Abstract: Embodiments provide a method, a system, and a computer program product for performing copy operations of one or more data units in a hierarchical storage management (HSM) system. The HSM system includes an upper layer and a lower layer. The upper layer includes multiple storage nodes having a grid configuration. The method comprises scheduling a copy operations of multiple data units each of which is stored in at least one of the multiple storage nodes such that loads on the copy operations are distributed among the multiple storage nodes in which the multiple data units are stored and copying the multiple data units to the lower layer in accordance with the scheduling.

Abstract: A set of data storage values is received. It is determined that a data storage value in the set will not fit in an available memory segment including variable data column widths based at least in part on data sizes specified in a plurality of segment layout maps. A memory segment is selected for which a column width of a column will be expanded. A column width of the selected memory segment is expanded. A segment layout map corresponding to the selected memory segment is updated. The set of data storage values is stored in the selected memory segment.

Abstract: The present disclosure describes technologies and techniques for use by a data storage controller—such as a controller for use with a NAND or other non-volatile memory (NVM)—to store crash-dump information in a boot partition following a system crash within the data storage controller. Within illustrative examples described herein, the boot partition may be read by a host device without the host first re-installing valid firmware into the data storage controller following the system crash. In the illustrative examples, the data storage controller is configured for use with versions of Peripheral Component Interconnect (PCI) Express—Non-Volatile Memory express (NVMe) that provide support for boot partitions in the NVM. The illustrative examples additionally describe virtual boot partitions in random access memory (RAM) for storing crash-dump information if the NAND has been corrupted, where the crash-dump information is retrieved from the RAM without power-cycling the RAM.

Abstract: A method and apparatus for detecting kernel data structure tampering are disclosed. In the method and apparatus, a memory region of a computer system is scanned for one or more characteristics of a kernel data structure of an operating system kernel. It is then determined, based at least in part on identifying whether the one or more characteristics are found in the memory region, whether the kernel data structure is stored in the memory region of the computer system for tampering with the kernel data structure.

Abstract: A computer-implemented method, computer program product and computing system for batch processing computing elements on a computing system are provided. The computer-implemented method, computer program product and computing system determine a configuration of the computing system hosting the computing elements, identify a plurality of computing elements to batch process, and batch process the plurality of computing elements based on the determined configuration of the computing system.

Abstract: A computer storage device having a host interface, a controller, non-volatile storage media, and firmware. The firmware instructs the controller to: allocate a named portion of the non-volatile storage device; generate, according to a first block size, first block-wise mapping data; translate, using the first block-wise mapping data, logical addresses defined in the named portion to logical addresses defined for the entire non-volatile storage media, which can then be further translated to physical addresses in a same way for all named portions; determine a second block size; generate, according to the second block size, second block-wise mapping data; translate, using the second block-wise mapping data, the logical addresses defined in the name portion to the logical addresses defined for the entire non-volatile storage media.

Abstract: Techniques for rebalancing computing workloads between a set of on-premises resources of an on-premises system and a set of cloud computing resources of a cloud computing system. A method embodiment commences upon detecting a rebalancing trigger event. Responsive to the trigger event, a set of resource data corresponding to observations pertaining to the on-premises resources and resource data corresponding to observations pertaining the cloud resources is accessed. Based at least in part on the resource data, a plurality of candidate workload rebalancing plans are generated. The candidate workload rebalancing plans are evaluated against the likelihood of achieving quantitative objectives resulting from the rebalancing. A portion of the scheduling commands to carry out the workload rebalancing plans are sent to the cloud computing system, and another portion of the scheduling commands are sent to the on-premises computing system. The receiving computing systems carry out the scheduling commands.

Abstract: The method includes identifying a first data table that includes a set of rows and a structure. The method further includes creating a second data table and a third data table having a matching structure as the first table. The method further includes distributing the set of rows of the first data table, wherein the set of rows is distributed between one or more of the second data table and the third data table based upon preset parameters. The method further includes, generating one or more operations for the set of rows. The method further includes executing one of the one or more generated operations on the second data table and the third data table.

Abstract: A distributed computing system for automatic constraint-based data resource arrangement, comprising a plurality of computing components being communicatively coupled to each other, each computing component comprising the following data resources: data storage media for storing client-related digital information, a data processor for processing said client-related digital information, and a network communications interface for communicating said client-related digital information; and a constraint engine for automatically determining alternate arrangements of said data resource assignments, said constraint engine comprising a constraint processor and a constraint database, said constraint database for receiving and storing changeable digital constraint parameters indicative of permissible operational constraints on said data resources, wherein said alternate arrangements comply with at least a first set of said changeable digital constraint parameters; wherein said data resource assignments are reassigned from

Abstract: A processor comprising a mode indicator, a plurality of processing cores, and a neural network unit (NNU), comprising a memory array, an array of neural processing units (NPU), cache control logic, and selection logic that selectively couples the plurality of NPUs and the cache control logic to the memory array. When the mode indicator indicates a first mode, the selection logic enables the plurality of NPUs to read neural network weights from the memory array to perform computations using the weights. When the mode indicator indicates a second mode, the selection logic enables the plurality of processing cores to access the memory array through the cache control logic as a cache memory.

Abstract: According to one embodiment, a storage system performs a first allocation operation of allocating, for a first namespace, a plurality of first blocks included in the blocks of a nonvolatile memory. The storage system performs a read operation, a write operation or an erase operation on one of the first blocks in response to a command received from a host to read, write or erase the one first block, counts the total number of erase operations performed on the first blocks, and notifies the host of the counted number of erase operations in response to a command received from the host to obtain an erase count associated with the first namespace.

Abstract: The method includes identifying a first data table that includes a set of rows and a structure. The method further includes creating a second data table and a third data table having a matching structure as the first table. The method further includes distributing the set of rows of the first data table, wherein the set of rows is distributed between one or more of the second data table and the third data table based upon preset parameters. The method further includes, generating one or more operations for the set of rows. The method further includes executing one of the one or more generated operations on the second data table and the third data table.

Abstract: The invention relates to a method, computer program product and computer system for providing attribute value information for a data extent comprising a set of data entries.

Abstract: A data processing technology is provided, and is applied to a partition management device. The partition management device stores a partition view, the partition view records a correspondence between an ID of a current partition and an address of a storage disk, and a total quantity of current partitions may be less than a total quantity of final partitions.

Abstract: A control device to be connected to a host and a disk device including at least one Redundant Arrays of Independent Disks (RAID) set, the control device includes circuitry that performs data communication with the host by a protocol capable of connecting one logical device in the RAID set to each port, virtually sets a plurality of ports which are defined by the protocol, sets a plurality of logical devices in the RAID set included in the disk device, and connects the plurality of logical devices to the plurality of ports, respectively, and the circuitry is to be connected to the host through an interface.

Abstract: Examples relate to purging storage partitions of a database. The examples disclosed herein identify a first partition of a database to be purged and identify a data entry in the first storage partition, where the data entry is to be copied. Examples herein copy an updated version of the data entry to a future storage partition of the database and purge the first storage partition. A dummy data entry is created in a second storage partition of the database, where the dummy data entry identifies the future storage partition.

Abstract: This disclosure generally relates to hypervisor memory virtualization. In an example, multiple page table stages may be used to provide a page table that may be used by a processor when processing a workload for a nested virtual machine. An intermediate (e.g., nested) hypervisor may request an additional page table stage from a parent hypervisor, which may be used to virtualize memory for one or more nested virtual machines managed by the intermediate hypervisor. Accordingly, a processor may use the additional page table stages to ultimately translate a virtual memory address for a nested virtual machine to a physical memory address.

Abstract: A network-based data store maintaining storage volumes for file systems at storage clients may implement dynamic storage volume scaling. A scaling event may be detected according to a specified scaling policy for a storage volume maintained at a network-based data store that is mounted at a storage client and configured for a file system. In a least some embodiments, the scaling policy for the storage volume may be received from the storage client. In response to detecting the storage event, storage maintaining the storage volume may be modified according to the scaling policy to scale the storage volume. An indication of the scaling event may be sent to the storage client in order to update the file system at the storage client so that the file system may utilize the scaled storage volume.

Abstract: An information handling system includes a dynamic random access memory, and a processor. The dynamic random access memory includes a lower memory portion and multi-channel dynamic random access memory portion. The dynamic random access memory is allocated to operations of a boot process of the information handling system. The processor communicates with the dynamic random access memory. The processor determines whether a fast memory allocation service is detected in the boot process. In response to the fast memory allocation being detected, the processor allocates the multi-channel dynamic random access memory portion of the dynamic random access memory to operations of the boot process.

Abstract: Various embodiments are presented herein that may allow an application direct access to graphical processing unit memory. An apparatus and a computer-implemented method may include accessing allocated graphical processing unit memory of a second resource via a link from a first resource. The allocated graphical processing unit memory may be mapped into one or more page tables of a central processing unit. A virtual address of the graphical processing unit memory from the one or more page tables of the central processing unit may be sent to the application.

Abstract: A storage device with a memory may have a hidden diagnostic partition that can only be accessed during debug or diagnostic mode. Debug or diagnostic mode allows a host device to access the debug or diagnostic analysis (e.g. error logs) stored in the hidden diagnostic partition. By default, the hidden diagnostic partition is invisible to the host. When accessed through a triggering event, such as a vendor specific command (“VSC”), the hidden diagnostic partition can be used to report debug and error events.

Abstract: Disclosed herein are techniques for managing partitions on a storage device. A method can include (1) identifying a storage capacity of the storage device, (2) generating a first data structure that defines a first partition on the storage device, where the first partition consumes a first amount of the storage capacity, and (3) generating a second data structure that defines a second partition on the storage device, where the second partition consumes at least a portion of a remaining amount of the storage capacity relative to the first amount. In response to receiving a shrink request directed to the first partition, the method can further include (4) identifying a first utilized area within the first partition that will no longer be utilized as a result of the shrink request, and (5) updating first information in the first data structure to indicate that the first utilized area is unutilized.

Abstract: Embodiments relate to enabling state manipulation of virtual machines (VMs) that have directly assigned hardware devices. A hypervisor manages execution of a VM with a guest operating system. The hypervisor directly assigns a physical (non-virtual) hardware device to the VM, such that the VM has exclusive use of the hardware device and the guest operating system uses the hardware without needing virtualization abstraction/mapping by the hypervisor. When the VM needs to be pauses, suspended, restored, or similarly manipulated, the hypervisor communicates with the VM to cause the guest operating system to perform whatever operations might be necessary by the hardware device. The hypervisor and VM/guest may communicate with a variety of mechanisms, such as a host driver communicating with a guest driver, a virtual bus service backed by a hardware bus of the computer, a virtual bus channel, or the like.

Abstract: An operating system (OS) of a processing system having a plurality of processor cores determines a cost associated with different mechanisms for performing a translation lookaside buffer (TLB) shootdown in response to, for example, a virtual address being remapped to a new physical address, and selects a TLB shootdown mechanism to purge outdated or invalid address translations from the TLB based on the determined cost. In some embodiments, the OS selects an inter-processor interrupt (IPI) as the TLB shootdown mechanism if the cost associated with sending an IPI is less than a threshold cost. In some embodiments, the OS compares the cost of using an IPI as the TLB shootdown mechanism versus the cost of sending a hardware broadcast to all processor cores of the processing system as the shootdown mechanism and selects the shootdown mechanism having the lower cost.

Abstract: Data is received describing attributes of a first mainframe computing system at a software appliance, where the software appliance is hosted at least in part on a second mainframe computing system and the software appliance includes a plurality of microservices hosted in a plurality of software containers. The data is sent to a first one of the plurality of microservices hosted in a first one of the plurality of software containers, where the first microservice includes data science logic. The data is analyzed using the data science logic to generate a result at the first microservice, and it is determined whether to send the result to a second one of the plurality of microservices hosted in a second one of the plurality of software containers, where the second microservice includes logic to determine an alert condition based on results received from other microservices in the software appliance.

Abstract: A certified checkpoint is provided for a ledger comprising a blockchain and a world state. The certified checkpoint enables a third party to recognize and verify that the ledger has integrity, a known starting state, and immutability properties starting at a specific point in time. Certification means that all of the validating peers reached consensus on the state of the ledger at that point in time. Thus, the certified checkpoint state represents an agreed-upon state, and that one or more subsequent operations on the ledger are relative to that agreed-upon state. Preferably, before a checkpoint is certified, it must be consistent, meaning that all validating peers have reached the same value for the checkpoint. Preferably, the checkpoint is a compression of the current blockchain world state into a compact representation (e.g., a hash) of the ledger that based on an agreed-upon consensus protocol is consistent across the (validating) peers. The approach also is extended to a permissionless blockchain.

Abstract: A region-based placement policy that can be used to achieve a better distribution of data in a clustered storage system is disclosed herein. The clustered storage system includes a master module to implement the region-based placement policy for storing one or more copies of a received data across many data nodes of the clustered storage system. When implementing the region-based placement policy, the master module splits the received data into one or more regions, where each region includes a contiguous portion of the received data. Further, for each of the plurality of regions, the master module stores complete copies of the region in a subset of the data nodes.

Abstract: Disclosed embodiments relate to generating an update package for updating software on an Electronic Control Unit (ECU) in a vehicle. Operations may include accessing a plurality of attributes of a software update to be stored on the ECU in the vehicle; accessing a corresponding plurality of attributes of current software stored on the ECU in the vehicle; comparing the plurality of attributes with the corresponding plurality of attributes; generating a delta file representing differences between the plurality of attributes and the corresponding plurality of attributes determined in the comparison; and providing the delta file to the ECU, wherein the delta file is configured to be processed by startup code in the ECU that enables the delta file to execute in the ECU in the vehicle.

Abstract: A key-value storage system is described herein for interacting with key-value entries in a content store using a resource-efficient index. The index provides a data structure that includes a plurality of hash buckets. Each hash bucket includes a linked list of hash bucket units. The key-value storage system stores hash entries in each linked list of hash bucket units in a distributed manner between an in-memory index store and a secondary index store, based on time of their creation. The key-value storage system is further configured to store hash entries in a particular collection of linked hash bucket units in a chronological order to reflect time of their creation. The index further includes various tunable parameters that affect the performance of the key-value storage system.

Abstract: A computing pod includes one or more computing racks. Each of the one or more computing racks includes one or more computing drawers. Each of the computing drawers includes one or more computing modules. Each of the one or more computing modules includes a computing blade. A first service processor is on a first computing blade of a first computing drawer of the one or more computing drawers of a first computing rack of the one or more computing racks. The first service processor executes a first BMC function. The first BMC function manages the first computing blade. The first service processor determines a type of the first computing blade. The first service processor configures a first PSME in accordance with the determined type.

Abstract: Disclosed are various examples for top-of-rack (TOR) switch replacement in hyper-converged computing environments. A hyper-converged management service can maintain network configurations applied to a first network switch of a rack. When the performance of the first network switch does not satisfy a performance threshold, the hyper-converged management service can decommission the first network switch in the rack and configure a second network switch for replacement of the first network switch in the rack by accessing the at least one network configuration applied to the first network switch from the memory and applying the at least one network configuration to the second network switch.

Abstract: Techniques are disclosed for splitting a database partition into two partitions. In embodiments, where the partition is a hash partition, the partition is split at its logical midpoint, so that half of the partition's rows are placed in each split partition. Where the partition is a hash-range partition, a number of rows of the partition are sampled. Where enough samples fall on each side of the logical midpoint, the logical midpoint is again used as the split point. Where not enough samples fall on one side of the logical midpoint, then the median of the samples is used as the split point.

Abstract: A system and method is disclosed for managing data in a non-volatile memory. The system may include a non-volatile memory having multiple non-volatile memory sub-drives. A controller of the memory system is configured to route incoming host data to a desired sub-drive, keep data within the same sub-drive as its source during a garbage collection operation, and re-map data between sub-drives, separate from any garbage collection operation, when a sub-drive overflows its designated amount logical address space. The method may include initial data sorting of host writes into sub-drives based on any number of hot/cold sorting functions. In one implementation, the initial host write data sorting may be based on a host list of recently written blocks for each sub-drive and a second write to a logical address encompassed by the list may trigger routing the host write to a hotter sub-drive than the current sub-drive.

Abstract: To provide an electronic device that is capable of maintaining a backup performance of a specified type of data for a long period of time. An MFP has a feature to includes a HDD, an SSD, a data type determination part that determines whether or not data is a specified type of data, a writing control part that controls a data writing, the writing control part writing the data that is determined to be the specified type of the data to both the HDD and the SSD, the writing control part writes the data that is determined not to be the specified type of the data to only the HDD, among the HDD and the SSD.

Abstract: Systems and methods are described for capturing and reproducing the full state of an application session. An application is captured by performing a dump of various components of the session and storing the components in an application session capture. Captured components include a memory image, CPU register values, open handles to kernel objects, and loaded libraries. The session is reproduced when requested based on the session capture. In cases where locations or references to certain items (e.g., memory locations, open handle references, library addresses, etc.) are different when the session is restored, a driver is used to remap the old locations or references to the new locations or references.

Abstract: A processor comprising a plurality of processing cores, a last level cache memory (LLC) shared by the plurality of processing cores, and a neural network unit (NNU) comprising an array of neural processing units (NPU) and a memory array. The LLC comprises a plurality of slices. To transition from a first mode in which the memory array operates to store neural network weights read by the plurality of NPUs to a second mode in which the memory array operates as a slice of the LLC in addition to the plurality of slices, the processor write-back-invalidates the LLC and updates a hashing algorithm to include the memory array as a slice of the LLC in addition to the plurality of slices. To transition from the second mode to the first mode, the processor write-back-invalidates the LLC and updates the hashing algorithm to exclude the memory array from the LLC.

Abstract: A data storage system operates to shrink an intermediate logical volume and lower file system upon deletion of a first upper data unit of an upper file system without moving file data on the physical storage devices. Initially, a second upper data unit of the upper file system is identified that is mapped to an end location of the intermediate logical volume. Within the lower file system, a non-end location of the volume file is re-mapped to a first location on the physical storage devices where the data of the end location of the intermediate logical volume is stored. Within the upper file system, the second upper data unit is re-mapped to the non-end location of the intermediate logical volume. Subsequently, the intermediate volume is truncated by removing the portion at the non-end location.

Abstract: Embodiments are described for repairing the locality of a namespace index on a deduplication storage device. A namespace index is selected from a plurality of namespace indices, each of which is mountable on the storage system. The selected namespace index is traversed and divided into regions, each of which was stored in one or more storage containers. Locality information for each region is generated, including a density metric and a distribution metric. If either metric is below a threshold value for the metric, then the region is enqueued for namespace locality repair. A region can be repaired by reading the region into memory and flushing the memory to co-located containers on the storage system.

Abstract: A system and method is disclosed for managing data in a non-volatile memory. The system may include a non-volatile memory having multiple non-volatile memory sub-drives. A controller of the memory system is configured to route incoming host data to a desired sub-drive, keep data within the same sub-drive as its source during a garbage collection operation, and re-map data between sub-drives, separate from any garbage collection operation, when a sub-drive overflows its designated amount logical address space. The method may include initial data sorting of host writes into sub-drives based on any number of hot/cold sorting functions. In one implementation, the initial host write data sorting may be based on a host list of recently written blocks for each sub-drive and a second write to a logical address encompassed by the list may trigger routing the host write to a hotter sub-drive than the current sub-drive.

Abstract: Described embodiments provide systems and methods for online system recovery orchestration. Embodiments may include receiving a request to recover a system management process from a checkpoint. The system management initialization parameter file is updated by setting a flag to start from the checkpoint. A system management process is started with the updated initialization parameter and a copy of the system management repository is loaded. Volumes are recovered before data path module activation. Data modules are then activated with checkpoint offset information and data module journals are reformatted. Control modules are then activated with recovered volume configuration and control module journals are reformatted. Defragmentation mode is set to rollback mode and data on checkpoint stripes that were written after checkpoint creation are deleted. After waiting for the defragmentation rollback to complete the host IO direction is set to checkpoint stripes.

Abstract: A computing device includes a host, one or more guest partitions, and one or more physical devices. A physical device can be virtualized, at least in part, by the host and made available to the guest partitions. A physical device includes both a control plane and a data plane. The host provides direct access to at least part of the data plane of a physical device to a guest partition. However, the host virtualizes the control plane of the physical device, exposing a control plane for the physical device to the guest partition that is not the actual control plane of the physical device. Requests to access (e.g., read, write, modify, etc.) the control plane of the physical device are received by the host from the guest partition, and converted as appropriate to the control plane for the physical device.

Abstract: Provided are a computer program product, system, and method for selecting resources to make available in local queues for processors to use. Each processor of a plurality of processors maintains a queue of resources for the processor to use when needed for processor operations. One of processors is selected. The selected processor accesses at least one available resource and includes the accessed at least one resource in the queue of the selected processor.

Abstract: A resource management device provides a communication service in a network in response to a service request. The resource management device includes a processor. The processor executes a process. The process includes converting a total resource amount in the network into a first value. The process includes assigning a second value to each type of the communication service from the first value. The process includes second converting a resource requesting amount of the service request into a third value. The process includes setting, determining that the third value is equal to or less than the second value that is assigned to the type of the communication service of the service request, a communication service associated with the service request for a communication device on the network.

Abstract: Systems and methods for automatic provisioning of a redundant array of independent disks (RAID) are disclosed. In particular, the systems and methods include automatically determining, using one or more processors, that a first partition associated with a first storage device and a second partition associated with a second storage device are associated with a common mount point; and automatically adding, using the one or more processors, the first partition and the second partition to a RAID.

Abstract: The present disclosure relates to a data science system that packages data science operations. The data science system packages a data science operation with a component descriptor or service descriptor to allow the data science system to easily apply and execute the data science operations using data from a variety sources. As described herein, the data science system also enables a user to provide data science packages to a marketplace as well as retrieve data science packages created by other users from the marketplace. Further, the data science system can customize a data science package obtained from the marketplace to perform data science operations using data belonging to the user or using user-specified parameters.

Abstract: An Input/Output (I/O) adapter device coupled to a host device can perform data mirroring or data striping of payload data for transmitting to multiple network destinations. In some embodiments, a virtual machine running on the host device or on the I/O adapter device may be aware of the capabilities of the I/O adapter device to perform data mirroring or data striping and configure the I/O adapter device for performing data mirroring or data striping operations. In some embodiments, a virtual machine may be agnostic to the capabilities of the I/O adapter device to perform data mirroring or data striping and the I/O adapter device may perform data mirroring or data striping “under the hood” or without being configured by the virtual machine.

Abstract: Efficiently generating effective address translations for memory management test cases including obtaining a first set of EAs, wherein each EA comprises an effective segment ID and a page, wherein each effective segment ID of each EA in the first set of EAs is mapped to a same first effective segment; obtaining a set of virtual address corresponding to the first set of EAs; translating the first set of EAs by applying a hash function to each virtual address in the set of virtual addresses to obtain a first set of PTEG addresses mapped to a first set of PTEGs; and generating a translation for a second set of EAs to obtain a second set of PTEG addresses mapped to the first set of PTEGs.

Abstract: Techniques are provided for managing cached data objects in a mixed workload environment. In an embodiment, a database system receives request to access a target data object. The database system determines whether the request to access the target data object is associated with a first type of workload or a second type of workload. In response to determining that the request is associated with the first type of workload, the target data object replaces a least recently used data object in a cache. In response to determining that the request is associated with the second type of workload, the target data object is cached based on an associated access-level value.

Abstract: The present disclosure introduces a method and an apparatus of shrinking virtual hard disk image file. The present techniques search a garbage data block in a file and revise a record of the garbage data block in a block allocation table (BAT). The file includes one or more data blocks and a BAT that records information of each data block. The garbage data block is a data block that does not store effective data. The present techniques move an effective data block subsequent to the garbage data block, revise a record of the effective data block in the BAT after the moving, and shrink a size of the file after the movement.

Abstract: Certain aspects direct to systems and methods for device or vendor independent network switch management on a management controller. The management controller is communicatively connected to a network switch through a Simple Network Management Protocol (SNMP) interface. The management controller receives parsed information of a management information base (MIB) file corresponding to the network switch, and establishes a communication between the management controller and the network switch through the SNMP interface based on the parsed information of the MIB file, in which the management controller functions as a client and the network switch functions as a server of the communication. Then the management controller receives an input to manage and configure the network switch, and manages and configures the network switch via the communication through the SNMP interface based on the input and the parsed information of the MIB file.