Abstract: Systems, apparatuses, and methods for maintaining a region-based cache directory are disclosed. A system includes multiple processing nodes, with each processing node including a cache subsystem. The system also includes a cache directory to help manage cache coherency among the different cache subsystems of the system. In order to reduce the number of entries in the cache directory, the cache directory tracks coherency on a region basis rather than on a cache line basis, wherein a region includes multiple cache lines. Accordingly, the system includes a region-based cache directory to track regions which have at least one cache line cached in any cache subsystem in the system. The cache directory includes a reference count in each entry to track the aggregate number of cache lines that are cached per region. If a reference count of a given entry goes to zero, the cache directory reclaims the given entry.

Abstract: A data storage system includes a logical space having logical block addresses (LBAs) divided into non-overlapping LBA ranges, and a physical space having pairs of physical bands. The system also includes a map in which first successive alternate LBAs of each different one of the non-overlapping LBA ranges are mapped to successive adjacent physical blocks of a first physical band of each different pair of the pairs of physical bands, and second successive alternate LBAs of each different one of the non-overlapping LBA ranges are mapped to successive adjacent physical blocks of a second physical band of each different pair of the pairs of physical bands. A controller employs the map to concurrently read data from a first physical block of the first physical band of one pair of physical bands and from a first physical block of the second physical band of the same pair of physical bands.

Abstract: Technique manage data within computerized memory. The techniques involve, in response to receiving host data in a write cache, updating a data order log that holds order information indicating a temporal order for the host data. The temporal order initially is the order that the host data was received in the write cache. The techniques further involve transferring the host data from the write cache to secondary storage. The techniques further involve, after the host data is transferred from the write cache to secondary storage, providing a garbage collection service that consolidates the host data within the secondary storage in accordance with the data order log that holds the order information indicating the temporal order for the host data. With the temporal order of the host data generally preserved, data access operations may enjoy various optimizations such as improved prefetching, more sequential reads, improved auto-tiering, and so on.

Abstract: Systems, methods and apparatuses to accelerate accessing of borrowed memory over network connection are described. For example, a memory management unit (MMU) of a computing device can be configured to be connected both to the random access memory over a memory bus and to a computer network via a communication device. The computing device can borrow an amount of memory from a remote device over a network connection using the communication device; and applications running in the computing device can use virtual memory addresses mapped to the borrowed memory. When a virtual address mapped to the borrowed memory is used, the MMU translates the virtual address into a physical address and instruct the communication device to access the borrowed memory.

Abstract: Techniques for protecting information may include: exposing a logical device of a data storage system to a host, wherein the logical device has an attribute identifying the logical device as a stealth device having accessibility controlled by the data storage system based on commands issued over a control path, wherein the logical device has a mode indicating whether the logical device is accessible to the host; sending, from the host to the data storage system, a write command that writes first data on the logical device when the mode indicates the logical device is accessible to the host; and subsequent to said sending, issuing a command over the control path to the data storage system, wherein the command sets the mode of the logical device to inaccessible indicating the logical device is not accessible to the host.

Abstract: Examples relate to an approximative memory deduplication method, a controller apparatus or controller device for a memory or storage controller, a memory or storage controller, a computer system and to a computer program. The approximative memory deduplication method comprises determining a hash value of a data block. The hash value is based on a user-defined approximative hashing function. The approximative memory deduplication method comprises storing a quantized version of the data block based on the hash value using a memory or storage device of the computer system.

Abstract: A data storage device is provided. The data storage device a buffer configured to store a mapping table comprising physical block addresses (PBAs) corresponding to logical block addresses (LBAs), a non-volatile memory configured to store data; and a controller configured to control the buffer and the non-volatile. The controller is configured to read data stored at a first PBA of the non-volatile memory corresponding to a first LBA by referring to the mapping table when receiving a command to read data corresponding to the first LBA from outside of the data storage device, and, when a second LBA included in the data read from the non-volatile memory is different from the first LBA, retry reading, from the non-volatile memory, the data corresponding to the first LBA.

Abstract: A system including a storage drive and a semiconductor apparatus coupled to the storage drive, is provided. The semiconductor apparatus may include one or more substrates and logic coupled to the one or more substrates, the logic coupled to the one or more substrates to: initiate managing resources of the storage drive and, if the storage drive loses capacity, determine an amount of capacity loss, create a reserved file that is associated with logical memory space in a file system, based on the amount of the capacity loss, and erase at least a portion of the reserved file so that logical memory space associated with an un-erased portion of the reserved file is usable by the storage drive.

Abstract: Garbage collection (GC) to reclaim memory in computing systems sometimes suspends applications, web services, and other mutator programs in order to scan their execution stacks, which reduces their responsiveness. But suspension times are decreased by behavior-driven stack scan optimization (BDSSO) functionality that increases the concurrency of mutator execution with GC stack scanning. BDSSO obtains execution stack frame occurrence data, determines frame execution likelihoods, selects a stack scan depth based on the likelihoods, and installs a scan return barrier at the selected depth. Then the GC scans the execution stack below the barrier while the mutator runs, thus increasing concurrency and improving mutator responsiveness. Selected barrier locations vary according to actual stack activity to provide optimized concurrency instead of using an inflexible approach to barrier placement. Existing profiler samples or virtual machine interfaces can be reused by BDSSO.

Abstract: A system and method for allocating memory to a heterogeneous address space includes identifying, by an operating system, at least one superset feature from an application configured to be executed on a host device. The address space associated with the application includes a plurality of supersets, and wherein the operating system allocates the memory to each of the plurality of supersets from a non-volatile memory or a volatile memory based upon the at least one superset feature.

Abstract: Methods and apparatus for profile-guided preloading for virtualized resources are described. A block-level storage volume whose contents are to be populated via data transfers from a repository service is programmatically attached to a compute instance. An indication of data transfers from the repository to a block storage service implementing the volume is obtained, corresponding to a particular phase of program execution at the compute instance. A storage profile is generated, based at least in part on the indication of data transfers. The storage profile is subsequently used to pre-load data from the repository service on behalf of other compute instances.

Abstract: An aspect of volume migration includes establishing a connection between hosts and a non-volatile memory over fabric (NVMeoF) controller for appliances, and creating and exposing a volume as a namespace to hosts. The namespace forms part of an asymmetric namespace access (ANA) group in which optimized paths are exposed to the hosts. Upon receiving a request to migrate the volume from a source appliance to a destination appliance, an aspect further includes creating a namespace on the destination, and configuring the ANA group of the namespace as inaccessible. Through the connection, an aspect includes initiating an asynchronous event notification (AEN) to the host to connect ports on the destination and, upon connecting to the ports, discovering by the hosts on the destination, the namespace created, verifying the hosts have connected to the ports and have discovered the namespace, and migrating data to the destination and performing input/output cutover.

Abstract: A method for limiting I/O access in shared storage systems is disclosed. In one embodiment, such a method includes establishing, for a volume, a list of address spaces that are authorized to access the volume. The method further receives an I/O request to access the volume and determines whether the I/O request originates from one of the address spaces identified in the list. If the I/O request originates from one of the address spaces in the list, the method passes the I/O request to the volume. If, on the other hand, the I/O request does not originate from one of the address spaces in the list, the method blocks the I/O request. A corresponding system and computer program product are also disclosed.

Abstract: Systems, apparatus and methods are provided to combine multiple channels in a multi-channel memory controller to save area and reduce power and cost. An apparatus may comprise a first memory controller configured to access a first channel using a first protocol, a second memory controller configured to access a second channel using a second protocol that is different from the first protocol and a physical interface coupled to the first memory controller and a second memory controller. The physical interface may comprise a set of pins for an address and command bus shared by the first memory controller and the second memory controller for the first memory controller and the second memory controller to send address or command to respective channels by time division multiplexing.

Abstract: Techniques are provided for providing a fully active and non-replicated block storage solution in a clustered filesystem that implements cache coherency. In a clustered filesystem where one or more data blocks are stored in a respective cache of each host node of a plurality of host nodes, a request is received at a host node of the plurality of host nodes from a client device to write the one or more data blocks to a shared storage device. In response to the request, the one or more data blocks are stored in the cache of the host node and a particular notification is sent to another host node of the plurality of host nodes that the one or more data blocks have been written to the shared storage device. In response to receiving the notification, the other host node invalidates a cached copy of the one or more data blocks in the respective cache of the other host node.

Abstract: Preparing a key block in a memory system. Various methods include: selecting a candidate key block of memory; checking a quality of the candidate key block using a word line of the candidate key block; altering operating parameters of the candidate key memory block; and registering the candidate key memory block as the key block. Where altering the operating parameters includes replacing a first set of parameters associated with the first memory block with a second set of parameters, where the first set of parameters includes a first erase parameter, a first program parameter, and a first read parameter, where the memory block operating in a normal block mode is accessed using the first set of parameters, and the second set of parameters includes a second erase parameter, a second program parameter, and a second read parameter, where the first memory block is accessed using the second set of parameters.

Abstract: A spatial disturbance that occurs when an access is concentrated in a specific memory area in a volatile semiconductor memory like DRAM is properly solved by a memory controller. The memory controller includes a concentration access detection part generating a concentration access detection signal when an address for accessing a specific memory area among memory areas of volatile semiconductor memory is concentratedly received. In the case that the concentration access detection signal is generated, the memory controller includes a controller for easing or preventing corruption of data which memory cells of the specific memory area and/or memory cells of memory areas adjacent to the specific memory area hold.

Abstract: A non-volatile storage system, configured to use a protocol that supports predictable latency, including: a memory array storing a data in a block of memory; a controller coupled to the memory array, where the controller is configured to: in response to determining that predictable latency is enabled, operate the storage system using a first mode for a duration of time, where during the first mode, the storage system operates such that a read latency is below a read latency threshold; and after the duration of time, operate, the storage system using a second mode for a second duration of time, where during the second mode: the storage system performs a management operation based on a second set of thresholds that are different from a first set of threshold used during the first mode.

Abstract: In a data storage system with distributed data storage units, initialization-less parity can be practiced with a storage controller connected to a storage memory and multiple data storage units. Data locations of data storage devices of the respective data storage units can be arranged as distributed data storage groups as directed by the storage controller prior to receiving a write request to a distributed data storage group. Unwritten data locations of the distributed data storage group are identified by consulting the storage memory and each unwritten data storage location may be assumed to have a zero value when computing parity data for the distributed data storage group.

Abstract: A method for mapping LUNs (logical unit numbers) in storage memory, performed by a storage system, is provided. The method includes determining a set of LUNs in the storage memory and generating a mapping from a logical address space to all of the LUNs in the set, based on the determining, so that each logical address in the logical address space maps to one LUN in the set. The method includes accessing one or more of the LUNs in accordance with the mapping.