Recently, I have been learning AWS Braket, so this series of tutorial posts will record the things or problems that I face.
What is Amazon Braket?
Amazon Braket is a service that gives you access to multiple different types of quantum computers or simulators.
Braket Available Quantum Computers and Simulators
AWS Braket offers several types of QPUs and Simulators that users can utilize. Below, each one’s pros, cons, and suitability are listed.
IonQ
IonQ uses trapped-ion technology, where qubits are represented by the energy states of individual ions (charged atoms) trapped in an electromagnetic field.
| Device Name | Qubit Numbers | Paradigm | Device ARN | Region |
|---|---|---|---|---|
| Aria 1 | 25 | gate based w/ error mitigation support | arn:aws:braket:us-east-1::device/qpu/ionq/Aria-1 | us-east-1 |
| Aria 2 | 25 | gate based w/ error mitigation support | arn:aws:braket:us-east-1::device/qpu/ionq/Aria-2 | us-east-1 |
| Forte 1 | 36 | gate based w/ error mitigation support | arn:aws:braket:us-east-1::device/qpu/ionq/Forte-1 | us-east-1 |
| Harmony | 11 | gate based | arn:aws:braket:us-east-1::device/qpu/ionq/Harmony | us-east-1 |
- Pros
- Trapped-ion technology with high gate fidelities and low error rates.
- Suitable for running quantum algorithms with high precision and accuracy.
- Cons
- Limited qubit count (currently up to 32 qubits).
- Relatively slower gate operations compared to some other architectures.
- Suitable for
- Simulations
- Quantum chemistry
- Algorithms that require high precision but don’t demand a large number of qubits
Rigetti
Rigetti is based on superconducting qubit technology, where qubits are created using superconducting circuits cooled to near absolute zero temperatures.
| Device Name | Qubit Numbers | Paradigm | Device ARN | Region |
|---|---|---|---|---|
| Aspen M-3 | 79 | gate based | arn:aws:braket:us-west-1::device/qpu/rigetti/Aspen-M-3 | us-west-1 |
- Pros
- Superconducting qubit architecture with relatively high qubit counts (up to 80 qubits).
- Fast gate operations
- High connectivity between qubits.
- Cons
- Lower gate fidelities compared to some other architectures.
- Susceptible to environmental noise and decoherence.
- Suitable for
- Algorithms that require a larger number of qubits and can tolerate some noise, such as quantum machine learning and optimization problems.
QuEra
QuEra is based on neutral atom technology, which uses arrays of individually controlled neutral atoms (typically rubidium or cesium) as qubits. It utilizes field-programmable qubit array (FPQA) hardware, which allows qubit configurations to be rearranged on demand without the need to change the hardware, offering flexibility and controllability.
| Device Name | Qubit Numbers | Paradigm | Device ARN | Region |
|---|---|---|---|---|
| Aquila | 256 | Analog Hamiltonian Simulation | arn:aws:braket:us-east-1::device/qpu/quera/Aquila | us-east-1 |
- Pros
- High qubit counts
- Long coherence times: maintaining their quantum states for relatively long periods, improving computation accuracy.
- Lower operational errors
- Utilization of field-programmable qubit array (FPQA) hardware
- Potential for high connectivity: The neutral atom architecture allows for high connectivity between qubits.
- Cons
- Limited information on gate fidelities and performance
- Suitable for
- Running quantum algorithms that require high qubit counts, long coherence times, and high connectivity between qubits.
Oxford Quantum Circuits
Oxford Quantum Circuits uses superconducting qubit technology, similar to Rigetti.
| Device Name | Qubit Numbers | Paradigm | Device ARN | Region |
|---|---|---|---|---|
| Lucy | 8 | gate based | arn:aws:braket:eu-west-2::device/qpu/oqc/Lucy | us-west-2 |
- Pros
- Superconducting qubit architecture with high connectivity and gate fidelities.
- Suitable for running quantum algorithms with high precision.
- Cons
- Currently limited qubit count (up to 8 qubits).
- Relatively new player in the quantum computing market.
- Suitable for
- Testing and developing quantum algorithms that require high precision but don’t demand a large number of qubits.
Amazon Web Services
| Device Name | Qubit Numbers | Paradigm | Device ARN | Region |
|---|---|---|---|---|
| SV1 | 34 | state vector | arn:aws:braket:::device/quantum-simulator/amazon/sv1 | All Regions |
| TN1 | 50 | tensor network | arn:aws:braket:::device/quantum-simulator/amazon/tn1 | us-west-2, us-east-1, and eu-west-2 |
| DM1 | 17 | density matrix | arn:aws:braket:::device/quantum-simulator/amazon/dm1 | All Regions |
| braket_sv | 25 | state vector | N/A | Local |
| braket_dm | 12 | tensor network | N/A | Local |
Conclusion
So far, we have introduced the QPUs and simulators that Braket offers, along with their pros, cons, and suitability.
In the next chapter, we will show how to use the Braket service.