will likely play a more dominant role [23], and the selection of the groups of qubits will need to take this into account [4].…”

“…A number of works have investigated the theory and implementation of parallelism (also called multi-programming) on NISQ devices. Algorithms have been proposed that partition chips into regions of reliable qubits, allowing for circuits of different depths and size to be run simultaneously [3][4][5][6]. On cloud systems, this could potentially allow multiple users to execute circuits on the same chip at the same time, increasing the throughput of quantum devices.…”

“…On cloud systems, this could potentially allow multiple users to execute circuits on the same chip at the same time, increasing the throughput of quantum devices. Demonstrations of the VQE algorithm in parallel have also been performed for two-qubit deuteron and molecular hydrogen, in particular executing circuits to measure all of the Pauli terms in a Hamiltonian simultaneously [4,6].…”

“…Indeed, we find below that the choice of which algorithm to use for VQE has crucial implications for the level of speedup via parallelism that can be achieved. Also, crosstalk and the use of lower-fidelity qubits might reduce the accuracy of parallel algorithms [3,4].…”