Analysis of Memory Footprints of Sparse Matrices Partitioned Into Uniformly-Sized Blocks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00010669%3A_____%2F18%3AN0000012" target="_blank" >RIV/00010669:_____/18:N0000012 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/68407700:21240/18:00322994

Výsledek na webu

<a href="https://www.researchgate.net/publication/327698036_Analysis_of_Memory_Footprints_of_Sparse_Matrices_Partitioned_Into_Uniformly-Sized_Blocks" target="_blank" >https://www.researchgate.net/publication/327698036_Analysis_of_Memory_Footprints_of_Sparse_Matrices_Partitioned_Into_Uniformly-Sized_Blocks</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.12694/scpe.v19i3.1358" target="_blank" >10.12694/scpe.v19i3.1358</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Analysis of Memory Footprints of Sparse Matrices Partitioned Into Uniformly-Sized Blocks

Popis výsledku v původním jazyce

The presented study analyses memory footprints of 563 representative benchmark sparse matrices with respect to their partitioning into uniformly-sized blocks. Different block sizes and different ways of storing blocks in memory are considered and statistically evaluated. Memory footprints of partitioned matrices are then compared with their lower bounds and CSR, index-compressed CSR, and EBF storage formats. The results show that block-based storage formats may significantly reduce memory footprints of sparse matrices arising from a wide range of application domains. Additionally, measured consistency of results is presented and discussed, benefits of individual formats for storing blocks are evaluated, and an analysis of best-case and worst-case matrices is provided for in-depth understanding of causes of memory savings of block-based formats.

Název v anglickém jazyce

Analysis of Memory Footprints of Sparse Matrices Partitioned Into Uniformly-Sized Blocks

Popis výsledku anglicky

The presented study analyses memory footprints of 563 representative benchmark sparse matrices with respect to their partitioning into uniformly-sized blocks. Different block sizes and different ways of storing blocks in memory are considered and statistically evaluated. Memory footprints of partitioned matrices are then compared with their lower bounds and CSR, index-compressed CSR, and EBF storage formats. The results show that block-based storage formats may significantly reduce memory footprints of sparse matrices arising from a wide range of application domains. Additionally, measured consistency of results is presented and discussed, benefits of individual formats for storing blocks are evaluated, and an analysis of best-case and worst-case matrices is provided for in-depth understanding of causes of memory savings of block-based formats.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA16-16772S" target="_blank" >GA16-16772S: Vývoj symetriemi-řízených metod pro modelování středně těžkých atomových jader z prvních principů</a><br>

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Scalable Computing: Practice and Experience

ISSN

1895-1767

e-ISSN

—

Svazek periodika

19

Číslo periodika v rámci svazku

3

Stát vydavatele periodika

RO - Rumunsko

Počet stran výsledku

17

Strana od-do

275-291

Kód UT WoS článku

000444726800005

EID výsledku v databázi Scopus

—