Multi-Core Quantum Compiler Mapping to Reduce Inter-Core Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing quantum compiler methods fail to optimize performance in multi-core quantum computing systems by accounting for both inter-operation and information exchange operations, leading to inefficient execution.
Innovation Solution
A processor-implemented method that partitions quantum program code into segments, maps logical qubits to physical cores using a graph solver, and minimizes inter-core operations through entanglement-based links and qubit teleportation, while considering latency and qubit distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If quantum code is executed on multi-core quantum computing systems, then computing power and scalability are improved, but execution performance deteriorates due to inter-core operations and information exchange overhead
Solution Approach 1:
The patent segments the quantum code into multiple segments and assigns each segment to a specific core for execution. This segmentation allows parallel processing across multiple cores while minimizing inter-core communication by ensuring that each segment can be executed relatively independently, thus reducing the performance degradation caused by inter-core operations.
Solution Approach 2:
The patent performs preliminary actions by pre-processing the quantum code to identify and separate operations that can be executed independently on different cores. It also pre-establishes the mapping between code segments and cores, and prepares the necessary entanglement links in advance, thereby reducing the overhead during actual execution and improving overall performance.
2Productivity
If code is partitioned into multiple segments for parallel execution, then productivity is improved, but device complexity increases due to segment management and inter-core coordination
Solution Approach 1:
The patent employs a universal compiler framework that can handle different quantum architectures and code segments through a common set of rules and algorithms. This universal approach simplifies the management of code segmentation and inter-core coordination by providing standardized interfaces and procedures, thereby reducing the apparent complexity despite the parallel execution model.
Solution Approach 2:
The patent introduces an intermediary compiler layer that acts as a mediator between the high-level quantum code and the multi-core execution system. This intermediary handles the complex tasks of code segmentation, core assignment, and synchronization, shielding the user from the underlying complexity while enabling efficient parallel execution.
3Adaptability or versatility
If entanglement-based inter-core links are used for communication, then information exchange capability is improved, but latency increases due to random generation time of entanglement links
Solution Approach 1:
The patent performs preliminary entanglement generation and link establishment before the actual quantum operations require communication between cores. By pre-setting up the entanglement links, the system avoids the random generation delays during critical execution phases, thereby reducing the effective latency experienced during code execution while maintaining the adaptability of entanglement-based communication.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method optimizes code execution performance and reliability in multi-core quantum computers by reducing inter-core operations and latency, achieving efficient resource utilization.
Implementation Method 1
the cores are linked by entanglement-based one-time-use links whose generation time duration varies randomly
Implementation Method 2
information exchange operations where the information encoded in a qubit is moved to another qubit on a different core e.g. using teleportation
Data Source
AI summary
A method is provided for mapping a quantum program code to a multi-core quantum computing system. The method includes partitioning code into code segments; identifying, from the code segments, a first group (Gc) comprising at least parts of possible contiguous sequences of the code segments; identifying a second group (Gnc) comprising at least part of possible non-overlapping combinations of Gc members; converting each Gc member to a corresponding graph group (Ggc); mapping logical qubits contained in each Gc member to different physical cores; generating a respective group of solver results (Gsc); determining an amount of inter-operations (Asc) related to the contiguous code part corresponding to a respective Gsc; determining a group of inter-operation amounts (Gamnt) based on the Asc; and determining an optimal compiled code structure having a smallest amount of Gamnt members.