Abstract: A method and system for storing data in portable storage devices. Specifically, the disclosed method and system provide a solution for the write-hole problem inflicting persistent storage, especially redundant array of independent disks (RAID) configured storage. The write-hole problem may arise from the occurrence of power failure during a write operation of data to RAID configured storage, subsequently resulting in disparities between the data and parity information thereof—the consistency there-between of which is critical to data reconstruction upon disk failure. To rectify these inconsistencies, a full-stripe (or full-block set) write is recommended, which the disclosed method and system implements through the use of, and re-mapping of relationships between, virtual, physical, and in-memory block sets.

Abstract: Example implementations relate to storing manifest portions in a metadata cache. An example includes receiving, by a storage controller, a read request associated with a first data unit. In response to receiving the read request, the storage controller stores a manifest portion in a metadata cache, the stored manifest portion comprising a plurality of records, the plurality of records including a first record associated with the first data unit. The storage controller determines storage information of the first data unit using pointer information included in the first record of the stored manifest portion, and replaces the pointer information in the first record with the determined storage information of the first data unit.

Abstract: A memory controller circuit is disclosed which is coupleable to a first memory circuit, such as DRAM, and includes: a first memory control circuit to read from or write to the first memory circuit; a second memory circuit, such as SRAM; a second memory control circuit adapted to read from the second memory circuit in response to a read request when the requested data is stored in the second memory circuit, and otherwise to transfer the read request to the first memory control circuit; predetermined atomic operations circuitry; and programmable atomic operations circuitry adapted to perform at least one programmable atomic operation. The second memory control circuit also transfers a received programmable atomic operation request to the programmable atomic operations circuitry and sets a hazard bit for a cache line of the second memory circuit.

Abstract: A drive cluster with RAID (D+P) protection groups initially has D+P=W drives and W partitions. One protection group is created in each partition of the original drive cluster. The original drive cluster is scaled by selecting RAID protection group members starting with a first partition of the new drive and characterized by decrementing drive numbers and incrementing partition numbers, relocating the selected RAID protection group members within partitions such that a selected RAID protection group member on partition X of an existing drive is relocated to partition X of the new drive, and creating a new protection group in vacated partitions and the first partition of the new drive. The scaled drive cluster is split into independent smaller drive clusters when there are at least 2*W drives.

Abstract: Partitions of drives are used to form a volume of a drive array. Each partition is associated with a trusted computing group (TCG) band. Each drive encrypts data stored on the partition with a key unique to the TCG band. The volume is formed using the partitions of the drives. In response to a band-based erasure being invoked on the volume, each drive of the plurality of drives overwrites the key of the TCG band associated with the partition and provides an erasure certificate attesting to the overwriting of the key. The erasure certifications from the drives are compiled into a consolidated erasure certification that attests to the erasure of the volume.

Abstract: A caching system including a first sub-cache and a second sub-cache in parallel with the first sub-cache, wherein the second sub-cache includes: line type bits configured to store an indication that a corresponding cache line of the second sub-cache is configured to store write-miss data, and an eviction controller configured to evict a cache line of the second sub-cache storing write-miss data based on an indication that the cache line has been fully written.

Abstract: A method for protecting data in a storage system is disclosed. In one embodiment, such a method includes detecting, by a first rack power controller, first battery-on status associated with a first uninterruptible power supply. The method further detects, by a second rack power controller, second battery-on status associated with a second uninterruptible power supply. The method communicates, from the first rack power controller to the second rack power controller, the first battery-on status. The method then triggers, by the second rack power controller, a dump of modified data from memory to more persistent storage upon detecting both the first battery-on status and the second battery-on status. A corresponding system and computer program product are also disclosed.

Abstract: A prefetcher, an operating method of the prefetcher, and a processor including the prefetcher are provided. The prefetcher includes a prefetch address generating circuit, an address tracking circuit, and an offset control circuit. The prefetch address generating circuit generates a prefetch address based on first prefetch information and an offset amount. The address tracking circuit stores the prefetch address and a plurality of historical prefetch addresses. When receiving an access address, the offset control circuit updates the offset amount based on second prefetch information, the access address, the prefetch address, and the historical prefetch addresses, and provides the prefetch address generating circuit with the updated offset amount.

Abstract: Distributed storage nodes having backup power supplies and write-back caching capabilities can be pooled for servicing write requests. For example, a management node of a distributed storage system can determine a subset of storage nodes, from among a group of storage nodes of the distributed storage system, coupled to backup power supplies based on status information received from the group of storage nodes. The status information can indicate whether a respective storage node of the group of storage nodes is coupled to a corresponding backup power supply. The management node can then generate a node pool that includes the subset of storage nodes with write-back caching enabled. The node pool can be configured to perform write-back caching in relation to a write request.

Abstract: An information handling system includes a non-volatile memory (NVRAM) and a processor. The NVRAM stores a plurality of NVRAM variables and a basic input/output system (BIOS) of the information handling system. The BIOS includes system BIOS variable services. The processor executes the system BIOS variable services. While executing the system BIOS variable services, the processor determines whether a holding area of a first NVRAM variable of the NVRAM variables is completely used. In response to the storage being completely used, the processor calculates a new size of the holding area based on metadata of the first NVRAM variable, and creates a new storage area for the first NVRAM variable. The size of a second holding area of the new storage area equals the new size.

Abstract: Provided is a memory management system that efficiently protects data in a cache memory adopting a virtual address cache method. The memory management system includes a cache memory that temporarily stores data for which memory access is requested by a processor core; a state storage unit that stores a security state communicated simultaneously with the memory access request from the processor core; and a memory management unit that manages access to a main memory. In a case where there is a change in the security state when memory access is requested by the processor core, a cache flush is performed for a cache line that hits the request.

Abstract: Systems and methods for adding backups to de-duplicated storage and for removing backups from de-duplicated storage are disclosed. Backups can be added to the de-duplicated storage with minimal reference count updates. Backup data sets can be removed without garbage collection processes being performed on the de-duplicated storage.

Abstract: A first host device establishes connectivity to a logical storage device of a storage system. The first host device obtains from the storage system host connectivity information identifying at least a second host device that has also established connectivity to the logical storage device, caches one or more extents of the logical storage device in a memory of the first host device, and maintains local cache metadata in the first host device regarding the one or more extents of the logical storage device cached in the memory of the first host device. In conjunction with processing of a write operation of the first host device involving at least one of the one or more cached extents of the logical storage device, the first host device invalidates corresponding entries in the local cache metadata of the first host device and in local cache metadata maintained in the second host device.

Abstract: A storage system has a volatile memory, a non-volatile memory, and a controller. The controller of the storage system can implement various mechanisms for improving random read performance. These mechanisms include improved read prediction cache management, using a pattern length for read prediction, and a time-based enhancement for read prediction. Each of these mechanisms can be used alone on in combination with some or all of the other mechanisms.

Abstract: Exemplary methods, apparatuses, and systems include a file system process determining to a flush a node in a first tree. The first node includes a buffer structured as a second tree. The file system process generates an input/output instruction to load the buffer from a first memory to a second memory. The second tree is stored in two more non-contiguous locations in the first memory and the input/output operation includes a read operation corresponding to each of the two or more non-contiguous locations. The file system process causes the input/output instruction to be executed concurrently on the first memory.

Abstract: Snapshot metadata may include a plurality of pages of nodes, including active nodes and free nodes. It may be determined whether a snapshot metadata object is eligible for de-allocation, for example, of one or more of the pages of its snapshot metadata nodes. This determination may be based on a number of free nodes in the snapshot metadata object, for example, in relation to the quantity of nodes that are included in a snapshot metadata page. This determination may be made based on previous usage of the nodes allocated to the snapshot metadata object, for example, the number of active nodes relative to a total size of the snapshot metadata object. For example, a maximum extent of active nodes during one or more periods may be compared to a current extent of active nodes to determine whether the snapshot metadata object is eligible.

Abstract: Techniques for transitioning between thread-confined memory segments and shared memory segments are disclosed. The system may instantiate a confined memory segment view. The confined memory segment view confines access to a memory segment to a particular thread. The system may further receive a request to change access permissions for the confined memory segment to allow access by a first set of one or more threads. Responsive to receiving the request to change access permissions for the confined memory segment, the system may instantiate a new memory segment view, wherein the new memory segment view permits access to the memory segment by the first set of one or more threads. The system may also copy metadata from the confined memory segment view to the new memory segment view. The system may de-allocate the memory segment in response to determining that there are no memory segment views associated with the memory segment.

Abstract: A solid state drive is provided for improving write consistency when replicating data. The solid state drive includes a plurality of memory die, a host interface, and a memory controller. The memory controller is configured to receive write commands from a host server, update a list of write commands and a status of execution of the write commands to include the write commands from the host server, and write data payloads associated with the write commands to at least a first memory die and a second memory die of the plurality of memory die. In response to the data payloads being written, the memory controller is configured to update the status of execution for the data payloads in the list of write commands and receive an indication that the data payloads have been written to an external device.

Abstract: In some examples, a control device includes a controller to receive, from a requester device that is separate from the control device, a request to access a first memory region of a memory. The controller is to determine, based on occurrence of a systems initialization event and according to permissions information that identifies access permissions for respective memory regions of the memory, whether access of content in the first memory region is allowed.

Abstract: A memory system with at least one namespace includes a memory device and a controller. The memory device includes a plurality of single-level cell (SLC) buffers and a plurality of memory blocks, wherein each memory block includes a plurality of memory cells, each memory cell storing multi-bit data, and is allocated for a respective one of a plurality of zones, wherein each of the at least one namespace is divided by at least some of the plurality of zones. The controller is configured to receive a program request related to at least one application program executed by a host, to determine at least one zone designated by the at least one application program as an open state, and to control the memory device to perform a program operation on at least one memory block allocated for an open state zone.