Abstract: Techniques described herein coordinate inline deduplication among nodes in a storage system. The method includes storing, in a page descriptor ring on a node, data and a fingerprint associated with the data in an entry. The method includes determining that a flushing work set (FWS) has been frozen. The node identifies, in the page descriptor ring, entries associated with the frozen FWS and having fingerprints with a parity associated with the node. The node deduplicates the entries based on a fingerprint database on the node. The node synchronizes deduplication of the frozen FWS with a peer node, so as to receive deduplication results concerning entries having fingerprints with a parity associated with the peer node. The node replaces entries in the page descriptor ring with the deduplication results from the peer node, and flushes entries in the frozen FWS to a storage device.

Abstract: Techniques described herein coordinate inline deduplication among nodes in a storage system. The method includes storing, in a page descriptor ring on a node, data and a fingerprint associated with the data in an entry. The method includes determining that a flushing work set (FWS) has been frozen. The node identifies, in the page descriptor ring, entries associated with the frozen FWS and having fingerprints with a parity associated with the node. The node deduplicates the entries based on a fingerprint database on the node. The node synchronizes deduplication of the frozen FWS with a peer node, so as to receive deduplication results concerning entries having fingerprints with a parity associated with the peer node. The node replaces entries in the page descriptor ring with the deduplication results from the peer node, and flushes entries in the frozen FWS to a storage device.

Abstract: Nodes in a storage system can autonomously ingest I/O requests and flush data to storage. First and second nodes determine a sequence separator, the sequence separator corresponding to an entry in a page descriptor ring that separates two flushing work sets (FWS). The first node receives an input/output (I/O) request and allocates a sequence identification (ID) number to the I/O request. The first node determines a FWS for the I/O request based on the sequence separator and the sequence ID number, and commits the I/O request using the sequence ID number. The I/O request and the sequence ID number are sent to the second node.

Abstract: A technique for managing a datapath of a data storage system includes receiving a request to access target data and creating a transaction that includes multiple datapath elements in a cache, where the datapath elements are used for accessing the target data. In response to detecting that one of the datapath elements is invalid, the technique further includes processing the transaction in a rescue mode. The rescue mode attempts to replace each invalid datapath element of the transaction with a valid version thereof obtained from elsewhere in the data storage system. The technique further includes committing the transaction as processed in the rescue mode.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for data compression. The method includes: comparing the size of a first data packet to be compressed with a first threshold size; if the size of the first data packet is greater than the first threshold size, determining at least two second data packets from the first data packet, wherein the size of each second data packet is less than a second threshold size; and respectively compressing the at least two second data packets. In this way, the delay of data compression can be shortened.

Abstract: A method is used for uninterrupted data flushing in a storage system. A barrier in a page descriptor ring is determined to distinguish between a filling flushing work set (FWS) and a frozen FWS. A sequence number of an I/O request is compared to the barrier. A FWS corresponding to the I/O request is identified based on the comparison. The I/O request is committed to the identified FWS.

Abstract: A method is used for uninterrupted data flushing in a storage system. An input/output (I/O) request is received. A pointer for a filling flushing work set (FWS) is atomically switching. A copy of the pointer for the filling FWS is saved, and the counter associated with a FWS referenced by the saved copy of the pointer is incremented. The saved copy of the pointer and the pointer for the filling FWS are compared, and upon determining that the saved copy of the pointer and the pointer for the filling are FWS equal, the I/O request is committed to a page descriptor ring and the counter associated with the FWS referenced by the saved copy of the pointer is decremented.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for managing a storage system. The method includes: determining multiple storage units provided by multiple storage devices, each storage unit in the multiple storage units having a storage space allocated from a first number of storage devices among the multiple storage devices; dividing the multiple storage units into at least one storage unit group based on a total number of the multiple storage devices and the first number, each storage unit group in the at least one storage unit group including a second number of storage units; and storing, based on a logical address of to-be-stored data, the to-be-stored data into the at least one storage unit group. Embodiments of the present disclosure can allocate storage resources more reasonably and thus improve the performance of a storage system for processing sequential data.

Abstract: A method is used for tracking storage utilization in a storage system. An amount of physical storage space available for use on an appliance is determined. The amount of storage space consumed by each type of data being stored on the appliance is normalized. These normalized amounts are used to determine an amount of storage space on the appliance consumed by the data. When a percentage of storage space consumed on the appliance exceeds a threshold, an alert is generated.

Abstract: Techniques provide for managing storage devices. Such techniques involve: combining storage devices in a redundant array of independent disks into a group of storage devices in an order of physical capacities of the storage devices; determining a first number of sets of storage devices based on a number of storage devices in the group and a threshold number of storage devices to be divided into each of the sets; and dividing, based on a configuration of the redundant array of independent disks, the group of storage devices into the first number of sets of storage devices, to optimize an effective capacity of the redundant array of independent disks.

Abstract: A method of validating metadata pages that map to user data in a data storage system is provided. The method includes (a) obtaining first information stored for a first metadata page and second information stored for a second metadata page, the first and second metadata pages having a relationship to each other within a hierarchy of metadata pages for accessing user data; (b) performing a consistency check between the first information and the second information, the consistency check producing a first result in response to the relationship being verified and a second result otherwise; and (c) in response to the consistency check yielding the second result, performing a corrective action to restore consistency between the first and second information. An apparatus, system, and computer program product for performing a similar method are also provided.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for upgrading a storage system. The method includes: acquiring information about a set of candidate periods related to a workload of a storage system, the storage system having a workload lower than a first predetermined threshold during the set of candidate periods; determining, based on user information of the storage system, a target period for upgrade from the set of candidate periods; and performing an upgrade operation on at least a part of components among multiple components of the storage system during the target period. In this manner, the upgrade operation for the storage system can be improved.

Abstract: Techniques for managing a storage system involve flushing a target page in a cache device to a persistent storage device of the storage system. The techniques further involve releasing a resource storing a page descriptor of the target page to a resource pool. The resource pool is configured to provide resources to store page descriptors of pages to be flushed in the cache device. The techniques further involve: if it is determined that an auxiliary descriptor of the target page is located at a tail of a queue of auxiliary descriptors of the pages to be flushed, removing the auxiliary descriptor of the target page from the queue. The auxiliary descriptors of the pages to be flushed are configured to describe the page descriptors of the pages to be flushed. Accordingly, the page flushing performance of the storage system can be improved, thereby improving the input/output performance.

Abstract: A technique for managing messaging between storage nodes of a storage system includes a first storage node delaying the sending of non-latency-critical messages to a second storage node until the first storage node has a latency-critical message to be sent. The technique further includes combining the non-latency-critical messages with the latency-critical message to form a single, aggregated message, and sending the aggregated message to the second storage node.

Abstract: A method is used for uninterrupted data flushing in a storage system. A barrier in a page descriptor ring is determined to distinguish between a filling flushing work set (FWS) and a frozen FWS. A sequence number of an I/O request is compared to the barrier. A FWS corresponding to the I/O request is identified based on the comparison. The I/O request is committed to the identified FWS.

Abstract: A method is used for uninterrupted data flushing in a storage system. An input/output (I/O) request is received. A pointer for a filling flushing work set (FWS) is atomically switching. A copy of the pointer for the filling FWS is saved, and the counter associated with a FWS referenced by the saved copy of the pointer is incremented. The saved copy of the pointer and the pointer for the filling FWS are compared, and upon determining that the saved copy of the pointer and the pointer for the filling are FWS equal, the I/O request is committed to a page descriptor ring and the counter associated with the FWS referenced by the saved copy of the pointer is decremented.

Abstract: Rotating disk drives can be partitioned into performance partitions (PPs) each including slices having similar performance in accordance with an allowable difference or tolerance for any of data throughput, bandwidth and data transfer rate. Each subdrive can correspond to slices of the rotating disk drives included in a single PP. Each PP can have a performance level denoting a relative performance ranking of PPs. Each PP tier can include PPs across multiple rotating disk drives have the same performance level. Resiliency sets (RSs) can be configured from the PP tiers where each RS can include slices of subdrives from only one of the PP tiers. Physical ubers or units (PUs) can be allocated from the RSs. Each PU can denote a mapped RAID group configured from slices of different subdrives of one RS. Spare slices can be reserved in each RS or a lowest performance PP tier.

Abstract: Techniques performed by a computing device of storing data in a data storage system are provided. A method includes (a) storing references to write commands within entries of a first chained hash table (CHT), the first CHT being pointed to by a first data structure representative of a logical disk; (b) keeping track of a load factor of the first CHT during operation; and (c) in response to determining that the load factor of the first CHT has transitioned outside of predetermined bounds: (1) creating a second CHT and a second data structure representative of the logical disk, the second CHT being pointed to by the second data structure; (2) linking the second data structure to the first data structure via a linked list; and (3) storing references to new write commands directed at the logical disk within entries of the second CHT rather than the first CHT.

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: Techniques manage a redundant array of independent disks. In such a technique, a response time of a first storage device in the RAID is compared to a first threshold. In response to the response time of the first storage device exceeding the first threshold, the first storage device is configured as a pseudo-degraded storage device, such that the pseudo-degraded storage device is responsive to write requests only.