Learning Library

← Back to Library

Running Quantum Programs with Qiskit Runtime

Key Points

  • Quantum computers are now accessible via the cloud, but sending circuits to remote hardware and receiving results for each iteration creates significant latency and inefficiency.
  • IBM’s 2021 introduction of **Qiskit Runtime** packages the entire quantum‑classical program in a container that runs close to the hardware, dramatically reducing round‑trip delays and improving scalability.
  • The runtime environment includes built‑in tools such as circuit optimization (gate remapping), result post‑processing (e.g., converting bit strings to expectation values), and error‑mitigation to enhance the quality of noisy quantum outputs.
  • By using the **qiskit‑ibm‑runtime** package, developers can submit both the quantum circuit and its classical processing logic as a single job, eliminating the need for repeated data transfers.
  • This cloud‑based execution model streamlines real‑world quantum applications, allowing hundreds of iterative steps to be performed efficiently without the overhead of constant communication with distant quantum hardware.

Full Transcript

# Running Quantum Programs with Qiskit Runtime **Source:** [https://www.youtube.com/watch?v=yh789q5qur0](https://www.youtube.com/watch?v=yh789q5qur0) **Duration:** 00:04:24 ## Summary - Quantum computers are now accessible via the cloud, but sending circuits to remote hardware and receiving results for each iteration creates significant latency and inefficiency. - IBM’s 2021 introduction of **Qiskit Runtime** packages the entire quantum‑classical program in a container that runs close to the hardware, dramatically reducing round‑trip delays and improving scalability. - The runtime environment includes built‑in tools such as circuit optimization (gate remapping), result post‑processing (e.g., converting bit strings to expectation values), and error‑mitigation to enhance the quality of noisy quantum outputs. - By using the **qiskit‑ibm‑runtime** package, developers can submit both the quantum circuit and its classical processing logic as a single job, eliminating the need for repeated data transfers. - This cloud‑based execution model streamlines real‑world quantum applications, allowing hundreds of iterative steps to be performed efficiently without the overhead of constant communication with distant quantum hardware. ## Sections - [00:00:00](https://www.youtube.com/watch?v=yh789q5qur0&t=0s) **Reducing Quantum Cloud Latency with Qiskit Runtime** - The speaker describes the inefficiency of repeatedly sending circuits to remote quantum hardware for each hybrid iteration and introduces IBM’s 2021 Qiskit Runtime as a way to keep classical processing near the quantum device, dramatically cutting data‑transfer overhead. ## Full Transcript
0:00in my last video i show you how to write 0:03a quantum circuit using kiskit 0:10but i'm sure that you are wondering how 0:12to actually take la quantum program 0:15and rang it on a real quantum 0:18computer 0:21quantum computers have existed since the 0:231990s but until recently they are only 0:26accessible 0:28to researchers in labs 0:30today quantum computers still require 0:33special equipment and maintenance 0:35but thanks to the development of cloud 0:37technology 0:38anyone can access our quantum computers 0:41through the cloud 0:45today most of the real world quantum 0:47applications 0:49uses some kind of interplay between the 0:51classical and the quantum resources 0:54that is sending my circuits to the real 0:57quantum hardware for execution is only 0:59half of the story 1:01once the program is the circuits are 1:03done being wrong 1:05the quantum hardware returns the results 1:08back to my laptop 1:10my quantum program then does some kind 1:12of classical processing 1:15and the output of this classical 1:16processing is then used to generate the 1:19next sets of circuits to run 1:22unfortunately this is a very inefficient 1:25model 1:26because my real world application can 1:28have hundreds of iterations 1:31and for every single iteration the 1:33circuit 1:34the circuits need to be sent all the way 1:36to the quantum hardware which can be 1:39extremely far away 1:41ellen needs to wait for the result to 1:42again make the return trip just to 1:44generate the circuits for the next 1:46iteration 1:47it is spending more time on the data 1:50transfer than actual execution 1:53and this is why in 2021 ibm introduced 1:57kisky runtime 2:03kisking runtime takes quantum programs 2:06and all its dependencies 2:08and package it in a containerized 2:10execution environment 2:19this environment is close to the quantum 2:21hardware so it can reduce the latency by 2:24creating a tight loop 2:26and because the program runs on the 2:28cloud 2:29it also improves 2:31the scalability of classical resources 2:35but there's more than that 2:37your programs now run in the cloud and 2:40can get all the benefits from 2:43of a cloud-based model 2:46most notably ksk runtime offers a set of 2:49tools and features 2:51that would 2:52make quantum computing easier for 2:54developers 2:57it has a number of predefined routines 3:00such as 3:02circuit optimization lot remaps the 3:05gates in your circuits so they would run 3:07faster 3:09similar to what modern compilers do 3:16circuit result post-processing law will 3:19take all those measured ones and zeros 3:21and 3:23convert them into higher level 3:25representation such as expectation 3:27values 3:29and 3:30the last one is error mitigation 3:36law gives you higher quality results 3:39the error mitigation is especially 3:41important because quantum hardware today 3:43is not perfect 3:45and the raw results often contain errors 3:48so without error mitigation you wouldn't 3:50be able to get the correct results back 3:53so how do you actually access cascade 3:55runtime 3:56the answer is simple you use kiskit 3:59kit has a package that allows you to 4:02send your entire quantum program 4:04including both the quantum and classical 4:06piece into this execution environment 4:10thanks for watching 4:11if you have any questions please leave 4:13them in the comments below also please 4:15remember to like this video and 4:17subscribe to our channel so we can 4:19continue to bring you content that 4:20matters to you