Docs
Composer

Composer

Composer: Smart Scheduling for Quantum-Classical Execution

Composer is Qoro Quantum’s intelligent orchestration layer for distributing quantum workloads across heterogeneous hardware and simulators. Its core purpose is to turn complex quantum-classical workflows into efficient, executable schedules—optimized for performance, reliability, and cost.

As quantum infrastructure grows more fragmented and varied (from NISQ-era hardware to high-fidelity simulators), Composer provides the logic required to make smart, context-aware decisions about where, when, and how to run each part of a quantum program.

Key Responsibilities

1. Network-Aware Scheduling

Composer analyzes available computing resources—including real quantum processors, classical co-processors (e.g. GPUs, CPUs), and simulators—and selects the most appropriate backend for each task.

  • Routes simpler jobs to simulators
  • Offloads larger circuits to quantum hardware
  • Manages queue times, resource constraints, and hardware fidelity
  • Adapts to multi-vendor environments and cloud-based quantum access

2. Circuit Cutting and Aggregation

For circuits too large to fit on a single quantum device, Composer can:

  • Cut circuits into smaller, independently executable fragments
  • Insert classical stitching logic for reconstruction
  • Aggregate results via post-processing to yield the final output

3. Circuit Packing

Quantum hardware is often underutilized due to uneven circuit sizes. Composer introduces circuit packing to optimize for hardware utilization:

  • Identifies unused qubits on target hardware
  • Packs multiple compatible circuits together into one larger circuit
  • Reduces overhead and maximizes throughput on constrained devices

This is especially useful in hybrid workloads or batched execution scenarios where many small circuits must be executed efficiently.

When to Use Composer

Use Composer when your quantum workloads:

  • Span multiple devices, hardware types, or geographic locations
  • Exceed current hardware qubit limits
  • Require queue-aware or latency-sensitive scheduling
  • Include a mix of simulable and hardware-only tasks
  • Can benefit from smart job packing to reduce cost or idle time

Composer works in concert with Maestro and Divi, feeding optimized schedules to simulation or hardware backends, and seamlessly integrating with Divi’s parallel execution model.