Abstract: A method, computer program product, and computer system for extending, by a computing device, transaction log page-buffers for Non-Volatile Random Access Memory (NVRAM) onto a solid state drive (SSD). It may be determined whether a bandwidth limit of the NVRAM has reached a threshold bandwidth. An IO may be processed on one of the NVRAM and the SSD based upon, at least in part, whether the bandwidth limit of the NVRAM has reached the threshold bandwidth.

Abstract: A method, computer program product, and computer system for submitting, by a computing device, a write for a write-request equal to a size of valid data in a segment. The write-request may be decomposed into RAID elements and corresponding parity stripe elements. Parity for RAID stripes may be calculated using only valid RAID elements of the RAID elements. A write for all the valid RAID elements and the corresponding parity stripe elements may be issued. An unmap command may be issued to at least a portion of the segment that is invalid.

Abstract: A technique maintains availability of a non-volatile cache. The technique involves arranging a plurality of non-volatile random-access memory (NVRAM) drives into initial drive sets that form the non-volatile cache. The technique further involves detecting a failed initial drive set among the initial drive sets. The plurality of NVRAM drives now includes failed NVRAM drives that belong to the failed initial drive set and remaining non-failed NVRAM drives. The technique further involves, in response to detecting the failed initial drive set, re arranging the remaining non-failed NVRAM drives of the plurality of NVRAM drives into new drive sets that form the non-volatile cache.

Abstract: A method, computer program product, and computer system for virtualizing, by a computing device, a physical metadata space into a virtual metadata space. A translation table from the virtual metadata space to the physical metadata space may be implemented. Metadata in the physical metadata space may be cached based upon the virtual metadata space. The metadata in the physical metadata space may be moved without updating references associated with the metadata in the physical metadata space.

Abstract: In a method used for maximizing performance of a storage system, saturation points of the storage system for I/O requests of different types and sizes are identified. Normalized tokens are determined based on the saturation points. Unique numbers of normalized tokens are associated with the I/O requests of different types and sizes. A number of normalized tokens for a queue depth limit of a storage device is determined. From the queue depth limit, normalized tokens are allocated to each node in the storage device. I/O requests are processed according to the allocations of normalized tokens.

Abstract: Techniques for managing storage may comprise: receiving a request for a first amount of free capacity, wherein the request includes a first priority denoting a purpose for which the first amount of free capacity is requested; determining whether a current utilization of storage exceeds a first threshold associated with the first priority; responsive to determining the current utilization of storage is less than the first threshold associated with the first priority, performing first processing including: determining whether there is a sufficient amount of existing free capacity to grant the first amount; and responsive to determining there is a sufficient amount of existing free capacity to grant the first amount, granting the first amount of free capacity; and responsive to determining the current utilization of storage is not less than the first threshold associated with the first priority, rejecting the request and not granting the first amount of free capacity.

Abstract: Techniques for providing lifecycle handling of faults associated with a storage appliance. The techniques can include recording information or details of one or more detected faults associated with a storage appliance in a fault database, analyzing the recorded fault information/details by an analytic tool, and determining whether to perform immediate recovery or defer recovery of the detected fault(s) based at least on a recommendation of the analytic tool. In this way, taking the storage appliance offline upon occurrence of a single fault can, for the most part, be avoided. In addition, because the detection of faults is not performed within a fault domain for a filesystem handling files for a single or limited number of volume families, a deduplication domain for the files of the filesystem is not restricted to the single or limited number of volume families.

Abstract: A protocol for processing write operations may include recording each write operation in a log using a PDESC(page descriptor)-PB(page block) pair. The protocol may be a data commit sequence that assigns ownership to containers of logged writes. Each container may be associated with a particular logical block address (LBA) range of a logical device, where logged writes that write to target addresses in the particular LBA range of the logical device may be included in the container. Ownership of a container may be determined by the initial write operation that is mapped to, and placed into, the container based on the target logical address of the write operation. A container may be allocated in response to the initial write operation of the container. Multiple nodes may independently flush containers of logged operations based on the particular node designated as the current owner of the containers.

Abstract: In a method used for maximizing performance of a storage system, saturation points of the storage system for I/O requests of different types and sizes are identified. Normalized tokens are determined based on the saturation points. Unique numbers of normalized tokens are associated with the I/O requests of different types and sizes. A number of normalized tokens for a queue depth limit of a storage device is determined. From the queue depth limit, normalized tokens are allocated to each node in the storage device. I/O requests are processed according to the allocations of normalized tokens.

Abstract: Technology is disclosed for recovering the consistency of a RAID (Redundant Array of Independent Disks) metadata database when data corruption is detected in the RAID metadata database. The RAID metadata database includes super sectors, stage sectors, and a data region. Valid data within the data region is a contiguous set of sectors extending from a head sector to a tail sector. In response to data corruption in one of the two super sectors, a set of pointers contained in the other super sector is used to identify the head sector and tail sector. In response to data corruption in both super sectors, the head sector and tail sector are located based on the contents of the sectors in the data region. Techniques are also disclosed for recovering consistency when the data corruption occurs in the stage sectors and/or data region.

Abstract: A technique manages data within a storage array. The technique involves forming a hybrid tier within the storage array, the hybrid tier including SSD storage and HDD storage. The technique further involves, after the hybrid tier is formed, providing hybrid ubers (or Redundant Array of Independent Disks (RAID) extents) from the SSD storage and the HDD storage of the hybrid tier. The technique further involves, after the hybrid ubers are provided, accessing the hybrid ubers to perform data storage operations.

Abstract: A technique performs Redundant Array of Independent Disks (RAID) transformation. The technique involves performing a garbage collection operation on a first uber within a storage array, the garbage collection operation freeing a set of disk slices of the first uber. The technique further involves, upon completing the garbage collection operation, reallocating storage of the set of disk slices from the first uber to a second uber within the storage array. The technique further involves, after the storage of the set of disk slices is reallocated from the first uber to the second uber, storing data within the second uber. Such leveraging of garbage collection when performing RAID transformation reduces overhead and wear without not negatively affecting system performance.

Abstract: Rebalancing the workload of logical devices across multiple nodes may include dynamically modifying preferred paths for one or more logical devices in order to rebalance the I/O workload of the logical devices among the nodes of the data storage system. Determining whether to rebalance the I/O workload between the two nodes may be performed in accordance with one or more criteria. Processing may include monitoring the current workloads of both nodes over time and periodically evaluating, in accordance with the one or more criteria, whether the current workloads of the nodes are imbalanced. Responsive to determining, in accordance with the criteria, that rebalancing of workload between the nodes is needed, the rebalancing may be performed. A notification may be sent to the host regarding any path state changes made as a result of the workload rebalancing.

Abstract: A method of operating a storage appliance is provided.

Abstract: An apparatus comprises at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured to: obtain a set of rebuild rate parameters for a given storage device from a storage array comprising a plurality of storage devices; and dynamically regulate a rebuild rate associated with a rebuild process for the given storage device based on the set of rebuild rate parameters obtained from the storage array for the given storage device. For example, the set of rebuild rate parameters include a rebuild capacity parameter and a rebuild time parameter.

Abstract: A method of correcting errors in a data storage system including a first node, a second node, and shared persistent storage (the first and second nodes being configured to process data storage requests) is provided. The method includes (a) reading cached pages from a first cache disposed within the first node, the cached pages being cached versions of respective persistent pages stored in the shared persistent storage; (b) in response to determining that one of the cached pages is corrupted, requesting that the second node return to the first node a corresponding remote page from a second cache disposed within the second node, the cached page and the remote page each caching a same persistent page of the shared persistent storage; and (c) in response to determining that the remote page received from the second node by the first node is not corrupted, correcting the cached page using the remote page.

Abstract: Techniques for rebuilding a failed drive of a plurality of drives are provided. The techniques include: (a) determining a number of empty data stripes that have been allocated from the plurality of drives; (b) applying a set of decision criteria to the determined number to yield a decision, wherein applying the set of decision criteria yields a first decision at a first time and a differing second decision at a second time; (c) in response to the first decision at the first time, deallocating an empty data stripe that was allocated from the plurality of drives to yield a plurality of data extents on disparate drives of the plurality of drives and continuing to rebuild the failed drive onto at least one of the plurality of data extents; and (d) in response to the second decision at the second time, pausing rebuilding the failed drive until applying the set of decision criteria yields the first decision.

Abstract: Dirty pages of cached user data are persistently stored to page buffers that are allocated from a page buffer pool in a persistent data storage resource of a data storage system, and are indicated by page descriptors that are stored at a head of a temporally ordered page descriptor ring as the dirty pages are stored to the page buffers. The disclosed technology performs a flush operation by selecting a work-set of non-sequential page descriptors within the page descriptor ring, flushing dirty pages from page buffers indicated by the page descriptors in the work-set to non-volatile data storage drives of the data storage system, and storing, for each one of the page buffers indicated by the page descriptors in the work-set, an indication that the page buffer is available for re-use.

Abstract: Processing requests may include: receiving a write request from a host at a first node of a system; and servicing the write comprising assigning, by the first node, a sequence identifier to the write request, wherein the sequence identifier is included in a subsequence of identifiers only assignable by the first node, performing in parallel a first operation that stores first data written by the write request in a cache, a second operation that stores a descriptor for the write request in the cache, and a third operation that sends the descriptor (including the sequence identifier) to a peer node of the system; determining by the first node that the first, second and third operations have successfully completed; and responsive to determining the first, second and third operations have successfully completed, sending an acknowledgement from the first node to a host indicating successful completion of the write request.

Abstract: Techniques for providing lifecycle handling of faults associated with a storage appliance. The techniques can include recording information or details of one or more detected faults associated with a storage appliance in a fault database, analyzing the recorded fault information/details by an analytic tool, and determining whether to perform immediate recovery or defer recovery of the detected fault(s) based at least on a recommendation of the analytic tool. In this way, taking the storage appliance offline upon occurrence of a single fault can, for the most part, be avoided. In addition, because the detection of faults is not performed within a fault domain for a filesystem handling files for a single or limited number of volume families, a deduplication domain for the files of the filesystem is not restricted to the single or limited number of volume families.