Abstract: Embodiments are described for prioritizing input/output (I/O) operations dispatched to a solid-state device (SSD) storage in a network, determining properties of the SSD storage based on SSD device type, where the SSD storage maintains a plurality of caches, receiving a plurality of SSD I/Os from an application I/O request comprising at least one of a cache read or a cache write operation, creating a respective queue for each SSD I/O of the SSD I/Os based on a type of SSD I/O, prioritizing the each SSD I/O based on the type of the SSD I/O and the properties of the SSD storage that is the target for a respective SSD I/O, and dispatching, in a transactional I/O SSD scheduler, all queued SSD I/Os to the SSD storage based on the prioritizing.

Abstract: One embodiment provides a computer implemented method for installing an operating system (OS) using a dual-flash device. The method includes burning an OS version to the dual-flash device located on a server. The method also includes setting BIOS to boot the server from the dual-flash device; triggering installation of the OS version on the server from the dual-flash device; and setting BIOS to boot from a hard disk of the server after installation of the OS version.

Abstract: One embodiment provides a computer implemented method for recovering an operating system (OS) after an upgrade hang using a dual-flash device. The method includes detecting a system hang during an OS upgrade; initiating a boot from the dual-flash device; checking for OS configuration data in the dual-flash device; creating a hook in OS boot scripts to recover the OS configuration data after OS boot; and applying the OS configuration data after OS boot.

Abstract: One embodiment provides a computer implemented method for recovering an operating system (OS) after a runtime hang using a dual-flash device. The method includes detecting a first runtime hang of a server; initiating a first reboot from a hard disk; detecting a second runtime hang of the server; and initiating a second reboot from the dual-flash device.

Abstract: One embodiment provides a computer implemented method of for monitoring and upgrading a dual-flash device. The method includes performing an OS upgrade on a server; writing an upgraded OS to the dual-flash device; updating a grub.cfg file corresponding to the upgraded OS; and deleting old OS files from the dual-flash device.

Abstract: One embodiment provides a computer implemented method of dynamically allocating compression jobs including monitoring compression performance at a plurality of compression levels within a hardware accelerator; comparing compression performance between the plurality of compression levels; and dynamically allocating compression jobs between the plurality of compression levels to achieve a guaranteed average performance.

Abstract: Embodiments are described for prioritizing input/output (I/O) operations dispatched to a solid-state device (SSD) cache in a network, by defining a maximum write I/O operation size for writing data to the SSD cache, splitting large write I/O operations into smaller write I/O operations, each with a size less than the maximum write I/O operation size, interleaving cache read I/O operations in between the smaller write I/O operations, and performing the cache read I/O operations and the smaller write I/O operations in an order created by the interleaving. The network may comprise a deduplication backup system storing data to storage media including the SSD cache.

Abstract: Embodiments are described for prioritizing input/output (I/O) operations dispatched to a solid-state device (SSD) storage in a network, determining properties of the SSD storage based on SSD device type, where the SSD storage maintains a plurality of caches, receiving a plurality of SSD I/Os from an application I/O request comprising at least one of a cache read or a cache write operation, creating a respective queue for each SSD I/O of the SSD I/Os based on a type of SSD I/O, prioritizing the each SSD I/O based on the type of the SSD I/O and the properties of the SSD storage that is the target for a respective SSD I/O, and dispatching, in a transactional I/O SSD scheduler, all queued SSD I/Os to the SSD storage based on the prioritizing.

Abstract: Embodiments are described for prioritizing input/output (I/O) operations dispatched from a storage media to a host bus adapter in a network, by tagging, in an I/O tagging module, the I/O operations in a file system supporting applications generating the I/O operations, wherein all child I/O operations initiated from a parent I/O operation are tagged with a same unique tag ID; tracking a time of arrival of each I/O operation of the I/O operations; and dispatching, in a transactional I/O scheduler, all sibling I/O's of the parent I/O operation based on a unique tag ID for the sibling I/Os, a respective time of arrival of each of the sibling I/Os, and defined quality of service (QoS) requirements.

Abstract: Configuring virtual machines for garbage collection in a deduplication backup system executed on a server computer in a cloud computing network, by configuring at least one virtual machine (VM) of a plurality of virtual machines in a VM cluster to be data protection virtual machines (DP VMs), by enabling only read and write operations to/from the DP VMs, interconnecting a garbage collection VM (GC VM) of the VM cluster to the DP VMs, configuring the GC VM to perform only garbage collection tasks for the DP VMs, and providing data access credentials from each DP VM to the GC VM to facilitate the garbage collection tasks.

Abstract: One embodiment provides a computer implemented method of improving backup and restore performance including sending a compression job to a hardware accelerator using a compression thread; providing a callback pointer for the compression job; monitoring the hardware accelerator using a polling thread; calling the callback pointer to notify the compression thread when the hardware accelerator is available; and retrieving data from a destination buffer using the compression thread via a destination buffer pointer.

Abstract: One embodiment provides a computer implemented method of dynamically allocating compression jobs including monitoring compression performance at a plurality of compression levels within a hardware accelerator; comparing compression performance between the plurality of compression levels; and dynamically allocating compression jobs between the plurality of compression levels to achieve a guaranteed average performance.

Abstract: One embodiment provides a computer implemented method of data compression including segmenting user data into data segments; deduplicating the data segments to form deduped data segments; compressing the deduped data segments into compression units using a hardware accelerator; packing the compression units into compression regions; and packing the compression regions into one or more containers.

Abstract: One embodiment provides a computer implemented method of balancing mixed workload performance including monitoring the compression and decompression workload at a hardware accelerator using the hardware accelerator quality of service (QoS) scheduler; monitoring the compression and decompression workload at a CPU using the CPU QoS scheduler; comparing the workload at the hardware accelerator and the workload at the CPU; and allocating tasks between the hardware accelerator and the CPU to obtain an optimal bandwidth at the hardware accelerator and the CPU.

Abstract: Embodiments of the present disclosure disclose methods and devices of data de-duplication. The method of data de-duplication performed at a client comprises: in response to receiving data to be backed up at a client, sampling the data to be backed up to obtain the sampled data; generating a signature for the sampled data; transmitting the signature to a master storage node in a storage cluster including a plurality of storage nodes, to allow the master storage node to select one storage node from the plurality of storage nodes; receiving an indication of the selected storage node from the master storage node; and transmitting, based on the indication, the data to be backed up to the selected storage node. Embodiments of the present disclosure also provide methods of data de-duplication performed at the master storage node and the slave storage node, and corresponding devices.

Abstract: One embodiment provides a computer implemented method of data compression using a hardware accelerator. A first thread pool for compression jobs, and a first polling thread is allocated for polling the status of a hardware accelerator. A compression thread is retrieved from the first thread pool in response to a compression request from a file system. Multiple source data buffers from the file system are aggregated into a compression unit, and a scatter gather list and destination buffer are submitted to the hardware accelerator. A checksum of result data is calculated from the destination buffer. A zlib header is added to the result data, and the checksum is added as a zlib footer to the result data.

Abstract: One embodiment provides a computer implemented method of data compression using a hardware accelerator. The method includes communicating, using a hardware accelerator, to a client device a set of instructions that compressed data be sent in LZ compressed format; receiving LZ compressed data at the hardware accelerator in LZ compressed format; decompressing the LZ compressed data using the hardware accelerator to create decompressed data; recompressing the decompressed data into GZ format, using the hardware accelerator, to create GZ compressed data; and sending the GZ compressed data to a storage device.

Abstract: Embodiments are described for prioritizing input/output (I/O) operations dispatched from a storage media to a host bus adapter in a virtual or cloud network by: determining certain latency properties of the storage media; tagging, in an I/O tagging module, the I/O operations in a file system supporting applications generating the I/O operations, wherein all child I/O operations initiated from a parent I/O operation are tagged with a same unique tag ID; and dispatching, in a transactional I/O scheduler, all sibling I/O's of the parent I/O operation based on a unique tag ID for the sibling I/Os, the storage media latency properties, and defined quality of service (QoS) requirements. The I/O operations may be sequential, random, or internal I/O operations with random I/O operations being more latency dependent and prioritized over the sequential or internal operations.

Abstract: Embodiments of the present disclosure relate to method, device and computer program product for data backup. The method comprises: in response to receiving from a backup server a data stream to be backed up, dividing the data stream into a plurality of data segments; distributing the plurality of data segments to at least one computing node; in response to receiving an index of a corresponding data segment from a first computing node of the at least one computing node, looking up the index in a global index cache, the index being generated by the first computing node to uniquely identify the data segment, the global index cache storing indexes of data in a backup storage device; in response to the missing index in the global index cache, adding the index into the global index cache; and sending to the first computing node an indication to store the data segment in the backup storage device.

Abstract: Embodiments for systems and methods of resetting network devices for failover operation, by receiving an operating system panic function call, disabling error reporting for all device ports of a device to be reset to prevent primary interrupts to a disk dump operation, performing a function level reset of the all the device ports to be reset, and performing the reset operation on the device in a single thread context to prevent secondary interrupts to the disk dump process.