Abstract: A data storage system can have one or more hosts connected to a data storage subsystem with the host having a host processor and the data storage subsystem having a controller. Write back data generated at the host triggers the host processor to allocate a cache location in the data storage subsystem where the generated data is subsequently stored. The generated write back data is written in a non-volatile destination address as directed by the controller prior to waiting for a secondary event with the generated data stored in both the cache location and the non-volatile destination address. Detection of the secondary event prompts the controller to signal the host processor that the cache location is free for new data.

Abstract: A data storage system can have one or more hosts connected to a data storage subsystem with the host having a host processor and the data storage subsystem having a controller. Write back data generated at the host triggers the host processor to allocate a cache location in the data storage subsystem where the generated data is subsequently stored. The generated write back data is written in a non-volatile destination address as directed by the controller prior to waiting for a secondary event with the generated data stored in both the cache location and the non-volatile destination address. Detection of the secondary event prompts the controller to signal the host processor that the cache location is free for new data.

Abstract: Methods and structure for selectively powering a storage device over a data network. An exemplary system includes a power module configured to detect power from a host system via a network port. The system also includes an input/output controller configured to receive power derived from the network port of the power module, and in response, to identify a disk drive for a read/write operation based on information from the host system. With power derived from the network port, the power module is further configured to supply power to an expander that connects the disk drive to the input/output controller, and to supply power to the disk drive to perform the read/write operation.

Abstract: Apparatus and systems for improved access to storage devices from the sides of sleds mounted in storage enclosures. Embodiments provide apparatus and systems for a sled in a storage enclosure that provides access to storage devices on either side of the sled when the sled is slid forward out of its enclosure. Multiple sleds may be enclosed within a single enclosure to permit access to a portion of the storage devices in the enclosure hence reducing the problems of instability of the rack if the enclosure is mounted near the top of the rack.

Abstract: Methods and structure for selectively powering a storage device over a data network. An exemplary system includes a power module configured to detect power from a host system via a network port. The system also includes an input/output controller configured to receive power derived from the network port of the power module, and in response, to identify a disk drive for a read/write operation based on information from the host system. With power derived from the network port, the power module is further configured to supply power to an expander that connects the disk drive to the input/output controller, and to supply power to the disk drive to perform the read/write operation.

Abstract: Methods and structure for powering storage devices over a SAS interface. An exemplary system includes a storage device that includes a receptacle configured to receive a plug that communicates Serial Attached Small Computer Systems Interface compliant signals from a host system. The storage device also includes a microcontroller configured to detect a first power level on a power contact of the receptacle. The storage device further includes a circuit board configured to power up to the first power level via the microcontroller, to receive an inter-integrated circuit communication that indicates a second power level is available, and to power up to the second power level to sufficiently power a storage appliance.

Abstract: Storage systems configured for improved N-way connectivity among all of a plurality of storage controllers and all of a plurality of storage devices in the system. All controllers of the storage system are coupled through a switched fabric communication medium to all of the storage devices of the storage system. Thus, the back-end interface of each storage controller of the storage system is used for all communications with any of the storage devices as well as for any communications among the controllers to coordinate the N-way distribution of stored data in a declustered RAID storage environment. This use of the back-end channel for all storage controller to storage device N-way connectivity as well as controller to controller N-way connectivity eliminates the need for a dedicated inter-controller interface for such N-way connectivity and eliminates the over-utilization of a front-end (e.g., network) communication path for providing N-way connectivity in the storage system.

Abstract: Methods and apparatus for improved calculation of redundancy information in RAID storage controllers. Features and aspects hereof provide for a firmware/software element (FPE) for generating redundancy information in combination with a custom logic circuit (HPE) designed to generate redundancy information. A scheduler element operable on a processor of a storage controller along with the FPE determines which of the FPE and HPE is best suited to rapidly complete a new redundancy computation operation and activates or queues the new operation for performance by the selected component.

Abstract: Systems and associated methods for flexible scalability of storage systems. In one aspect, a storage controller may include an interface to a fabric adapted to permit each storage controller coupled to the fabric to directly access memory mapped components of all other storage controllers coupled to the fabric. The CPU and other master device circuits within a storage controller may directly address memory an I/O devices directly coupled thereto within the same storage controller and may use RDMA features to directly address memory an I/O devices of other storage controllers through the fabric interface.

Abstract: Systems and associated methods for flexible scalability of storage systems. In one aspect, a storage controller may include an interface to a fabric adapted to permit each storage controller coupled to the fabric to directly access memory mapped components of all other storage controllers coupled to the fabric. The CPU and other master device circuits within a storage controller may directly address memory an I/O devices directly coupled thereto within the same storage controller and may use RDMA features to directly address memory an I/O devices of other storage controllers through the fabric interface.

Abstract: Methods and structure for an improved solid-state drive (SSD) for use in caching applications. An improved SSD comprises both volatile and non-volatile memory. The volatile memory provides improved performance as compared to present SSDs for use in caching application. The improved SSD senses impending failure of external power applied to the SSD and, while adequate power remains, copies cached data from the volatile memory to the non-volatile memory to retain the data through the power loss. In some embodiments, a local power source may be present to assure sufficient time for the SSD to save cached data in the non-volatile memory. Since the volatile memory (e.g., DRAM) is used for the primary caching function and the non-volatile memory is rarely used, performance, reliability and cost goals are achieved for write cache applications.

Abstract: Methods and apparatus for improved calculation of redundancy information in RAID storage controllers. Features and aspects hereof provide for a firmware/software element (FPE) for generating redundancy information in combination with a custom logic circuit (HPE) designed to generate redundancy information. A scheduler element operable on a processor of a storage controller along with the FPE determines which of the FPE and HPE is best suited to rapidly complete a new redundancy computation operation and activates or queues the new operation for performance by the selected component.

Abstract: Apparatus and systems for improved access to storage devices from the sides of sleds mounted in storage enclosures. Embodiments provide apparatus and systems for a sled in a storage enclosure that provides access to storage devices on either side of the sled when the sled is slid forward out of its enclosure. Multiple sleds may be enclosed within a single enclosure to permit access to a portion of the storage devices in the enclosure hence reducing the problems of instability of the rack if the enclosure is mounted near the top of the rack.

Abstract: Storage systems configured for improved N-way connectivity among all of a plurality of storage controllers and all of a plurality of storage devices in the system. All controllers of the storage system are coupled through a switched fabric communication medium to all of the storage devices of the storage system. Thus, the back-end interface of each storage controller of the storage system is used for all communications with any of the storage devices as well as for any communications among the controllers to coordinate the N-way distribution of stored data in a declustered RAID storage environment. This use of the back-end channel for all storage controller to storage device N-way connectivity as well as controller to controller N-way connectivity eliminates the need for a dedicated inter-controller interface for such N-way connectivity and eliminates the over-utilization of a front-end (e.g., network) communication path for providing N-way connectivity in the storage system.

Abstract: Methods and systems for assuring persistence of battery backed cache memory in a storage system comprising multiple virtual machines. In one exemplary embodiment, an additional process is added to the storage controller that senses the loss of power and copies the entire content of the cache memory including portions used by each of the multiple virtual machines to a nonvolatile persistent storage that does not rely on the battery capacity of the storage system. In another exemplary embodiment, the additional process calls a plug-in procedure associated with each of the virtual machines to permit the virtual machine to assure that the content of its portion of the cache memory is consistent before the additional process copies the cache memory to nonvolatile memory. The additional process may be integrated with the hypervisor or may be operable as a separate process in yet another virtual machine.

Abstract: A method for offloading a cache memory is disclosed. The method generally includes the steps of (A) reading all of a plurality of cache lines from the cache memory in response to an assertion of a signal to offload of the cache memory, (B) generating a plurality of blocks by dividing the cache lines in accordance with a RAID configuration and (C) writing the blocks among a plurality of nonvolatile memories in the RAID configuration, wherein each of the nonvolatile memories has a write bandwidth less than a read bandwidth of the cache memory.

Abstract: Apparatus, systems and methods for clustering multiple PCI Express hierarchies to enable access of components in different hierarchies. Each PCI Express hierarchy includes a root device as well as a cluster port for coupling each hierarchy to each other hierarchy of a physical cluster through a PCI Express switched fabric. Memory addresses and transaction IDs are re-mapped by the cluster port of each hierarchy to partition the PCI Express space of the system of multiple hierarchies. A first portion of the partitioned space is reserved for access to local components within a hierarchy. A second portion of the space is used to access remote components of other hierarchies from within a different first hierarchy. The address and transaction ID values exchanged in such remote transactions are re-mapped and used by the cluster port of each hierarchy to route exchanges between hierarchies using standard PCIe root devices, endpoint devices, and switches.

Abstract: Snapshot sets comprising snapshot criteria are created, maintained, and executed via a command set designed to facilitate managing and conducting snapshots on data distributed across multiple volumes. Snapshot criteria such as source volume, target volume, redundancy level, copy mode, and the like are added as desired to a snapshot set. Upon invocation of an execute command, the fast replication operations defined by the snapshot set are executed, providing a logically atomic data replication utility potentially involving multiple sources and targets. In one embodiment auto selection of a target may be specified, thus facilitating over-subscription of target resources. The present invention reduces the complexity of archiving data—particularly data distributed across multiple volumes such as data associated with database applications and the like.

Abstract: Apparatus, systems and methods for clustering multiple PCI Express hierarchies to enable access of components in different hierarchies. Each PCI Express hierarchy includes a root device as well as a cluster port for coupling each hierarchy to each other hierarchy of a physical cluster through a PCI Express switched fabric. Memory addresses and transaction IDs are re-mapped by the cluster port of each hierarchy to partition the PCI Express space of the system of multiple hierarchies. A first portion of the partitioned space is reserved for access to local components within a hierarchy. A second portion of the space is used to access remote components of other hierarchies from within a different first hierarchy. The address and transaction ID values exchanged in such remote transactions are re-mapped and used by the cluster port of each hierarchy to route exchanges between hierarchies using standard PCIe root devices, endpoint devices, and switches.

Abstract: A method for offloading a cache memory is disclosed. The method generally includes the steps of (A) reading all of a plurality of cache lines from the cache memory in response to an assertion of a signal to offload of the cache memory, (B) generating a plurality of blocks by dividing the cache lines in accordance with a RAID configuration and (C) writing the blocks among a plurality of nonvolatile memories in the RAID configuration, wherein each of the nonvolatile memories has a write bandwidth less than a read bandwidth of the cache memory.