摘要:
A system and method for receiving and responding to issued commands in a storage enclosure is disclosed in which each storage enclosure processor of the storage enclosure is coupled to each expander of the storage enclosure. Each storage enclosure processor receives each interrupt and command issued by another expander of the storage enclosure. In the event of a failure of one of the storage enclosure processors, any interrupts or commands issued by the expander associated with the failed storage enclosure processor will be handled by the operational storage enclosure processor. In this configuration, the storage enclosure processor can also arbitrate or determine the storage enclosure processor that will handle each interrupt and any associated command.
摘要:
An architecture and a method for the automated configuration of a storage enclosure are disclosed. The disclosed storage enclosure includes an internal communications link between the storage controllers of the storage enclosure. The storage enclosure is operable to determine if the storage enclosure is a terminal storage enclosure or if the expansion ports of the storage enclosure are externally cabled. If the storage enclosure is the terminal storage enclosure, the internal communications link is enabled, providing a communications pathway between the two controllers that is internal to the storage enclosure.
摘要:
A method and apparatus for performing a hold operation while keeping the data in place as the data is in a hold state. Such a method and apparatus substantially eliminates the need for a copy operation and thus provides advantages cost and management savings. The method and apparatus define a hold delete operation along with hold life points in a CAS system.
摘要:
A port width module operating in association with SAS controllers determines the operational state of communication across an SAS cable by comparing the number of links at opposing controllers that are capable of supporting communication with the number of links in DWord synchronization. For example, an information handling system capable of communication of four links at an SAS connector communicates in a normal state across two links where the opposing device is a switch that supports only two links at an SAS connector. An LED at the information handling system indicates normal operations with two links in DWord synchronization by receiving the port width of the opposing device, retrieving its own port width, and comparing the port widths with the number of links in DWord synchronization.
摘要:
An integrated cable tester detects cable faults by coupling a single cable to host and expander connectors of an interface module and determining whether each of plural ports of the connectors has an associated PHY Ready signal. An LED interfaced with the cable tester illuminates to indicate a normal cable and fails to illuminate if the cable tests faulty. In one embodiment a module tester determines whether the module has degraded performance when a single cable is detected as coupled to the host and expander connectors of the module. The module tester clears the interface module's error log and initiates a reset of communication between the host and expander ports. Upon completion of the reset, such as detection of all PHY Ready signals for the plural ports, the module tester reads the error log and indicates errors as degrading the performance of the interface module.
摘要:
A Write-Once Read-Many (WORM) memory controller receives data from a processing system that is addressed to a location in a storage device, stores the data and a tag at the location, receives second data from the processing system that is addressed to the location, determines that the location includes the tag, and prevents the second data from being stored at the location based upon the presence of the tag. A WORM memory device sends a reply to a controller in response to an initialization command. The reply includes an address number that corresponds with the storage capacity of the WORM memory device. The WORM memory device sends another reply to another controller in response to another initialization command. The initialization commands are different from each other. The other initialization reply includes an address number of zero.
摘要:
Systems and methods for redundant object storage are disclosed. A method may include storing at least two copies of each of a plurality of objects among a plurality of nodes communicatively coupled to one another in order to provide redundancy of each of the plurality of objects in the event of a fault of one of the plurality of nodes. The method may also include monitoring access to each object to determine a frequency of access for each object. The method may additionally include redistributing one or more of the copies of the objects such that at least one particular node of the plurality of nodes includes copies of only objects accessed at a frequency below a predetermined frequency threshold based on the determined frequency of access for each object. The method may further include placing the at least one particular node in a reduced-power mode.
摘要:
An SAS domain map is automatically generated at an SAS concentrator switch by a virtual mapping device that presents itself as a target for discovery by SAS devices interfaced with the concentrator, such as information handling systems and storage devices. During the SAS protocol discovery process, the virtual mapping device generates the SAS domain map by acquiring the device name and the device port for each concentrator port that interfaces with a device. A management application running on the concentrator applies the SAS domain map to provide network functions, such as zoning or diagnostics.
摘要:
An object storage system, such as a content addressed storage system, manages replication of objects across network locations to balance storage space and data security. Network locations set a policy of replicating each object at the object's primary network location and a secondary network location. The secondary network location creates a first replica of the object and a virtual unique identifier representing a second replica of the object at the secondary network location. Creation of the second replica is suppressed unless the first replica becomes invalid so that storage space is conserved without substantially increasing the risk of loss of information represented by the object.
摘要:
Systems and methods for redundant object storage are disclosed. A method may include storing at least two copies of each of a plurality of objects among a plurality of nodes communicatively coupled to one another in order to provide redundancy of each of the plurality of objects in the event of a fault of one of the plurality of nodes. The method may also include monitoring access to each object to determine a frequency of access for each object. The method may additionally include redistributing one or more of the copies of the objects such that at least one particular node of the plurality of nodes includes copies of only objects accessed at a frequency below a predetermined frequency threshold based on the determined frequency of access for each object. The method may further include placing the at least one particular node in a reduced-power mode.