Consiglio Nazionale delle Ricerche

Tipo di prodottoArticolo in rivista
TitoloSolving linear equation systems on vector computers with maximum efficiency
Anno di pubblicazione1988
Formato
  • Elettronico
  • Cartaceo
Autore/iCorana A.; Martini C.; Morando M.; Ridella S.; Rolando C.
Affiliazioni autoriIstituto per i Circuiti Elettronici, Consiglio Nazionale delle Ricerche, 16145 Genova, Italy
Autori CNR e affiliazioni
  • MAURO MORANDO
  • CLAUDIO MARTINI
  • CLAUDIA ROLANDO
  • ANGELO CORANA
Lingua/e
  • inglese
AbstractA technique is presented for solving dense systems of linear equations by LU factorization with maximum performance on processors like FPS-120, FPS-5000 and X64 series, using FORTRAN with calls to elementary vector routines. A rearrangement of the matrix elements is done in order to compute all the matrix-vector operations involved in the LU factorization with only stride-1 dot-product operations, which are executed at peak speed in the FPS processors. Since only vector instructions are used, the algorithm is fully portable on all FPS 38/64 bit machines and in general on all vector computers with a similar memory structure. The performance obtained on FPS-100 and FPS M64/60 (FPS-264) processors is reported: the asymptotic speed is always the peak speed of the machine and the half performance length is N_1/2 = 238 for the FPS-100 and N_1/2 = 200 for the FFS M64/60. The N_1/2 vaLues could he lowered by using the APAL Assembly Language to code some critical parts, losing however the code portability.
Lingua abstractinglese
Altro abstract-
Lingua altro abstract-
Pagine da133
Pagine a139
Pagine totali-
RivistaParallel computing
Attiva dal 1984
Editore: North-Holland : Amsterdam - [poi] Elsevier Science [S.l.]
Paese di pubblicazione: Paesi Bassi
Lingua: inglese
ISSN: 0167-8191
Titolo chiave: Parallel computing
Titolo proprio: Parallel computing.
Titolo abbreviato: Parallel comput.
Numero volume della rivista8
Fascicolo della rivista1-3
DOI10.1016/0167-8191(88)90117-2
Verificato da refereeSì: Internazionale
Stato della pubblicazionePublished version
Indicizzazione (in banche dati controllate)
  • Scopus (Codice:2-s2.0-0024087735)
Parole chiavealgorithms; LU factorization; vector computers; FPS Attached Array Processors; efficiency
Link (URL, URI)http://www.scopus.com/inward/record.url?eid=2-s2.0-0024087735&partnerID=q2rCbXpz
Titolo parallelo-
Data di accettazione-
Note/Altre informazioni-
Strutture CNR
  • IEIIT — IEIIT - Sede secondaria di Genova
Moduli CNR-
Progetti Europei-
Allegati
  • Solving linear equation systems on vector computers with maximum efficiency