Pipelined Preconditioned Conjugate Gradient Methods for real and complex linear systems for distributed memory architectures
•We introduce PIPECG-OATI-c to solve complex Hermitian and symmetric linear systems.•PIPECG-OATI-c reduces synchronizations in PCG and overlaps them with computations.•We provide optimized implementations for PIPECG-OATI-c.•We obtain 25% performance improvement over PCG for 1B problem on 16K cores.•...
Gespeichert in:
| Veröffentlicht in: | Journal of parallel and distributed computing Jg. 163; S. 147 - 155 |
|---|---|
| Hauptverfasser: | , |
| Format: | Journal Article |
| Sprache: | Englisch |
| Veröffentlicht: |
Elsevier Inc
01.05.2022
|
| Schlagworte: | |
| ISSN: | 0743-7315, 1096-0848 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | •We introduce PIPECG-OATI-c to solve complex Hermitian and symmetric linear systems.•PIPECG-OATI-c reduces synchronizations in PCG and overlaps them with computations.•We provide optimized implementations for PIPECG-OATI-c.•We obtain 25% performance improvement over PCG for 1B problem on 16K cores.•We obtain 2.48X performance improvement over PCG for 20M problem on 16K cores.
Preconditioned Conjugate Gradient (PCG) is a popular method for solving large and sparse linear systems of equations. The performance of PCG at scale is affected due to the costly global synchronization steps that arise in dot-products on distributed memory systems. Pipelined PCG (PIPECG) removes the costly global synchronization steps from PCG by only executing a single non-blocking allreduce per iteration and overlapping it with independent computations.
In our previous work, we have developed a novel pipelined PCG algorithm called PIPECG-OATI (One Allreduce per Two Iterations) for real linear systems which executes a single non-blocking allreduce per two iterations and provides a large overlap of global communication with independent computations at higher number of cores. Our method achieves this overlap by using iteration combination and by introducing new recurrence and non-recurrence computations. We implement optimizations in the PIPECG-OATI method to use cache memory efficiently.
In this work, we present PIPECG-OATI-c method for linear systems with complex Hermitian positive definite and complex symmetric matrices. We compare our method with various pipelined CG methods on a variety of problems and demonstrate that our method always gives the least run times. We performed experiments with our method using 20M and 30M unknowns on up to 16K cores and obtained up to 2.48X performance improvement over PCG and 2.14X performance improvement over PIPECG methods. We also experimented with up to 1-billion unknowns on 16K cores, the largest problem size explored for the CG problem, to our knowledge, and obtained about 25% improvement over PCG. |
|---|---|
| AbstractList | •We introduce PIPECG-OATI-c to solve complex Hermitian and symmetric linear systems.•PIPECG-OATI-c reduces synchronizations in PCG and overlaps them with computations.•We provide optimized implementations for PIPECG-OATI-c.•We obtain 25% performance improvement over PCG for 1B problem on 16K cores.•We obtain 2.48X performance improvement over PCG for 20M problem on 16K cores.
Preconditioned Conjugate Gradient (PCG) is a popular method for solving large and sparse linear systems of equations. The performance of PCG at scale is affected due to the costly global synchronization steps that arise in dot-products on distributed memory systems. Pipelined PCG (PIPECG) removes the costly global synchronization steps from PCG by only executing a single non-blocking allreduce per iteration and overlapping it with independent computations.
In our previous work, we have developed a novel pipelined PCG algorithm called PIPECG-OATI (One Allreduce per Two Iterations) for real linear systems which executes a single non-blocking allreduce per two iterations and provides a large overlap of global communication with independent computations at higher number of cores. Our method achieves this overlap by using iteration combination and by introducing new recurrence and non-recurrence computations. We implement optimizations in the PIPECG-OATI method to use cache memory efficiently.
In this work, we present PIPECG-OATI-c method for linear systems with complex Hermitian positive definite and complex symmetric matrices. We compare our method with various pipelined CG methods on a variety of problems and demonstrate that our method always gives the least run times. We performed experiments with our method using 20M and 30M unknowns on up to 16K cores and obtained up to 2.48X performance improvement over PCG and 2.14X performance improvement over PIPECG methods. We also experimented with up to 1-billion unknowns on 16K cores, the largest problem size explored for the CG problem, to our knowledge, and obtained about 25% improvement over PCG. |
| Author | Tiwari, Manasi Vadhiyar, Sathish |
| Author_xml | – sequence: 1 givenname: Manasi orcidid: 0000-0002-7012-7846 surname: Tiwari fullname: Tiwari, Manasi email: manasitiwari@iisc.ac.in – sequence: 2 givenname: Sathish surname: Vadhiyar fullname: Vadhiyar, Sathish email: vss@iisc.ac.in |
| BookMark | eNp9kM1KAzEUhYNUsFZfwFVeYMab-elkwI0U_6BiF7oOmeTGZpiZlCQVCz68M9SVi64uB-534HyXZDa4AQm5YZAyYMvbNm13WqUZZFkKLAXgZ2TOoF4mwAs-I3OoijypclZekMsQWgDGyorPyc_G7rCzA2q68ajcoG20boorN7T7TxmRPnmpLQ6RvmLcOh2ocZ56lB2Vg6bK9bsOv-lUIj0NhxCxP_5oG6K3zT6OdT32zh-o9GprI6q49xiuyLmRXcDrv7sgH48P76vnZP329LK6XycqB4hJBszIutZ5rvKirivNTGVqKGuZZ40pG66Bm4pXhTIll7gsltpIWTCV8abUjcoXhB97lXcheDRC2SinndFL2wkGYrIoWjFZFJNFAUyMFkc0-4fuvO2lP5yG7o4QjqO-LHoR1ChQobaj4yi0s6fwX7oUkdI |
| CitedBy_id | crossref_primary_10_1016_j_istruc_2023_06_063 crossref_primary_10_1016_j_gsd_2024_101172 crossref_primary_10_3390_math12172629 |
| Cites_doi | 10.1177/1094342015593158 10.1137/0913023 10.1137/0905015 10.1016/0167-8191(87)90037-8 10.1016/j.parco.2013.06.001 10.1016/0167-8191(94)00087-Q 10.1137/0906018 10.1109/71.780863 10.6028/jres.049.044 10.1007/BF02145754 10.1137/0727091 10.1016/0377-0427(89)90045-9 10.1137/0720025 |
| ContentType | Journal Article |
| Copyright | 2022 Elsevier Inc. |
| Copyright_xml | – notice: 2022 Elsevier Inc. |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.jpdc.2022.01.008 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISSN | 1096-0848 |
| EndPage | 155 |
| ExternalDocumentID | 10_1016_j_jpdc_2022_01_008 S0743731522000156 |
| GroupedDBID | --K --M -~X .~1 0R~ 1B1 1~. 1~5 29L 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABEFU ABFNM ABFSI ABJNI ABMAC ABTAH ABXDB ABYKQ ACDAQ ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADFGL ADHUB ADJOM ADMUD ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHHHB AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CAG COF CS3 DM4 DU5 E.L EBS EFBJH EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-2 G-Q G8K GBLVA GBOLZ HLZ HVGLF HZ~ H~9 IHE J1W JJJVA K-O KOM LG5 LG9 LY7 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG ROL RPZ SBC SDF SDG SDP SES SET SEW SPC SPCBC SST SSV SSZ T5K TN5 TWZ WUQ XJT XOL XPP ZMT ZU3 ZY4 ~G- ~G0 9DU AATTM AAXKI AAYWO AAYXX ABDPE ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO ADVLN AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD |
| ID | FETCH-LOGICAL-c300t-201fa99d33c34997d1f7f9059a32bf5b8d08f7874cf58ae646dfaa41c28b5dbc3 |
| ISICitedReferencesCount | 3 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000777831100010&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0743-7315 |
| IngestDate | Sat Nov 29 07:19:55 EST 2025 Tue Nov 18 20:27:40 EST 2025 Fri Feb 23 02:40:25 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Complex Hermitian positive definite systems Overlapping communication and computations Pipelining Complex symmetric systems Preconditioned Conjugate Gradient |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c300t-201fa99d33c34997d1f7f9059a32bf5b8d08f7874cf58ae646dfaa41c28b5dbc3 |
| ORCID | 0000-0002-7012-7846 |
| PageCount | 9 |
| ParticipantIDs | crossref_citationtrail_10_1016_j_jpdc_2022_01_008 crossref_primary_10_1016_j_jpdc_2022_01_008 elsevier_sciencedirect_doi_10_1016_j_jpdc_2022_01_008 |
| PublicationCentury | 2000 |
| PublicationDate | May 2022 2022-05-00 |
| PublicationDateYYYYMMDD | 2022-05-01 |
| PublicationDate_xml | – month: 05 year: 2022 text: May 2022 |
| PublicationDecade | 2020 |
| PublicationTitle | Journal of parallel and distributed computing |
| PublicationYear | 2022 |
| Publisher | Elsevier Inc |
| Publisher_xml | – name: Elsevier Inc |
| References | J. Dongarra, M. Heroux, P. Luszczek, HPCG Benchmark: a new metric for ranking high performance computing systems Gropp (br0150) Meurant (br0110) 1987; 5 D'Azevedo, Eijkhout, Romine (br0120) 1993 Ghysels, Vanroose (br0160) 2014; 40 Hestenes, Stiefel (br0010) 1952; 49 Catalyurek, Aykanat (br0230) 1999; 10 Freund (br0060) 1992; 13 Tiwari, Vadhiyar (br0200) 2020 Pinar, Heath (br0210) 1999 Balay, Gropp, McInnes, Smith (br0090) 1997 Balay, Abhyankar, Adams, Brown, Brune, Buschelman, Dalcin, Dener, Eijkhout, Gropp, Karpeyev, Kaushik, Knepley, May, McInnes, Mills, Munson, Rupp, Sanan, Smith, Zampini, Zhang, Zhang (br0070) 2019 (br0290) 2019 (br0140) 2019 2015. Saad (br0100) 1984; 5 Ashby, Manteuffel, Saylor (br0040) 1990; 27 Cornelis, Cools, Vanroose (br0170) 2018 (br0300) 2020 Concus, Golub, Meurant (br0270) 1985; 6 Tatebe, Oyanagi (br0250) 1994 Balay, Abhyankar, Adams, Brown, Brune, Buschelman, Dalcin, Dener, Eijkhout, Gropp, Karpeyev, Kaushik, Knepley, May, McInnes, Mills, Munson, Rupp, Sanan, Smith, Zampini, Zhang, Zhang (br0080) 2020 White, Sadayappan (br0220) 1997 Romero, Zapata (br0240) 1995; 21 Johnson, Micchelli, Paul (br0260) 1983; 20 Serra-Capizzano (br0030) 1995; 2563 Cools, Cornelis, Vanroose (br0180) 2019 Joly, Meurant (br0050) 1993; 4 Chronopoulos, Gear (br0130) 1989; 25 Eller, Gropp (br0190) 2016 Saad (br0020) 2003 Pinar (10.1016/j.jpdc.2022.01.008_br0210) 1999 Concus (10.1016/j.jpdc.2022.01.008_br0270) 1985; 6 White (10.1016/j.jpdc.2022.01.008_br0220) 1997 Tatebe (10.1016/j.jpdc.2022.01.008_br0250) 1994 Freund (10.1016/j.jpdc.2022.01.008_br0060) 1992; 13 Meurant (10.1016/j.jpdc.2022.01.008_br0110) 1987; 5 Balay (10.1016/j.jpdc.2022.01.008_br0090) 1997 D'Azevedo (10.1016/j.jpdc.2022.01.008_br0120) 1993 Chronopoulos (10.1016/j.jpdc.2022.01.008_br0130) 1989; 25 Eller (10.1016/j.jpdc.2022.01.008_br0190) 2016 Johnson (10.1016/j.jpdc.2022.01.008_br0260) 1983; 20 Serra-Capizzano (10.1016/j.jpdc.2022.01.008_br0030) 1995; 2563 Saad (10.1016/j.jpdc.2022.01.008_br0100) 1984; 5 Ashby (10.1016/j.jpdc.2022.01.008_br0040) 1990; 27 10.1016/j.jpdc.2022.01.008_br0280 Ghysels (10.1016/j.jpdc.2022.01.008_br0160) 2014; 40 Saad (10.1016/j.jpdc.2022.01.008_br0020) 2003 Balay (10.1016/j.jpdc.2022.01.008_br0080) 2020 Hestenes (10.1016/j.jpdc.2022.01.008_br0010) 1952; 49 Catalyurek (10.1016/j.jpdc.2022.01.008_br0230) 1999; 10 Balay (10.1016/j.jpdc.2022.01.008_br0070) Tiwari (10.1016/j.jpdc.2022.01.008_br0200) 2020 Cornelis (10.1016/j.jpdc.2022.01.008_br0170) Romero (10.1016/j.jpdc.2022.01.008_br0240) 1995; 21 Gropp (10.1016/j.jpdc.2022.01.008_br0150) Joly (10.1016/j.jpdc.2022.01.008_br0050) 1993; 4 Cools (10.1016/j.jpdc.2022.01.008_br0180) |
| References_xml | – year: 2019 ident: br0180 article-title: Numerically stable recurrence relations for the communication hiding pipelined conjugate gradient method – year: 2003 ident: br0020 article-title: Iterative Methods for Sparse Linear Systems – volume: 5 start-page: 267 year: 1987 end-page: 280 ident: br0110 article-title: Multitasking the conjugate gradient method on the cray x-mp/48 publication-title: Parallel Comput. – volume: 40 start-page: 224 year: 2014 end-page: 238 ident: br0160 article-title: Hiding global synchronization latency in the preconditioned conjugate gradient algorithm publication-title: Parallel Comput. – volume: 49 start-page: 409 year: 1952 end-page: 436 ident: br0010 article-title: Methods of conjugate gradients for solving linear systems publication-title: J. Res. Natl. Bur. Stand. – reference: J. Dongarra, M. Heroux, P. Luszczek, HPCG Benchmark: a new metric for ranking high performance computing systems – volume: 10 start-page: 673 year: 1999 end-page: 693 ident: br0230 article-title: Hypergraph-partitioning-based decomposition for parallel sparse-matrix vector multiplication publication-title: IEEE Trans. Parallel Distrib. Syst. – volume: 21 start-page: 583 year: 1995 end-page: 605 ident: br0240 article-title: Data distributions for sparse matrix vector multiplication publication-title: Parallel Comput. – volume: 20 start-page: 362 year: 1983 end-page: 376 ident: br0260 article-title: Polynomial preconditioners for conjugate gradient calculations publication-title: SIAM J. Numer. Anal. – reference: , 2015. – start-page: 163 year: 1997 end-page: 202 ident: br0090 article-title: Efficient management of parallelism in object oriented numerical software libraries publication-title: Modern Software Tools in Scientific Computing – volume: 25 start-page: 153 year: 1989 end-page: 168 ident: br0130 article-title: S-step iterative methods for symmetric linear systems publication-title: J. Comput. Appl. Math. – year: 2018 ident: br0170 article-title: The communication-hiding conjugate gradient method with deep pipelines – year: 2020 ident: br0300 article-title: OpenFOAM Poisson solver example – start-page: 151 year: 2020 end-page: 160 ident: br0200 article-title: Pipelined preconditioned conjugate gradient methods for distributed memory systems publication-title: 27th IEEE International Conference on High Performance Computing, Data, and Analytics – year: 2019 ident: br0290 – volume: 27 start-page: 1542 year: 1990 end-page: 1568 ident: br0040 article-title: A taxonomy for conjugate gradient methods publication-title: SIAM J. Numer. Anal. – ident: br0150 article-title: Update on Libraries for Blue Waters – year: 2016 ident: br0190 article-title: Scalable non-blocking preconditioned conjugate gradient methods publication-title: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis – year: 2019 ident: br0140 – year: 1993 ident: br0120 article-title: Lapack working note 56: Reducing communication costs in the conjugate gradient algorithm on distributed memory multiprocessors – volume: 5 start-page: 203 year: 1984 end-page: 228 ident: br0100 article-title: Practical use of some Krylov subspace methods for solving indefinite and nonsymmetric linear systems publication-title: SIAM J. Sci. Stat. Comput. – volume: 6 start-page: 220 year: 1985 end-page: 252 ident: br0270 article-title: Block preconditioning for the conjugate gradient method publication-title: SIAM J. Sci. Stat. Comput. – start-page: 194 year: 1994 end-page: 203 ident: br0250 article-title: Efficient implementation of the multigrid preconditioned conjugate gradient method on distributed memory machines publication-title: Supercomputing '94:Proceedings of the 1994 ACM/IEEE Conference on Supercomputing – start-page: 66 year: 1997 end-page: 71 ident: br0220 article-title: On improving the performance of sparse matrix-vector multiplication publication-title: Proceedings Fourth International Conference on High-Performance Computing – year: 2019 ident: br0070 – volume: 13 start-page: 425 year: 1992 end-page: 448 ident: br0060 article-title: Conjugate gradient-type methods for linear systems with complex symmetric coefficient matrices publication-title: SIAM J. Sci. Stat. Comput. – volume: 2563 start-page: 326 year: 1995 end-page: 337 ident: br0030 article-title: Conditioning and solution of Hermitian (block) Toeplitz systems by means of preconditioned conjugate gradient methods publication-title: Proc. SPIE Int. Soc. Opt. Eng. – volume: 4 start-page: 379 year: 1993 end-page: 406 ident: br0050 article-title: Complex conjugate gradient methods publication-title: Numer. Algorithms – start-page: 30 year: 1999 ident: br0210 article-title: Improving performance of sparse matrix-vector multiplication publication-title: SC '99: Proceedings of the 1999 ACM/IEEE Conference on Supercomputing – year: 2020 ident: br0080 article-title: PETSc users manual – volume: 2563 start-page: 326 year: 1995 ident: 10.1016/j.jpdc.2022.01.008_br0030 article-title: Conditioning and solution of Hermitian (block) Toeplitz systems by means of preconditioned conjugate gradient methods publication-title: Proc. SPIE Int. Soc. Opt. Eng. – ident: 10.1016/j.jpdc.2022.01.008_br0170 – ident: 10.1016/j.jpdc.2022.01.008_br0280 doi: 10.1177/1094342015593158 – volume: 13 start-page: 425 issue: 1 year: 1992 ident: 10.1016/j.jpdc.2022.01.008_br0060 article-title: Conjugate gradient-type methods for linear systems with complex symmetric coefficient matrices publication-title: SIAM J. Sci. Stat. Comput. doi: 10.1137/0913023 – volume: 5 start-page: 203 issue: 1 year: 1984 ident: 10.1016/j.jpdc.2022.01.008_br0100 article-title: Practical use of some Krylov subspace methods for solving indefinite and nonsymmetric linear systems publication-title: SIAM J. Sci. Stat. Comput. doi: 10.1137/0905015 – start-page: 194 year: 1994 ident: 10.1016/j.jpdc.2022.01.008_br0250 article-title: Efficient implementation of the multigrid preconditioned conjugate gradient method on distributed memory machines – year: 2016 ident: 10.1016/j.jpdc.2022.01.008_br0190 article-title: Scalable non-blocking preconditioned conjugate gradient methods – volume: 5 start-page: 267 issue: 3 year: 1987 ident: 10.1016/j.jpdc.2022.01.008_br0110 article-title: Multitasking the conjugate gradient method on the cray x-mp/48 publication-title: Parallel Comput. doi: 10.1016/0167-8191(87)90037-8 – volume: 40 start-page: 224 issue: 7 year: 2014 ident: 10.1016/j.jpdc.2022.01.008_br0160 article-title: Hiding global synchronization latency in the preconditioned conjugate gradient algorithm publication-title: Parallel Comput. doi: 10.1016/j.parco.2013.06.001 – volume: 21 start-page: 583 issue: 4 year: 1995 ident: 10.1016/j.jpdc.2022.01.008_br0240 article-title: Data distributions for sparse matrix vector multiplication publication-title: Parallel Comput. doi: 10.1016/0167-8191(94)00087-Q – year: 2020 ident: 10.1016/j.jpdc.2022.01.008_br0080 – volume: 6 start-page: 220 issue: 1 year: 1985 ident: 10.1016/j.jpdc.2022.01.008_br0270 article-title: Block preconditioning for the conjugate gradient method publication-title: SIAM J. Sci. Stat. Comput. doi: 10.1137/0906018 – start-page: 163 year: 1997 ident: 10.1016/j.jpdc.2022.01.008_br0090 article-title: Efficient management of parallelism in object oriented numerical software libraries – year: 1993 ident: 10.1016/j.jpdc.2022.01.008_br0120 – start-page: 66 year: 1997 ident: 10.1016/j.jpdc.2022.01.008_br0220 article-title: On improving the performance of sparse matrix-vector multiplication – volume: 10 start-page: 673 issue: 7 year: 1999 ident: 10.1016/j.jpdc.2022.01.008_br0230 article-title: Hypergraph-partitioning-based decomposition for parallel sparse-matrix vector multiplication publication-title: IEEE Trans. Parallel Distrib. Syst. doi: 10.1109/71.780863 – volume: 49 start-page: 409 year: 1952 ident: 10.1016/j.jpdc.2022.01.008_br0010 article-title: Methods of conjugate gradients for solving linear systems publication-title: J. Res. Natl. Bur. Stand. doi: 10.6028/jres.049.044 – ident: 10.1016/j.jpdc.2022.01.008_br0180 – start-page: 30 year: 1999 ident: 10.1016/j.jpdc.2022.01.008_br0210 article-title: Improving performance of sparse matrix-vector multiplication – volume: 4 start-page: 379 year: 1993 ident: 10.1016/j.jpdc.2022.01.008_br0050 article-title: Complex conjugate gradient methods publication-title: Numer. Algorithms doi: 10.1007/BF02145754 – volume: 27 start-page: 1542 issue: 6 year: 1990 ident: 10.1016/j.jpdc.2022.01.008_br0040 article-title: A taxonomy for conjugate gradient methods publication-title: SIAM J. Numer. Anal. doi: 10.1137/0727091 – volume: 25 start-page: 153 issue: 2 year: 1989 ident: 10.1016/j.jpdc.2022.01.008_br0130 article-title: S-step iterative methods for symmetric linear systems publication-title: J. Comput. Appl. Math. doi: 10.1016/0377-0427(89)90045-9 – ident: 10.1016/j.jpdc.2022.01.008_br0070 – year: 2003 ident: 10.1016/j.jpdc.2022.01.008_br0020 – volume: 20 start-page: 362 issue: 2 year: 1983 ident: 10.1016/j.jpdc.2022.01.008_br0260 article-title: Polynomial preconditioners for conjugate gradient calculations publication-title: SIAM J. Numer. Anal. doi: 10.1137/0720025 – start-page: 151 year: 2020 ident: 10.1016/j.jpdc.2022.01.008_br0200 article-title: Pipelined preconditioned conjugate gradient methods for distributed memory systems – ident: 10.1016/j.jpdc.2022.01.008_br0150 |
| SSID | ssj0011578 |
| Score | 2.3349729 |
| Snippet | •We introduce PIPECG-OATI-c to solve complex Hermitian and symmetric linear systems.•PIPECG-OATI-c reduces synchronizations in PCG and overlaps them with... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 147 |
| SubjectTerms | Complex Hermitian positive definite systems Complex symmetric systems Overlapping communication and computations Pipelining Preconditioned Conjugate Gradient |
| Title | Pipelined Preconditioned Conjugate Gradient Methods for real and complex linear systems for distributed memory architectures |
| URI | https://dx.doi.org/10.1016/j.jpdc.2022.01.008 |
| Volume | 163 |
| WOSCitedRecordID | wos000777831100010&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1096-0848 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0011578 issn: 0743-7315 databaseCode: AIEXJ dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1La9wwEBbbpIde-i5N-kCH3oyDbVkr6xhK-qINgaRhb0a2JLLLdll2N-kG8kP6czvSSI63j9AWejGLkbSy5_PMSJr5hpBXJThsivEm1bywaTkcNqnKK5sKzsGfs7AA8Qftpx_F4WE1GsmjweBbzIW5mIrZrFqv5fy_ihrugbBd6uxfiLsbFG7AbxA6XEHscP0jwR-N5y7H3LgkALfa1chGpF1y3-Tc7Zolbxc-0GuVfPL1oz0lQ7IwgTbAR5mbdeIGUZHrGdtoR7PrKmTBcF9cjO5l0j-JWP7G1XX84tOpweH7Y7S-pkS0nj4a-KvC3HcXl7Mcx_unSp-NLzEY_NgXUT7r71fAUreLDsRNtJhIsxHn6ZlSBcPMzj2Dujhz8dEVEnF2yjqoQ1S3ObJ1BsudI-HvT0YB9ycme5O5dqSVReGJWrPq2gR2gYnHbiJuHkWBWea3yHYhuAR9ub3__mD0oTuhyjla-TjxkJCFsYM__tOvnZ6eI3Nyn9wNYqH7iJwHZGBmD8m9WN2DBmX_iFx1QKKbQKIdkGgEEg1AogAS6oBEQdI0AIkikGgAkm_TAwFFINENID0mn98cnLx-l4ZaHWnLsmwFX2BulZSasZbBIlro3AorwXdXrGgsbyqdVRaMQ9laXikzLIfaKlXmbVE1XDcte0K2ZvAMTwk1QjdlJsH3bUrwloxkWrbCCNYIeNGF3CF5fJl1G4jsXT2VaR0jFie1E0DtBFBneQ0C2CFJ12eONC43tuZRRnVwRNHBrAFSN_Tb_cd-z8id6y_lOdlaLc7NC3K7vViNl4uXAXnfAZEVs-Q |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Pipelined+Preconditioned+Conjugate+Gradient+Methods+for+real+and+complex+linear+systems+for+distributed+memory+architectures&rft.jtitle=Journal+of+parallel+and+distributed+computing&rft.au=Tiwari%2C+Manasi&rft.au=Vadhiyar%2C+Sathish&rft.date=2022-05-01&rft.pub=Elsevier+Inc&rft.issn=0743-7315&rft.eissn=1096-0848&rft.volume=163&rft.spage=147&rft.epage=155&rft_id=info:doi/10.1016%2Fj.jpdc.2022.01.008&rft.externalDocID=S0743731522000156 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0743-7315&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0743-7315&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0743-7315&client=summon |