QPU Submission Checklist
Things to do before submitting to IonQ’s hardware systems.
Once you’ve gotten access to IonQ’s systems and set up a quantum circuit using the IonQ API, Qiskit, or any other method, you can submit a job to one of our QPUs. However, before you start running on quantum hardware, we have a few recommendations to ensure that your circuits run successfully.
Run your job on the cloud simulator
The ideal simulator is a great starting point for making sure that your code runs successfully, you can submit jobs to the IonQ Quantum Cloud, and your circuits compile without errors.
Run your job on the cloud simulator with noise model
Once you’ve run a job on the ideal simulator, try running with the noisy simulator, using the noise model based on your target QPU. While our noise models don’t capture all sources of noise for the QPUs, they can give you a general idea of how noise will affect your results. This can also help you determine how many shots to use.
Check that you have access to the QPU
On the “Backends” tab of the IonQ Cloud Console, you can see our current QPUs. If a QPU is marked with “Out of Plan”, you won’t be able to submit to it. You can request access with this form or ask your IonQ point of contact.
Check the QPU status
The “Backends” tab of the IonQ Cloud Console and the IonQ Status page provide information about system availability and maintenance. If you submit a job to a system that is not currently available, it will still go into the queue and will show as “ready” on the “My Jobs” tab, but jobs in the queue won’t run until the system is available again.
Check the queue status
The “Backends” tab of the IonQ Cloud Console shows the average time in queue for jobs that are currently waiting, and the IonQ Status page shows this average over time. This information doesn’t tell you exactly when your job will run, but it can give you a general idea of how crowded the systems are when you submit.
Check your project budget
On the “Projects” tab of the IonQ Cloud Console, you can see your spend so far and the total budget for each of your projects. If your spend is approaching your total available budget, you may not be able to run a job. Your organization owner can adjust project budgets.
Double-check your API key and project
Each API key is tied to a specific project. If you have multiple projects, we suggest verifying that your code is using the API key for the right project. You can view the jobs submitted to a project by clicking on it from the “Projects” tab. This list includes simulator jobs, so if you’ve tested your code on the simulator and those jobs are showing up in the intended project, switching to a QPU backend with the same API key will keep those jobs in the same project.
Make a reservation
If you’re running a hybrid or variational job, you may want to make a reservation to ensure that multiple circuits submitted (i.e., multiple iterations of your variational optimization) will run consecutively rather than via the queue. You can request a reservation by emailing [email protected] with information about the target system, reservation length, and scheduling preferences.
Once you’ve confirmed that you’re ready to run on the QPU, you should be good to go! Please reach out to [email protected] or submit a ticket on support.ionq.com if you run into any problems or you have any additional questions.
Was this page helpful?