1063:
111:(RAM) banks. These processors pass work to one another through a reconfigurable interconnect of channels. By harnessing many processors working in parallel, an MPPA chip can accomplish more demanding tasks than conventional chips. MPPAs are based on a software parallel programming model for developing high-performance embedded system applications.
389:
479:
331:
299:
460:
273:
727:
750:
31:
639:
100:
745:
722:
247:
227:
324:
717:
532:
824:
687:
1048:
882:
500:
420:
163:
117:
was an early implementation of a massively parallel computer architecture. MPP architectures are the second most common
1088:
1067:
1013:
473:
317:
168:
992:
787:
672:
634:
484:
374:
139:'s PDW commonly implement an MPP architecture to handle the processing of very large amounts of data in parallel.
1008:
987:
932:
819:
809:
782:
644:
174:
962:
588:
527:
440:
124:
104:
85:
877:
1093:
1023:
1018:
468:
153:
762:
694:
598:
490:
445:
552:
854:
814:
767:
757:
495:
415:
354:
66:
43:
794:
682:
677:
667:
654:
450:
108:
88:
becomes very important, and modern supercomputers have used various approaches ranging from enhanced
277:
957:
912:
738:
733:
712:
578:
74:
982:
804:
629:
593:
583:
384:
359:
340:
195:
185:
180:
158:
93:
47:
542:
1028:
704:
662:
557:
293:
243:
223:
1038:
837:
772:
619:
435:
430:
425:
394:
200:
81:
62:
17:
902:
842:
777:
624:
614:
547:
379:
369:
148:
77:, opportunistic grid system, whereby the grid provides power only on a best effort basis.
537:
1033:
849:
506:
399:
80:
Another approach is grouping many processors in close proximity to each other, as in a
58:
1082:
922:
799:
118:
46:(or separate computers) to simultaneously perform a set of coordinated computations
522:
114:
220:
Grid computing: experiment management, tool integration, and scientific workflows
1043:
103:(MPPAs), a type of integrated circuit with an array of hundreds or thousands of
89:
917:
892:
136:
69:
is opportunistically used whenever a computer is available. An example is
967:
947:
872:
128:
972:
952:
927:
562:
132:
54:
are massively parallel architecture with tens of thousands of threads.
942:
937:
309:
70:
977:
907:
897:
190:
84:. In such a centralized system the speed and flexibility of the
313:
887:
864:
51:
259:
Knight, Will: "IBM creates world's most powerful computer",
27:
Use of many processors to perform simultaneous operations
121:
implementations after clusters, as of
November 2013.
1001:
863:
703:
653:
607:
571:
515:
459:
408:
347:
240:
325:
8:
332:
318:
310:
65:of many computers in distributed, diverse
42:is the term for using a large number of
212:
298:: CS1 maint: archived copy as title (
291:
222:by Radu Prodan, Thomas Fahringer 2007
7:
242:by Francisco Fernández de Vega 2010
101:massively parallel processor arrays
32:Massively parallel (disambiguation)
25:
1062:
1061:
533:Analysis of parallel algorithms
1:
480:Simultaneous and heterogenous
261:NewScientist.com news service
92:systems to three-dimensional
1068:Category: Parallel computing
164:Process-oriented programming
169:Shared-nothing architecture
18:Massively parallel computer
1110:
375:High-performance computing
29:
1057:
1009:Automatic parallelization
645:Application checkpointing
175:Symmetric multiprocessing
125:Data warehouse appliances
99:The term also applies to
105:central processing units
1024:Embarrassingly parallel
1019:Deterministic algorithm
154:Embarrassingly parallel
739:Associative processing
695:Non-blocking algorithm
501:Clustered multi-thread
67:administrative domains
855:Hardware acceleration
768:Superscalar processor
758:Dataflow architecture
355:Distributed computing
734:Pipelined processing
683:Explicit parallelism
678:Implicit parallelism
668:Dataflow programming
109:random-access memory
30:For other uses, see
958:Parallel Extensions
763:Pipelined processor
94:torus interconnects
44:computer processors
1089:Parallel computing
832:Massively parallel
810:distributed shared
630:Cache invalidation
594:Instruction window
385:Manycore processor
365:Massively parallel
360:Parallel computing
341:Parallel computing
280:on 6 December 2013
196:Manycore processor
186:Cellular automaton
181:Connection Machine
159:Parallel computing
40:Massively parallel
1076:
1075:
1029:Parallel slowdown
663:Stream processing
553:Karp–Flatt metric
16:(Redirected from
1101:
1065:
1064:
1039:Software lockout
838:Computer cluster
773:Vector processor
728:Array processing
713:Flynn's taxonomy
620:Memory coherence
395:Computer network
334:
327:
320:
311:
304:
303:
297:
289:
287:
285:
276:. Archived from
270:
264:
257:
251:
237:
231:
217:
201:Vector processor
82:computer cluster
63:processing power
57:One approach is
21:
1109:
1108:
1104:
1103:
1102:
1100:
1099:
1098:
1079:
1078:
1077:
1072:
1053:
997:
903:Coarray Fortran
859:
843:Beowulf cluster
699:
649:
640:Synchronization
625:Cache coherence
615:Multiprocessing
603:
567:
548:Cost efficiency
543:Gustafson's law
511:
455:
404:
380:Multiprocessing
370:Cloud computing
343:
338:
308:
307:
290:
283:
281:
274:"Archived copy"
272:
271:
267:
258:
254:
238:
234:
218:
214:
209:
149:Multiprocessing
145:
75:volunteer-based
35:
28:
23:
22:
15:
12:
11:
5:
1107:
1105:
1097:
1096:
1094:Supercomputing
1091:
1081:
1080:
1074:
1073:
1071:
1070:
1058:
1055:
1054:
1052:
1051:
1046:
1041:
1036:
1034:Race condition
1031:
1026:
1021:
1016:
1011:
1005:
1003:
999:
998:
996:
995:
990:
985:
980:
975:
970:
965:
960:
955:
950:
945:
940:
935:
930:
925:
920:
915:
910:
905:
900:
895:
890:
885:
880:
875:
869:
867:
861:
860:
858:
857:
852:
847:
846:
845:
835:
829:
828:
827:
822:
817:
812:
807:
802:
792:
791:
790:
785:
778:Multiprocessor
775:
770:
765:
760:
755:
754:
753:
748:
743:
742:
741:
736:
731:
720:
709:
707:
701:
700:
698:
697:
692:
691:
690:
685:
680:
670:
665:
659:
657:
651:
650:
648:
647:
642:
637:
632:
627:
622:
617:
611:
609:
605:
604:
602:
601:
596:
591:
586:
581:
575:
573:
569:
568:
566:
565:
560:
555:
550:
545:
540:
535:
530:
525:
519:
517:
513:
512:
510:
509:
507:Hardware scout
504:
498:
493:
488:
482:
477:
471:
465:
463:
461:Multithreading
457:
456:
454:
453:
448:
443:
438:
433:
428:
423:
418:
412:
410:
406:
405:
403:
402:
400:Systolic array
397:
392:
387:
382:
377:
372:
367:
362:
357:
351:
349:
345:
344:
339:
337:
336:
329:
322:
314:
306:
305:
265:
252:
232:
211:
210:
208:
205:
204:
203:
198:
193:
191:CUDA framework
188:
183:
178:
172:
166:
161:
156:
151:
144:
141:
59:grid computing
26:
24:
14:
13:
10:
9:
6:
4:
3:
2:
1106:
1095:
1092:
1090:
1087:
1086:
1084:
1069:
1060:
1059:
1056:
1050:
1047:
1045:
1042:
1040:
1037:
1035:
1032:
1030:
1027:
1025:
1022:
1020:
1017:
1015:
1012:
1010:
1007:
1006:
1004:
1000:
994:
991:
989:
986:
984:
981:
979:
976:
974:
971:
969:
966:
964:
961:
959:
956:
954:
951:
949:
946:
944:
941:
939:
936:
934:
931:
929:
926:
924:
923:Global Arrays
921:
919:
916:
914:
911:
909:
906:
904:
901:
899:
896:
894:
891:
889:
886:
884:
881:
879:
876:
874:
871:
870:
868:
866:
862:
856:
853:
851:
850:Grid computer
848:
844:
841:
840:
839:
836:
833:
830:
826:
823:
821:
818:
816:
813:
811:
808:
806:
803:
801:
798:
797:
796:
793:
789:
786:
784:
781:
780:
779:
776:
774:
771:
769:
766:
764:
761:
759:
756:
752:
749:
747:
744:
740:
737:
735:
732:
729:
726:
725:
724:
721:
719:
716:
715:
714:
711:
710:
708:
706:
702:
696:
693:
689:
686:
684:
681:
679:
676:
675:
674:
671:
669:
666:
664:
661:
660:
658:
656:
652:
646:
643:
641:
638:
636:
633:
631:
628:
626:
623:
621:
618:
616:
613:
612:
610:
606:
600:
597:
595:
592:
590:
587:
585:
582:
580:
577:
576:
574:
570:
564:
561:
559:
556:
554:
551:
549:
546:
544:
541:
539:
536:
534:
531:
529:
526:
524:
521:
520:
518:
514:
508:
505:
502:
499:
497:
494:
492:
489:
486:
483:
481:
478:
475:
472:
470:
467:
466:
464:
462:
458:
452:
449:
447:
444:
442:
439:
437:
434:
432:
429:
427:
424:
422:
419:
417:
414:
413:
411:
407:
401:
398:
396:
393:
391:
388:
386:
383:
381:
378:
376:
373:
371:
368:
366:
363:
361:
358:
356:
353:
352:
350:
346:
342:
335:
330:
328:
323:
321:
316:
315:
312:
301:
295:
279:
275:
269:
266:
262:
256:
253:
249:
248:3-642-10674-9
245:
241:
236:
233:
229:
228:3-540-69261-4
225:
221:
216:
213:
206:
202:
199:
197:
194:
192:
189:
187:
184:
182:
179:
176:
173:
170:
167:
165:
162:
160:
157:
155:
152:
150:
147:
146:
142:
140:
138:
134:
130:
126:
122:
120:
119:supercomputer
116:
112:
110:
106:
102:
97:
95:
91:
87:
83:
78:
76:
72:
68:
64:
60:
55:
53:
49:
45:
41:
37:
33:
19:
831:
608:Coordination
538:Amdahl's law
474:Simultaneous
364:
282:. Retrieved
278:the original
268:
260:
255:
239:
235:
219:
215:
123:
115:Goodyear MPP
113:
98:
86:interconnect
79:
61:, where the
56:
39:
38:
36:
1044:Scalability
805:distributed
688:Concurrency
655:Programming
496:Cooperative
485:Speculative
421:Instruction
263:, June 2007
250:pages 65–68
107:(CPUs) and
48:in parallel
1083:Categories
1049:Starvation
788:asymmetric
523:PRAM model
491:Preemptive
284:12 January
207:References
90:InfiniBand
783:symmetric
528:PEM model
230:pages 1–4
137:Microsoft
1014:Deadlock
1002:Problems
968:pthreads
948:OpenHMPP
873:Ateji PX
834:computer
705:Hardware
572:Elements
558:Slowdown
469:Temporal
451:Pipeline
294:cite web
143:See also
129:Teradata
127:such as
973:RaftLib
953:OpenACC
928:GPUOpen
918:C++ AMP
893:Charm++
635:Barrier
579:Process
563:Speedup
348:General
133:Netezza
1066:
943:OpenCL
938:OpenMP
883:Chapel
800:shared
795:Memory
730:(SIMT)
673:Models
584:Thread
516:Theory
487:(SpMT)
441:Memory
426:Thread
409:Levels
246:
226:
913:Dryad
878:Boost
599:Array
589:Fiber
503:(CMT)
476:(SMT)
390:GPGPU
177:(SMP)
71:BOINC
978:ROCm
908:CUDA
898:Cilk
865:APIs
825:COMA
820:NUMA
751:MIMD
746:MISD
723:SIMD
718:SISD
446:Loop
436:Data
431:Task
300:link
286:2014
244:ISBN
224:ISBN
171:(SN)
73:, a
52:GPUs
993:ZPL
988:TBB
983:UPC
963:PVM
933:MPI
888:HPX
815:UMA
416:Bit
135:or
1085::
296:}}
292:{{
131:,
96:.
50:.
333:e
326:t
319:v
302:)
288:.
34:.
20:)
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.