公开(公告)号：JP2003015825A

公开(公告)日：2003-01-17

申请号：JP2002136492

申请日：2002-05-13

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーション

Inventor： HODGES PAUL ,  HURLEY MICHAEL GARWOOD ,  OUCHI NORMAN KENNETH ,  SHIH MIEN

IPC: G06F3/06 ,  G06F13/16

CPC classification number: G06F13/1642

Abstract: PROBLEM TO BE SOLVED: To provide a high performance storage subsystem with a plurality of independent paths for communication between a storage controller and a storage device.
SOLUTION: This data processing system has a plurality of independent paths 352, 354 for the communication between a host and a plurality of storage devices. Each path has its own queue 413, 415, services a requests generated by the host and accesses the storage devices 345. Each request is assigned a unique sequential ID and after that, stored in all the queues along with the ID, Each storage device has a 'mailbox' register 500 to store the ID and a status of the executed latest ID. Queues are serviced and their status are updated based on the contents of the mailbox. Consequently, the queue in each path is allowed to be completely out of synchronization with each queue in other paths.
COPYRIGHT: (C)2003,JPO

Abstract translation: 要解决的问题：提供具有用于存储控制器和存储设备之间的通信的多个独立路径的高性能存储子系统。 解决方案：该数据处理系统具有用于主机和多个存储设备之间的通信的多个独立路径352,354。 每个路径具有自己的队列413,415，服务于由主机产生的请求并访问存储设备345.每个请求被分配唯一的顺序ID，之后与所有ID一起存储在所有队列中，每个存储设备具有 用于存储ID和所执行的最新ID的状态的“邮箱”寄存器500。 服务队列并根据邮箱的内容更新其状态。 因此，允许每个路径中的队列与其他路径中的每个队列完全不同步。

公开(公告)号：DE69724846D1

公开(公告)日：2003-10-23

申请号：DE69724846

申请日：1997-01-23

Applicant: IBM

Inventor： HODGES PAUL ,  HURLEY MICHAEL GARWOOD ,  OUCHI NORMAN KENNETH ,  SHIH MIEN

IPC: G06F3/06 ,  G06F13/16

Abstract: A data processing system (300, 800, 1000) having multiple independent paths for communication between a host (310, 810, 1010) and a plurality of storage devices (340, 840, 1040) where each path has its own queue (326, 828) for servicing requests generated by the host for accessing the storage devices. Each request is assigned a unique sequential ID before it is stored, along with its unique ID, in all the queues. Each storage device has a "mailbox" register (500) where the ID and the status (520) of the latest request being carried out is stored. Queues are serviced and their status updated based on the content of the mailbox in each storage device. The combination of assigning a unique task ID to each request and a "mailbox" register in each storage device allows the queue in each path to be completely out of sync with each of the queues in the other paths without causing data integrity problems, duplication of requests at the device level, or a need for complex locking schemes to keep the queues in sync with each other.

公开(公告)号：DE3278258D1

公开(公告)日：1988-04-21

申请号：DE3278258

申请日：1982-05-14

Applicant: IBM

Inventor： COLLINS THOMAS WILLIAM ,  HURLEY MICHAEL GARWOOD

IPC: G11C11/14 ,  G11C19/08

Abstract: A major/minor loop bubble memory system includes a passive replicator (34) in the major loop read channel (33) which is connected by a first path (42) to a mode switch-annihilator (44) and a merge point in the major loop write channel (30) and by a second path (36) to an off-chip decision-making means (38) and the merge point in the write channel. The decision-making means (38) is positioned the same or fewer propagation steps than the mode switch-annihilator (44) is from the replicator (34). The decision-making means (38) is activated to cause either the replicated data to pass through the mode switch-annihilator (44) into the write channel or the replicated data to be annihilated in the mode switch-annihilator and the data from a generator (40) to pass into the write channel.

公开(公告)号：DE69724846T2

公开(公告)日：2004-07-08

申请号：DE69724846

申请日：1997-01-23

Applicant: IBM

Inventor： HODGES PAUL ,  HURLEY MICHAEL GARWOOD ,  OUCHI NORMAN KENNETH ,  SHIH MIEN

IPC: G06F3/06 ,  G06F13/16

Abstract: A data processing system (300, 800, 1000) having multiple independent paths for communication between a host (310, 810, 1010) and a plurality of storage devices (340, 840, 1040) where each path has its own queue (326, 828) for servicing requests generated by the host for accessing the storage devices. Each request is assigned a unique sequential ID before it is stored, along with its unique ID, in all the queues. Each storage device has a "mailbox" register (500) where the ID and the status (520) of the latest request being carried out is stored. Queues are serviced and their status updated based on the content of the mailbox in each storage device. The combination of assigning a unique task ID to each request and a "mailbox" register in each storage device allows the queue in each path to be completely out of sync with each of the queues in the other paths without causing data integrity problems, duplication of requests at the device level, or a need for complex locking schemes to keep the queues in sync with each other.