With the rapid advancement of quantum technology, researchers and developers need controllers capable of performing complex classical calculations in real-time and employing flexible control flow. They require any-to-any qubit data sharing encompassing measurements and calculated results (not solely limited to Boolean decisions) with ultra-fast feedback for large-scale QPUs. This is the rationale behind Quantum Machines’ Pulse Processing Unit (PPU).

Enabling Quantum Breakthroughs

Hosted within every OPX+ controller and OPX1000 frontend module (FEM) is a PPU. PPU orchestrates protocols that were not previously possible on the quantum coherence timescale. Thus, dramatically accelerating runtime and enabling quantum technology breakthroughs.

Unmatched Real-time Processing Capabilities

Implemented on a robust UltraScale+ FPGA, PPU is a multi-core processor featuring 18 cores with OPX+ and 16 cores with every OPX1000’s FEM (over 100 cores with a fully populated OPX1000 chassis). These cores operate in parallel, generating pulses to operate qubits and receive data from qubits. They manage control flow and execute real-time heavy lifting for classical calculations, which is crucial for the algorithms to run with optimal efficiency. With the PPU technology at their core, OPX quantum controllers deliver unmatched performance. They significantly reduce algorithm execution times, achieving speeds tens or even hundreds of times faster than other controllers.

Easy as Pseudo Coding

PPUs are programmed using QUA, a quantum pulse-level language. QUA enables intuitive programming of highly complex sequences, including multi-parameter calibrations, active reset with adaptive thresholds, NISQ algorithms, quantum error correction, sequences for experimenting with weak measurements, quasiparticle poisoning, and many others. When multiple OPX units and FEM modules are used to control a large-scale quantum processor, they are programmed together and operate as one big controller. The PPUs handle synchronization automatically, eliminating the need for manual alignment.

Gate-level programming is supported through the QUA OpenQASM3 extension, and HPC classical-quantum application development is facilitated via QUA integration with CUDA Quantum.

Accelerate
Experiments

Advanced control flow and parametric scans mean no more uploading of memory exhaustive sample-by-sample waveforms.

Easily Code
Complex Algorithms

Write high-level QUA programs executed in real-time on PPUs, saving months of tedious low-level FPGA programming.

Make Quantum
Breakthroughs

Run complex algorithms that were previously impossible to run.

Achieve Best
Fidelity

Implement ground state preparation with adaptive thresholds, real-time frequency tracking, QEC, and more. Use built-in pre-distortion FIR/IIR filters and crosstalk matrix.