Abstract: An apparatus for providing storage control in a network of storage controllers is disclosed. The apparatus includes an owner storage controller; an I/O performing component, an ownership assignment component, a lock manager and a messaging component. The ownership assignment component assigns ownership of metadata for data to an owner storage controller. The lock manager controls the locking of metadata during I/O. The messaging component passes messages among storage controllers to request metadata state, to grant locks, to request release of locks, and to signal lock release. The I/O is performed on data whose metadata is owned by an owner storage controller, subject to compliance with metadata lock protocols controlled by the owner storage controller, and any copy of the data held from time to time is maintained in a coherency relation with the data.

Abstract: A storage network includes a path processor for directing data onto one of two or more paths between an application host and two or more storage controllers in a storage subsystem. To determine a data path within the storage network, the operational status of a storage controller in the storage subsystem is monitored. It is determined whether a change in the operational status of the storage controller is a failure or a potential failure. This determination is sent to the path processor. The path processor redirects application host data from the path controller on the basis of the determination.

Abstract: A storage control system comprising: one or more I/O processors for receiving requests to write data; a batch controller for selecting one or more write requests to be placed in a first batch; a first server for transmitting the first batch to a second server in a remote copy pair system; one or more I/O handlers for writing the first batch to completion on a data storage medium, in response to receiving permission from the second server; and one or more clients for communicating with the I/O processors and the I/O handlers, wherein the permission is granted when the second server is notified by the I/O handlers that previous write requests have been completed, wherein the first server is responsive to the completion of the first batch.

Abstract: This invention relates to an asynchronous remote copying (ARC) system adapted to operate as a remote copy pair by communicating between primary storage and remote storage of a remote copy pair. The system comprises a primary controller for receiving a write command and writing data to primary storage; a remote controller for receiving a write command and writing data to remote storage; an ARC controller for suppressing the release of the write command to the remote controller if a delay between receiving the write command and the time when the remote storage may accept the data is more than a threshold delay, and the primary controller suppresses recording the data in primary storage until release of the write command to the remote storage.

Abstract: An apparatus configured to operate as a primary for a remote copy pair is provided. The apparatus comprises a journal component at said primary for generating journal entries for one or more uncompleted sequenced batches; a configurator responsive to an error indication from a secondary for enumeration of said one or more uncompleted sequenced batches; and a primary server responsive to said enumeration of said one or more uncompleted sequenced batches to reissue to said secondary one or more write requests for said one or more uncompleted sequenced batches, during recovery processing.

Abstract: A storage apparatus is operable as primary in a remote copy pair and comprises a remote copy component operable to establish a remote copy relationship between said primary and a secondary; a copy component operable at the primary to create a copy for download onto a portable physical storage medium for offline transport to the secondary for upload; a synchronization component for synchronizing data at said secondary with data at said primary using an online link in response to a request for synchronization from the secondary; a metadata component operable to store a dirty state indicator of a portion of a storage space at the primary after establishment of the remote copy relationship at the primary; and the metadata component being operable to limit synchronization at the secondary to the portion of storage having a dirty state indicator at the primary.

Abstract: A method and apparatus for managing a loop network, the loop network (200) including at least one loop (206, 208), a plurality of devices (210) connected to the at least one loop (206, 208) via ports (211, 212), wherein at least two of the devices are initiators (207, 209). The method includes each initiator (207, 209) sending a frame to all other initiators (207, 209) in the loop network (200) identifying any ports (211, 212) which should not be used. Each initiator (207, 209) merges the information from all other initiators (207, 209) with its own information identifying any ports (211, 212) which should not be used resulting in all the initiators (207, 209) generating a single list of ports (211, 212) to be used which is consistent across all the initiators (207, 209). Each initiator (207, 209) applies an algorithm (300) to determine a common set of ports (211, 212) to be used by all the initiators (207, 209) and to balance port accesses across the loop network (200).

Abstract: A copy engine (104) is provided as an interface between firmware (108) and memory space (106) for carrying out copy operations. The copy engine has a first register (202, 203) to point to a first address and a second register (204, 205) to point to a second address. One of the first and second addresses is a source address and one is a destination address for data to be copied. The copy engine (104) also has a control register (201). The control register (201) includes a count of the amount of memory space required by a copy operation, an indication of the direction of the copy operation from the first address to the second address or from the second address to the first address, and an indication of whether the first memory address is incremented or decremented. The copy engine (104) includes a locking mechanism for locking the copy engine (104) during a copy operation. The copy engine (104) also includes a serialisation mechanism in which a write is made to the control register (201) of zero count.

Abstract: In a stacked system, errors detected at a user application interface of one or more host systems are associated with root cause errors at a stack level below a virtualization layer by detecting an error at a user application interface; identifying an associated root cause error at a lower stack level; creating an error trace entry for the error; associating an error log identifier with the error trace entry; making the combined error log identifier and the error trace entry into an error identifier that is unique within the plurality of host systems in said stacked system; and communicating the error identifier to any requester of a service at a user application interface of one or more host systems when the service must be failed because of the root cause error.

Abstract: A copy engine (104) is provided as an interface between firmware (108) and memory space (106) for carrying out copy operations. The copy engine has a first register (202, 203) to point to a first address and a second register (204, 205) to point to a second address. One of the first and second addresses is a source address and one is a destination address for data to be copied. The copy engine (104) also has a control register (201). The control register (201) includes a count of the amount of memory space required by a copy operation, an indication of the direction of the copy operation from the first address to the second address or from the second address to the first address, and an indication of whether the first memory address is incremented or decremented. The copy engine (104) includes a locking mechanism for locking the copy engine (104) during a copy operation. The copy engine (104) also includes a serialisation mechanism in which a write is made to the control register (201) of zero count.

Abstract: A salad container assembly is disclosed including a base, a cover removably mounted on the base, and a sealed salad dressing insert for nesting within the cover having a flexible top, a salad dressing reservoir, an opening, a seal over the opening, and a puncturing mechanism capable of breaking the seal on application of sufficient pressure to the flexible top thereby allowing the contents of the reservoir to dispense into the base.

Abstract: A data processing system comprises a master processor (10), a slave processor (30), a memory (50), and a bus subsystem (20) interconnecting the master processor (10), the slave processor (30), and the memory (50). The master processor (10) is configured to generate, in response to a memory access instruction, a read request comprising a read command for execution by the slave processor (30) to read data stored in a location in the memory (50) specified by the memory access instruction, and to write the read request to the slave processor (30) via the bus subsystem (20). The slave processor (30) is configured to execute the read command received in the read request from the master processor (10) to obtain the data stored at the specified location in the memory (50) and to write the data thus obtained to the master processor (10) via the bus subsystem (20).

Abstract: A method and system for reading data from a redundant array of computer data storage media elements without having to wait for a timeout in a drive in the redundant array. If the requested data is from a drive that is temporarily in a timeout, the requested data is reconstructed using other drives in the redundant array, and the reconstructed data is sent to the requester via an array controller. When the drive holding the originally requested data comes back from the timeout and responds to the original request by transmitting the original data, the array controller ignores the original data since the request has already been fulfilled by the reconstructed data.

Abstract: In a stacked system, errors detected at a user application interface of one or more host systems are associated with root cause errors at a stack level below a virtualization layer by detecting an error at a user application interface; identifying an associated root cause error at a lower stack level; creating an error trace entry for the error; associating an error log identifier with the error trace entry; making the combined error log identifier and the error trace entry into an error identifier that is unique within the plurality of host systems in said stacked system; and communicating the error identifier to any requester of a service at a user application interface of one or more host systems when the service must be failed because of the root cause error.

Abstract: A method and apparatus for detecting and preemptively ameliorating potential logic unit thrashing in a storage system having multiple I/O requesters is disclosed. In response to detecting that each of two requesters has usable access to both of the active-passive pair of controllers, one of the active-passive pair of controllers is selected to be designated as an active resource controller. In response to detecting that one of the two requesters has usable access to only one of the active-passive pair of controllers, only one of the active-passive pair of controllers is selected to be designated as an active resource controller. In response to detecting that each of the two requesters has usable access only to different ones of the active-passive pair of controllers, one of the active-passive pair of controllers is selected to be designated as an active resource controller.

Abstract: An apparatus for providing storage control in a network of storage controllers is disclosed. The apparatus includes an owner storage controller; an I/O performing component, an ownership assignment component, a lock manager and a messaging component. The ownership assignment component assigns ownership of metadata for data to an owner storage controller. The lock manager controls the locking of metadata during I/O. The messaging component passes messages among storage controllers to request metadata state, to grant locks, to request release of locks, and to signal lock release. The I/O is performed on data whose metadata is owned by an owner storage controller, subject to compliance with metadata lock protocols controlled by the owner storage controller, and any copy of the data held from time to time is maintained in a coherency relation with the data.

Abstract: A method and system for reading data from a redundant array of computer data storage media elements without having to wait for a timeout in a drive in the redundant array. If the requested data is from a drive that is temporarily in a timeout, the requested data is reconstructed using other drives in the redundant array, and the reconstructed data is sent to the requester via an array controller. When the drive holding the originally requested data comes back from the timeout and responds to the original request by transmitting the original data, the array controller ignores the original data since the request has already been fulfilled by the reconstructed data.

Abstract: Providing high availability cache coherency in a distributed cache environment for a storage cluster. An extent owner node in a cluster controls extent disk requests and the locking and demotion of extent data. An extent client node may access the data but does not control destaging or locking. The extent client waits for a lock state grant prior to completing any I/O request. An extent client is allowed to receive data from a host for a write request and later let the extent owner sort out the proper cache coherency order. An extent client is not required to request a lock state change from the extent owner when a read cache hit occurs in the extent client. When a read miss occurs in an extent client, the extent client can initiate a stage request from disk and request the owner for a lock state change at a later time.

Abstract: A method and apparatus for managing a loop network, the loop network (200) including at least one loop (206, 208), a plurality of devices (210) connected to the at least one loop (206, 208) via ports (211, 212), wherein at least two of the devices are initiators (207, 209). The method includes each initiator (207, 209) sending a frame to all other initiators (207, 209) in the loop network (200) identifying any ports (211, 212) which should not be used. Each initiator (207, 209) merges the information from all other initiators (207, 209) with its own information identifying any ports (211, 212) which should not be used resulting in all the initiators (207, 209) generating a single list of ports (211, 212) to be used which is consistent across all the initiators (207, 209). Each initiator (207, 209) applies an algorithm (300) to determine a common set of ports (211, 212) to be used by all the initiators (207, 209) and to balance port accesses across the loop network (200).

Abstract: In a data processing network including a plurality of array controllers connected for communication to each other and a plurality of data storage devices, a method for coordinating exclusive write access by the plurality of controllers to a shared data region on said plurality of data storage devices comprising the steps of: at a first array controller, broadcasting an exclusive access request to all other array controllers having access to the shared data region and storing a non-volatile record of the write operation in the first controller; and at each controller receiving the exclusive access request, storing a non-volatile record of the write operation prior to sending an exclusive access grant to the first controller.