Picasso: Memory-Efficient Graph Coloring Using Palettes With Applications in Quantum Computing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ferdous, S M, Neff, Reece, Peng, Bo, Shuvo, Salman, Minutoli, Marco, Mukherjee, Sayak, Kowalski, Karol, Becchi, Michela, Halappanavar, Mahantesh
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910328294473728
author Ferdous, S M
Neff, Reece
Peng, Bo
Shuvo, Salman
Minutoli, Marco
Mukherjee, Sayak
Kowalski, Karol
Becchi, Michela
Halappanavar, Mahantesh
author_facet Ferdous, S M
Neff, Reece
Peng, Bo
Shuvo, Salman
Minutoli, Marco
Mukherjee, Sayak
Kowalski, Karol
Becchi, Michela
Halappanavar, Mahantesh
contents A coloring of a graph is an assignment of colors to vertices such that no two neighboring vertices have the same color. The need for memory-efficient coloring algorithms is motivated by their application in computing clique partitions of graphs arising in quantum computations where the objective is to map a large set of Pauli strings into a compact set of unitaries. We present Picasso, a randomized memory-efficient iterative parallel graph coloring algorithm with theoretical sublinear space guarantees under practical assumptions. The parameters of our algorithm provide a trade-off between coloring quality and resource consumption. To assist the user, we also propose a machine learning model to predict the coloring algorithm's parameters considering these trade-offs. We provide a sequential and a parallel implementation of the proposed algorithm. We perform an experimental evaluation on a 64-core AMD CPU equipped with 512 GB of memory and an Nvidia A100 GPU with 40GB of memory. For a small dataset where existing coloring algorithms can be executed within the 512 GB memory budget, we show up to 68x memory savings. On massive datasets we demonstrate that GPU-accelerated Picasso can process inputs with 49.5x more Pauli strings (vertex set in our graph) and 2,478x more edges than state-of-the-art parallel approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2401_06713
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Picasso: Memory-Efficient Graph Coloring Using Palettes With Applications in Quantum Computing
Ferdous, S M
Neff, Reece
Peng, Bo
Shuvo, Salman
Minutoli, Marco
Mukherjee, Sayak
Kowalski, Karol
Becchi, Michela
Halappanavar, Mahantesh
Distributed, Parallel, and Cluster Computing
A coloring of a graph is an assignment of colors to vertices such that no two neighboring vertices have the same color. The need for memory-efficient coloring algorithms is motivated by their application in computing clique partitions of graphs arising in quantum computations where the objective is to map a large set of Pauli strings into a compact set of unitaries. We present Picasso, a randomized memory-efficient iterative parallel graph coloring algorithm with theoretical sublinear space guarantees under practical assumptions. The parameters of our algorithm provide a trade-off between coloring quality and resource consumption. To assist the user, we also propose a machine learning model to predict the coloring algorithm's parameters considering these trade-offs. We provide a sequential and a parallel implementation of the proposed algorithm. We perform an experimental evaluation on a 64-core AMD CPU equipped with 512 GB of memory and an Nvidia A100 GPU with 40GB of memory. For a small dataset where existing coloring algorithms can be executed within the 512 GB memory budget, we show up to 68x memory savings. On massive datasets we demonstrate that GPU-accelerated Picasso can process inputs with 49.5x more Pauli strings (vertex set in our graph) and 2,478x more edges than state-of-the-art parallel approaches.
title Picasso: Memory-Efficient Graph Coloring Using Palettes With Applications in Quantum Computing
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2401.06713