Abstract: One example method includes evaluating a function invoked by a request that is received at a local classical computing execution environment, and the request also implies performance of a quantum computing function in a quantum computing execution environment, based on an outcome of the evaluating, determining whether or not the function should be run in the local classical computing execution environment, or whether the function should be run in a separate classical computing execution environment, and when the determining indicates that the function should be run in the separate classical computing execution environment, forwarding the request to the separate classical computing environment for execution of the function. The local classical computing execution environment, the separate classical computing execution environment, and the quantum computing execution environment, are respective first, second, and third, tiers of a hybrid computing execution environment.

Abstract: Techniques for verifying a software package version are disclosed. A new package version for a software package is obtained. A repository maintains package versions for the software package. The new package version includes a new dependency and a new installation instruction for the software package. The repository lists one or more other package versions, each of which includes one or more other dependencies, for the software package. The new dependency and the new installation instruction are verified. In response, the new package version, which includes the new dependency and the new installation instruction, is permitted to be added to the repository, resulting in the software package being associated with the new package version.

Abstract: Cutting quantum circuits is disclosed. Probability distributions for quantum circuit parameters are generated from historical quantum circuit data. When cutting the quantum circuit, the probability distributions are sampled to obtain a set of initial circuit parameters that can function as constraints in the cutting operation.

Abstract: Dynamic orchestration of quantum circuit execution. A runtime prediction is performed on a quantum circuit to determine or estimate an amount of resources and execution time that are needed to execute the quantum circuit. The runtime prediction may also estimate the resources, time, and other factors associated with cutting the quantum circuit. When cutting the circuit is beneficial, the quantum subcircuits generated by cutting the quantum circuit are subject to runtime prediction. This information generated by the runtime prediction, along with telemetry data from computing and quantum resources and service level objectives, is used to generate an execution plan for performing or executing the quantum circuit or the quantum subcircuits when the quantum circuit is cut.

Abstract: One example method includes obtaining, for a quantum gate of a quantum circuit, an estimate of a distribution of values expected to result from a matrix multiplication relating to the quantum gate, applying a binning process to the distribution to generate a bin table, quantizing an FP value of the matrix multiplication by mapping the bin table to a bin index that corresponds to the FP value, storing the bin index, and performing the matrix multiplication using the bin index.

Abstract: Techniques for achieving improved checkpointing are disclosed. A library that has been implemented in a process is accessed. This library facilitates checkpoint-and-resume functionality to enable the process to checkpoint itself. Tags are used to annotate code of the process. The tags define which data of the process is to be saved in an event in which the process terminates in an unexpected manner. The tags define a block within the code and define the data that is to be saved. After the data has been saved, the process terminates unexpectedly. The process restarts from its beginning state and is progressed through its code. This progression includes skipping code for which data was previously saved. The process continues to progress and to skip through the code until reaching the defined block. At that point, the process resumes at the defined block, such that the process resumes at a user-defined location.

Abstract: One example method includes obtaining information about a first pre-defined implementation of a QUBO (quadratic unconstrained binary optimization) problem configured for execution on a gate-based device, obtaining information about a second pre-defined implementation of the QUBO problem configured for execution on an annealing device, receiving information about a QUBO job that is to be executed, identifying first hardware and second hardware that are available to execute the QUBO job, and the first hardware is different from the second hardware, using the information about the first and second pre-defined implementations of the QUBO problem to generate respective predictions concerning performance of the QUBO job on the first hardware and the second hardware, comparing the predictions, and based on the comparing, selecting one of the first hardware and the second hardware for execution of the QUBO job.

Abstract: Replacing a sequence of gates with a replacement matrix. A quantum circuit may include a series of gates that includes a sequence of gates that is repeated within the series or in the quantum circuit. The sequence of gates is computed to generate a replacement matrix. The sequence of gates is then replaced with the replacement matrix. The replacement matrix improves performance by reducing the number of operations that would otherwise be required. The replacement matrix can be cached and allows the circuit to be executed locally, remotely, or both. The cached replacement matrix can be used in other quantum circuits that include the same sequence of gates represented by the replacement matrix.

Abstract: One example method includes simulating execution of a quantum circuit on a classical computing infrastructure, after one or more times that a gate of the quantum circuit is executed as part of the simulating, creating, after execution of that gate, a hash of a state vector that captures a state of the execution of the quantum circuit, storing the hash, and respective associated data structure, in storage, then as part of a simulated execution process, calculating a hash of each gate across the new quantum circuit, looking up, in the storage, a hash of a state vector associated with execution of one of the gates of the new quantum circuit, and restoring, from storage, the latest hash of the state vector associated with the one gate of the new quantum circuit.

Abstract: A method includes predicting, using a machine learning model, runtime characteristics concerning a quantum computing function, predicting, using the machine learning model, resources needed to perform the quantum computing function, selecting an execution environment for the quantum computing function, and executing the quantum computing function in the execution environment. The quantum computing function may be a quantum circuit cutting operation, or the quantum computing function may be a quantum circuit execution.

Abstract: One example method includes defining a quantum circuit, orchestrating the quantum circuit to a computing infrastructure for execution, executing the quantum circuit on the infrastructure and, while the quantum circuit is being executed, checkpointing the quantum circuit. One or more of the defining, executing, and checkpointing, includes using a mechanism that improves and/or enhances performance of the checkpointing. Example mechanisms include including a custom gate in the quantum circuit, using quantization when storing state data concerning the quantum circuit, using persistent memory and orchestration for the checkpointing, dynamically determining when checkpointing will be performed, and performing automated flattening of quantum circuit checkpoint images.

Abstract: One example method includes receiving parameter values relating to execution of a simulation of a quantum algorithm, deriving quantum attributes from the parameter values, generating, based on the quantum attributes, a classical computing resource prediction, and translating the classical computing resource prediction into elements of a classical computing infrastructure. The classical computing infrastructure may be sized and configured to support computationally efficient, and cost efficient, execution of the simulation of the quantum algorithm.

Abstract: One example method includes evaluating a function invoked by a request that is received at a local classical computing execution environment, and the request also implies performance of a quantum computing function in a quantum computing execution environment, based on an outcome of the evaluating, determining whether or not the function should be run in the local classical computing execution environment, or whether the function should be run in a separate classical computing execution environment, and when the determining indicates that the function should be run in the separate classical computing execution environment, forwarding the request to the separate classical computing environment for execution of the function. The local classical computing execution environment, the separate classical computing execution environment, and the quantum computing execution environment, are respective first, second, and third, tiers of a hybrid computing execution environment.

Abstract: One example method includes identifying an accelerator service instance associated with a workload, calling a service broker associated with the accelerator service instance to obtain information needed to use the accelerator service instance, receiving an accelerator call, and accelerator job information concerning an accelerator job, from the workload, in response to the accelerator call, spinning up a new process dedicated to the accelerator job, as part of the new process, running the accelerator job using either the accelerator service instance, or a locally available accelerator, and returning data, generated by running the accelerator job, to the workload.

Abstract: One example method includes receiving job configuration information from a user with a quantum computing job to be performed, receiving quantum computing information from a quantum computing service vendor, generating, based on the quantum computing information, a vendor score for the quantum computing service vendor, and transmitting the vendor score to the user. The quantum computing information received from the quantum computing service vendor may include information about an accuracy of results produced by execution of a quantum circuit or other quantum hardware operated by the quantum computing service vendor.