-
公开(公告)号:DE69230093T2
公开(公告)日:2000-04-13
申请号:DE69230093
申请日:1992-10-27
Applicant: IBM
Inventor: COLEMAN JOHN JOSEPH , COLEMAN RONALD GERALD , MONROE OWEN KEITH , STUCKE ROBERT FREDERICK , VANDERBECK ELIZABETH ANNE , BELLO STEPHEN EDWARD , HATTERSLEY JOHN ROBERT , HUA KIEN ANH , PRUETT DAVID RAYMOND , ROLLO GERALD FRANKLIN
IPC: G06F15/16 , G06F15/173
-
公开(公告)号:DE69230093D1
公开(公告)日:1999-11-11
申请号:DE69230093
申请日:1992-10-27
Applicant: IBM
Inventor: COLEMAN JOHN JOSEPH , COLEMAN RONALD GERALD , MONROE OWEN KEITH , STUCKE ROBERT FREDERICK , VANDERBECK ELIZABETH ANNE , BELLO STEPHEN EDWARD , HATTERSLEY JOHN ROBERT , HUA KIEN ANH , PRUETT DAVID RAYMOND , ROLLO GERALD FRANKLIN
IPC: G06F15/16 , G06F15/173
-
公开(公告)号:CA2107303C
公开(公告)日:1998-09-15
申请号:CA2107303
申请日:1993-09-29
Applicant: IBM
Inventor: COLEMAN RONALD GERALD , ADVANI DEEPAK MOHAN , MONROE OWEN KEITH , STRAUB ROBERT MICHAEL , GLEASON MINER HOLDON III , HUANG YUN-PONG
IPC: G06F13/00 , G06F15/16 , G06F15/173 , G06F15/177 , H04L29/06 , H04L7/00 , G06F13/368
Abstract: Apparatus and an accompanying method for implementing a communications protocol for use in generally a distributed, and particularly a highly parallel, multi-processing system (5), for communicating application data of arbitrarily v arying strides between separate processors and generally without the need for intermediate data storage. Specifically, application data is transferred through a succession of p ackets (770) between originating and destination application programs (252), with each packet containing a partial data message. Each partial data message contains a portion of the data that is stored in a stride one portion of an originating application data m emory (700). Upon reception, each data portion is written into a stride one portion of a dest ination application data memory (760). An application interface (254), executing indepen dently at both the originating and destination processing elements (10 1, 10J+1) locall y and respectively determines the starting address and length of each separate stride one data portion stored in the originating application data memory and for each separate stride one area in the destination application program memory that is to receive applicatio n data. Advantageously, through use of these independent application interfaces, applica tion data can be transferred from the originating application data memory at one arbitrari ly varying stride to the destination application memory with a different arbitrarily varyin g stride but without the need to pass the data through any input/output (I/O) buffers (632,63 4).
-
-