Improving Locality-Aware Scheduling with Acyclic Directed Graph Partitioning

Benoît, Anne; Çatalyürek, Ümit V.

doi:10.1007/978-3-030-43229-4_19

Cited by 1 publication

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3) Acylic-partitioning-based Partitioning (dagP): We propose a novel acyclic DAG-partitioning-based heuristic by extending and modifying a recent open-source, state-of-the-art acyclic DAG partitioner [20], [21]. It presents an acyclic DAG partitioning algorithm that consists of an acyclic agglomerative clustering, recursive-bisection-based initial partitioning, and refinement phases.…”

Section: B Proposed Partitioning Methodsmentioning

confidence: 99%

“…However, as the number of qubits needed by the quantum circuit scales up, the working set size of the simulation task would inevitably exceed the cache size. For a modern CPU usually featuring a L3 cache (e.g., 32MB) shared by all cores, a L2 cache (e.g., 1MB) per core and a L1 cache (e.g.. 64KB) per core, the performance of the circuit simulation on more than 21 qubits (i.e., 2 21 •16 = 32MB) would be degraded by the capacity cache misses on L3 cache.…”

Section: A Sv-based Computational Patternmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Hierarchical State Vector Simulation of Quantum Circuits via Acyclic Graph Partitioning

Fang¹,

Li²,

Çatalyürek³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Early but promising results in quantum computing have been enabled by the concurrent development of quantum algorithms, devices, and materials. Classical simulation of quantum programs has enabled the design and analysis of algorithms and implementation strategies targeting current and anticipated quantum device architectures. In this paper, we present a graph-based approach to achieve efficient quantum circuit simulation. Our approach involves partitioning the graph representation of a given quantum circuit into acyclic subgraphs/circuits that exhibit better data locality. Simulation of each sub-circuit is organized hierarchically, with the iterative construction and simulation of smaller state vectors, improving overall performance. Also, this partitioning reduces the number of passes through data, improving the total computation time. We present three partitioning strategies and observe that acyclic graph partitioning typically results in the best time-to-solution. In contrast, other strategies reduce the partitioning time at the expense of potentially increased simulation times. Experimental evaluation demonstrates the effectiveness of our approach.

show abstract

Section: B Proposed Partitioning Methodsmentioning

confidence: 99%

Section: A Sv-based Computational Patternmentioning

confidence: 99%