Announcing Amazon Braket Hybrid Jobs for running hybrid quantum-classical workloads on Amazon Braket
Amazon Braket Hybrid Jobs enables you to easily run hybrid quantum-classical algorithms such as the Variational Quantum Eigensolver (VQE) and the Quantum Approximate Optimization Algorithm (QAOA), that combine classical compute resources with quantum computing devices to optimize the performance of today’s quantum systems. With this new feature, you only have to provide your algorithm script and choose a target device — a quantum processing unit (QPU) or quantum circuit simulator. Amazon Braket Hybrid Jobs is designed to spin up the requested classical resources when your target quantum device is available, run your algorithm, and release the instances after completion so you only pay for what you use. Braket Hybrid Jobs can provide live insights into algorithm metrics to monitor your algorithm as it progresses, enabling you to make adjustments more quickly. Most importantly, your jobs have priority access to the selected QPU for the duration of your experiment, putting you in control, and helping to provide faster and more predictable execution.
Amazon Braket Hybrid Jobs enables you to easily run hybrid quantum-classical algorithms such as the Variational Quantum Eigensolver (VQE) and the Quantum Approximate Optimization Algorithm (QAOA), that combine classical compute resources with quantum computing devices to optimize the performance of today’s quantum systems. With this new feature, you only have to provide your algorithm script and choose a target device — a quantum processing unit (QPU) or quantum circuit simulator. Amazon Braket Hybrid Jobs is designed to spin up the requested classical resources when your target quantum device is available, run your algorithm, and release the instances after completion so you only pay for what you use. Braket Hybrid Jobs can provide live insights into algorithm metrics to monitor your algorithm as it progresses, enabling you to make adjustments more quickly. Most importantly, your jobs have priority access to the selected QPU for the duration of your experiment, putting you in control, and helping to provide faster and more predictable execution.
To run a job with Braket Hybrid Jobs, you need to first define your algorithm using either the Amazon Braket SDK or PennyLane. You can also use TensorFlow and PyTorch or create a custom Docker container image. Next, you create a job via the Amazon Braket API or console, where you provide your algorithm script (or custom container), select your target quantum device, and choose from a variety of optional settings including the choice of classical resources, hyper-parameter values, and data locations. If your target device is a simulator, Braket Hybrid Jobs is designed to start executing right away. If your target device is a QPU, your job will run when the device is available and your job is first in the queue. You can define custom metrics as part of your algorithm, which can be automatically reported to Amazon CloudWatch and displayed in real time in the Amazon Braket console. Upon completion, Braket Hybrid Jobs writes your results to Amazon S3 and releases your resources.
The Hybrid Jobs feature is available in all regions where Amazon Braket is available and can be used with all devices on Amazon Braket. To learn more, you can read the AWS News Blog post and the documentation. To get started you can follow one of the example notebooks on Github that also come preinstalled in Amazon Braket notebooks.