An example method includes identifying, based on a request to write data to a volume provided by storage nodes in a storage node cluster, a cluster hierarchy associated with the storage node cluster, the cluster hierarchy identifying storage characteristics of the storage nodes, wherein the cluster hierarchy is based on data characterizing performance and storage capabilities of the storage nodes; based on the request, selecting a storage node for storing the data in the volume, the selecting being based, at least in part, on the storage characteristics identified by the cluster hierarchy and on one or more data distribution parameters associated with the storage volume; and transmitting the data to the storage node.

A total estimated occupancy value of a first data on a first data block of a plurality of data blocks is determined. To determine the total estimated occupancy value of the first data block, a total block power-on-time (POT) value of the first data block is determined. Then, a scaling factor is applied to the total block POT value to determine the total estimated occupancy value of the first data block. Whether the total estimated occupancy value of the first data block satisfies a threshold criterion is determined. Responsive to determining that the total estimated occupancy value of the first data block satisfies the threshold criterion, data stored at the first data block is relocated to a second data block of the plurality of data blocks.

A multiple storage node system including a first and second node is provided. The first node includes a first baseboard management controller (BMC), a first flash ROM configured to store a first flash image, and a first switch device configured to connect the first BMC to the first flash ROM. The second node includes an exact configuration of the first node. The first BMC is connected to the second switch device, and the second flash image is the same as the first flash.

Zero copy message reception for guests is disclosed. For example, a host has a memory, a device with access to device memory addresses, a processor, and a supervisor. An application with access to application memory addresses (AMA) executes on the host. An AMA is mapped to a page table entry (PTE). The application shares access to a first page of memory addressed by the AMA with the device to store data received by the device for the first application, where the first page is mapped as a device memory address of the plurality of device memory addresses. The application later sends a request to disconnect from the device. The supervisor is configured to copy contents of the first page to a second page in the memory after receiving the request to disconnect, and then update the PTE to address the second page instead of the first page.

A memory device includes several normal memory circuits and a redundant memory circuit is disclosed. The several normal memory circuits include several normal memory arrays. The redundant memory circuit includes a redundant memory array. The several normal memory arrays share the redundant memory array. When a first normal memory cell of a first normal memory array of the several normal memory arrays is destructed, a first redundant memory cell of the redundant memory array replaces the first normal memory cell. When a second normal memory cell of a second normal memory array of the several normal memory arrays is destructed, a second redundant memory cell of the redundant memory array replaces the second normal memory cell.

Local management of data stream throttling in data movement operations, such as secondary-copy operations in a storage management system, is disclosed. A local throttling manager may interoperate with co-resident data agents and/or a media agent executing on any given local computing device, whether a client computing device or a secondary storage computing device. The local throttling manager may allocate and manage the available bandwidth for various jobs and their constituent data streams—across the data agents and/or media agent. Bandwidth is allocated and re-allocated to data streams used by ongoing jobs, in response to new jobs starting and old jobs completing, without having to pause and restart ongoing jobs to accommodate bandwidth adjustments. The illustrative embodiment also provides local users with a measure of control over data streams—to suspend, pause, and/or resume them—independently from the centralized storage manager that manages the overall storage system.

The method and apparatus that are applied to a machine learning system which includes at least one parameter collection group and at least one parameter delivery group. Each parameter collection group is corresponding to at least one parameter delivery group. The method includes: when any parameter collection group meets an intra-group combination condition, combining model parameters of M nodes in the parameter collection group to obtain a first model parameter of the parameter collection group, where a smallest quantity s of combination nodes in the parameter collection group≤M≤a total quantity of nodes included in the parameter collection group; and sending the first model parameter of the parameter collection group to N nodes in a parameter delivery group corresponding to the parameter collection group, where 1≤N≤a total quantity of nodes included in the parameter delivery group corresponding to the parameter collection group.

Undo logging for persistent memory transactions may permit concurrent transactions to write to the same persistent object. After an undo log record has been written, a single persist barrier may be issued. The tail pointer of the undo log may be updated after the persist barrier, and without another persist barrier, so the tail update may be persisted when the next log record is written and persisted. Undo logging for persistent memory transactions may rely on inferring the tail of an undo log after a failure rather than relying on a guaranteed correct tail pointer based on persisting the tail after every append. Additionally, transaction version numbers and checksum information may be stored to the undo log enabling failure recovery.

The present disclosure includes apparatuses and methods related to a command bus in memory. A memory module may be equipped with multiple memory media types that are responsive to perform various operations in response to a common command. The operations may be carried out during the same clock cycle in response to the command. An example apparatus can include a first number of memory devices coupled to a host via a first number of ports and a second number of memory devices each coupled to the first number of memory devices via a second number of ports, wherein the second number of memory devices each include a controller, and wherein the first number of memory devices and the second number of memory devices can receive a command from the host to perform the various (e.g., the same or different) operations, sometime concurrently.

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.