Abstract: Based on a system configuration change (e.g., of a Decentralized, or Distributed, Agreement Protocol (DAP)) within a dispersed storage network (DSN), a computing device identifies a reallocating encoded data slice (EDS) number that is no more than a pillar width minus a performance threshold. The computing device then directs storage units (SUs) to update system configuration of the DAP (e.g., from a first to a second system configuration) by throttling and controlling the number of SUs permitted to update at a time. For example, the computing device permits no more than the reallocating EDS number of SUs to perform simultaneous (or substantially or approximately simultaneous) update of the system configuration of the DAP. The computing device also directs the SUs to operate based on the first system configuration before the condition(s) is/are met and then to operate based on the second system configuration after the condition(s) is/are met.

Abstract: Mechanisms for controlling access to storage volumes on the secondary storage system is provided. A determination is made as to whether a first site computing device has sent a notification of a failure condition of a first site. In response to a determination that the notification of the failure condition of the first site has not been received, secondary workloads of a second site computing device are permitted to access storage volumes on the secondary storage system. In response to a determination that the notification of the failure condition of the first site has been received, a mode of operation of the second site is modified from a normal mode of operation to a failure mode of operation. In the failure mode of operation, the storage system controller of the second site blocks at least a portion of access requests from secondary workloads of the second site computing device.

Abstract: A server-based desktop-virtual machines architecture may be extended to a client machine. In one embodiment, a user desktop is remotely accessed from a client system. The remote desktop is generated by a first virtual machine running on a server system, which may comprise one or more server computers. During execution of the first virtual machine, writes to a corresponding virtual disk are directed to a delta disk file or redo log. A copy of the virtual disk is created on the client system. When a user decides to “check out” his or her desktop, the first virtual machine is terminated (if it is running) and a copy of the delta disk is created on the client system. Once the delta disk is present on the client system, a second virtual machine can be started on the client system using the virtual disk and delta disk to provide local access to the user's desktop at the client system. This allows the user to then access his or her desktop without being connected to a network.

Abstract: According to one embodiment, an electronic device connectable to a host via an interface includes a nonvolatile memory and a controller electrically connected to the nonvolatile memory and capable of processing commands issued by the host in parallel. When the electronic device is connected to the host, the controller determines, when one or more commands to be processed by one or more deadline times, respectively, are issued by the host, scheduling indicative of timings at which the one or more commands are processed, respectively, based on the one or more deadline times. The controller performs processing corresponding to the one or more commands in accordance with the scheduling.

Abstract: A method and apparatus for incremental RAID stripe update parity calculations. The method includes: receiving, at a first set of solid state drives, a last portion of a redundant array of independent disks (RAID) stripe among multiple portions of the RAID stripe, wherein the RAID stripe includes multiple shards, and wherein each previous portion of the RAID stripe is written to the first set of solid state drives; calculating a current parity value based on the last portion of the RAID stripe and a previous parity value updated after receiving each previous portion of the RAID stripe; and responsive to receiving all portions of a shard of the RAID stripe, copying the shard of the RAID stripe from the first set of solid state drives to a second set of solid state drives.

Abstract: Managing a redundant array of independent disks (RAID) storage array involves assigning first and second stripes to span respective first and second sets of disks. A subset of drives in the first set fails such that the first stripe is in a first state wherein a failure of another drive in the first set will lead to data loss in the first stripe. It is determined that the second stripe is in a fault-tolerant state such that the second stripe can have failures of two drives in the second set before the second stripe is in the first state, Part of an operational disk of the second set used by the second stripe is allocated to the first stripe to replace at least part of the subset of failed drives.

Abstract: Techniques are described for clustering data at the point of ingestion for storage using scalable storage resources. The clustering techniques described herein are used to cluster time series data in a manner such that data that is likely to be queried together is localized to a same partition, or to a minimal set of partitions if the data set is large, where the partitions are mapped to physical storage resources where the data is to be stored for subsequent processing. Among other benefits, the clustered storage of the data at the physical storage resources can reduce an amount of data that needs to be filtered by many types of queries, thereby improving the performance of any applications or processes that rely on querying the data.

Abstract: Synchronizing metadata among storage systems synchronously replicating a dataset, where synchronizing the metadata includes: receiving, at a first storage system of the storage systems, an I/O operation directed to the dataset; determining, in dependence upon the I/O operation, a metadata update describing a mapping of segments of content to a virtual address within a storage object, wherein the storage object includes the dataset; and synchronizing metadata on a second storage system of the storage systems by sending the metadata update to the second storage system to update a metadata representation on the second storage system in accordance with the metadata update.

Abstract: A method of operating a remote procedure call cache in a storage cluster is provided. The method includes receiving a remote procedure call at a first storage node having solid-state memory and writing information, relating to the remote procedure call, to a remote procedure call cache of the first storage node. The method includes mirroring the remote procedure call cache of the first storage node in a mirrored remote procedure call cache of a second storage node. A plurality of storage nodes and a storage cluster are also provided.

Abstract: An apparatus may comprise a port to couple to a plurality of communication paths comprising a first, second, third, and fourth communication path, the communication paths each having a direction of either a receive path or a transmit path, the communication paths comprising a first communication path having a direction that may be selectively configured as a receive path or a transmit path. The apparatus further comprises a controller comprising circuitry, the controller to in response to a communication path reconfiguration command, reconfigure the direction of the first communication path such that the first communication path, second communication path, and third communication path have the same direction and the fourth communication path has a direction opposite to the direction of the first, second, and third communication paths.

Abstract: A multi-layer rate sizing stack may generate prescriptive operation-rate tokens for controlling sizing adjustments for operation-rates. The input layer of the rate sizing stack may generate operation pattern data based on operation-rate data received via network connection. A prescriptive engine layer may determine a prescriptive allowed operation-rate based on the operation pattern data. Based on the prescriptive allowed operation-rate, the prescriptive engine layer may generate the operation-rate tokens that that may be used to control rate sizing adjustments prescriptively.

Abstract: In an embodiment of the invention, a method comprises: A method, comprising: issuing, by a Direct Memory Access (DMA) engine, an update request to a dependency table if the DMA engine has finished executing a first descriptor; and issuing, by the DMA engine, a monitoring request if the DMA engine is executing a second descriptor that depends on a completion of a data transfer so that the DMA engine can monitor a status of a selected subindex related to the data transfer, wherein the subindex is in the dependency table.

Abstract: Embodiments include a multi-stream storage device, a system including a multi-stream storage device, and a method, comprising: receiving an access to a multi-stream storage device; converting at least one parameter of the access into a stream identifier; and accessing the multi-stream storage device using the stream identifier.

Abstract: A method for use in a distributed storage network (DSN) includes capturing data associated with a user device, which includes a collection module, and determining whether to archive captured data in the DSN based on information associated with the captured data. In response to a determination to archive the captured data, the captured data is processed to produce a representation of the captured data. Operational parameters associated with the DSN are determined, and the representation of the captured data is transmitted to the DSN for storage in a DSN memory.

Abstract: A method for reclaiming storage space in RAID arrays made up of heterogeneous storage drives is disclosed. In one embodiment, such a method includes determining a most common storage capacity for a set of storage drives utilized in a storage system. The method further identifies physical storage drives from the set that contain unused storage space. The method pools the unused storage space of the physical storage drives to create virtual storage drives with storage capacities substantially equal to the most common storage capacity. The method then utilizes the virtual storage drives in existing or new RAID arrays. A corresponding system and computer program product are also disclosed.

Abstract: A system and a method of managing a plurality of storage tiers, may include: receiving a write-access request, including at least one data element for storage, a logical address associated with the data element and a tier identifier for a selected tier; computing a unique reference name based on the content of the data element, and associating the unique reference name with the logical address; using a global API to generate a first write-access command, that includes the unique reference name and the data element; customizing the write-access command according to an addressing scheme of the selected tier by a dedicated driver, to obtain at least one second write-access command; using the second write-access command to associate the unique reference name with at least one physical location on the selected tier; and storing the at least one data element at the associated at least one physical location.

Abstract: Embodiments of the present disclosure provide a method, a system and a computer readable storage medium for data backup and restoration in a cluster system. The method comprises obtaining performance data of the nodes, the performance data being collected periodically by the backup and restore proxy modules; predicting workloads of the nodes based on the performance data; selecting, based on the predicted workloads, a backup node from the nodes, to take over a data backup and restore task from a working node which is currently in charge of the data backup and restore task.

Abstract: Examples include translation of a source m-node identifier (ID) to a target m-node ID, the source m-node ID identifying a source m-node representing a file system object in a source file system, and the source m-node ID including a source replication set ID, a source m-node container ID, and at least one local ID. In examples, a source m-node container ID is replaced with a target m-node container ID in the target m-node ID.

Abstract: Embodiments of the present disclosure relate to a method and device for flushing pages from a solid-state storage device. Specifically, the present disclosure discloses a method of flushing pages from a solid-state storage device comprising: determining a first number based on a period length of one flushing cycle and a period length required for building one flushing transaction, the first number indicating a maximum number of flushing transactions that can be built in the flushing cycle; and flushing pages from the solid-state storage device with an upper limit of the first number in the flushing cycle. The present disclosure also discloses a device for flushing pages from a solid-state storage device and a computer program product for implementing steps of a method of flushing pages from a solid-state storage device.

Abstract: One embodiment provides a computer system. The computer system comprises: a plurality of storage devices; and a first component functioning both as a network interface card and as an access switch, wherein the first component is configured to manage connections to the plurality of storage devices. A respective storage device comprises: an Ethernet port coupled to the first component; at least one microprocessor; a plurality of PCIe lanes; and a plurality of storage drives with non-volatile memory.

Abstract: A method is provided. The method includes receiving a plurality of data streams at a computing device, processing first event data of a first data stream of the plurality of data streams in accordance with a first key, and determining whether second event data of a second data stream of the plurality of data streams is stored in a cache memory in accordance with the first key. The second event data is to be joined with the first event data. The method further includes performing a join operation of the first event data and the second event data at least partially in view of whether the second event data is stored in the cache memory.

Abstract: An uneven distributed storage across a mesh fabric storage system may include receiving storage operations from one or more client devices and/or applications contemporaneously with receiving availability messaging from a set of multiple storage devices that may be of the same or different types. One or more of the storage operations may be assigned to a storage device that has signaled its readiness to perform the one or more storage operations via an issued availability message. Each storage device may thereby perform a subset of the collective set of storage operations with the uneven distribution allocating load that is directly commensurate with the performance of each storage device. Stored data may be moved between storage devices using a similar availability-driven methodology so as to reallocate capacity usage while still providing the fastest storage performance associated with all storage devices writing the data as it is generated.

Abstract: Multiple IHSs (Information Handling Systems) may be installed as components of a chassis that has access to a plurality of storage devices via a chassis management controller. An IHS requests configuration of a virtual storage profile, such as a RAID configuration. A remote access controller of the IHS determines physical storage requirements for implementing the requested virtual storage profile. Based on the physical storage requirements, the chassis management controller selects storage devices from idle storage devices mapped to one of the storage controllers installed in one of the IHSs supported by a chassis management controller. The selected storage devices are mapped to the storage controller and used to implement the virtual storage profile. The chassis management controller manages a global pool of spares from the idle storage device for virtual storage profiles supported by the supported storage controllers.

Abstract: Techniques for processing data may include: determining a first amount denoting an amount of write pending data stored in cache to be redirected through storage class memory (SCM) when destaging cached write pending data from the cache; performing first processing that destages write pending data from the cache, the first processing including: selecting, in accordance with the first amount, a first portion of write pending data that is destaged from the cache and stored in the SCM and a second portion of write pending data that is destaged directly from the cache and stored on one or more physical storage devices providing back-end non-volatile physical storage; and subsequent to storing the first portion of write pending data to the SCM, transferring the first portion of write pending data from the SCM to the one or more physical storage devices providing back-end non-volatile physical storage.

Abstract: A distributed storage schemes manages implementation of QoS targets for IOPs across compute nodes executing applications, primary storage nodes storing a primary copy of a logical storage volume, and clone storage nodes. On the compute node, a maximum priority is assigned to a minimum number of IOPs in a queue within a time window from a time of receipt of a last unexecuted IOP. Other IOPs are assigned a minimum priority. On the storage node, maximum priority IOPs are assigned to high priority queues, from which IOPs are executed first, and low priority IOPs are assigned to low priority queues. Methods for determining the capacity of storage nodes and allocating storage requests are also disclosed.

Abstract: A doubly mapped redundant array of independent nodes (doubly mapped RAIN) for data storage is disclosed. A doubly mapped RAIN cluster can be allocated on top of one or more real data clusters. The real data cluster can be N nodes by M disks by L extents. A doubly mapped RAIN cluster can be N? nodes by M? disks by L? extents, where N? is less than, or equal to N. Mapping of data storage locations in a doubly mapped RAIN cluster can facilitate use of a real cluster at a different granularity than in either a conventionally administered real cluster or in a singly mapped RAIN. The topology of a doubly mapped RAIN can be constrained via rules related to correlation of real disks to mapped nodes, correlation of real extents to mapped nodes, correlation or real extents to mapped disks, etc.

Abstract: A method for improving I/O response times in a data replication environment is disclosed. In one embodiment, such a method includes receiving, at a primary storage system, a request to read data on the primary storage system. The method starts a timer upon receiving the request. In the event the data cannot be retrieved from the primary storage system by the time the timer expires, the method requests the data from a secondary storage system. In the event the primary storage system returns the data before the secondary storage system, the method returns data retrieved from the primary storage system to a host system. In the event the secondary storage system returns the data before the primary storage system, the method returns the data retrieved from the secondary storage system to the host system. A corresponding system and computer program product are also disclosed.

Abstract: Disclosed are a semiconductor memory device, a memory system including the semiconductor memory device, and a method of operating the same. The memory system includes a controller receiving a host command and program data from a host, queueing a command in response to the host command, and generating grouping information by grouping a plurality of program states into two or more program groups, and a semiconductor memory device receiving a queued command, the program data, and the grouping information from the controller, performing a program operation, and sequentially programming the two or more program groups on the basis of the grouping information.

Abstract: A disclosed method for managing a RAID (redundant array of independent devices) file system memory comprises coalescing small read and write chunks into reasonably sized coalesced chunks up to 1 MB without any small size I/Os. The method also includes creating a new mapping layer configured to map the small size chunks in the coalesced chunks. The method additionally comprises flushing the coalesced chunks by allocating large chunks based on a chunk size coalescence or on a check point hit in terms of a SPA Sync. The method may also include applying a dynamic parity algorithm on the coalesced chunks to generate parities and data fragmented columns in reasonably big sized chunks. The method may yet include creating an appropriate coalesced chunk header to maintain the new mapping layer. The method may still comprise generating a chunked block pointer and a regular block pointer configured to co-exist in dual process threads.

Abstract: A reconstruction method based on a hard disk array comprises determining whether at least one damaged hard disk among a plurality of hard disks of the hard disk array has been replaced by at least one to-be-reconstructed hard disk, and writing a backup data corresponding to the at least one damaged hard disk into the at least one to-be-reconstructed hard disk according to a mapping table and information of the at least one damaged hard disk.

Abstract: Embodiments of the present disclosure relate to a method for managing a storage system, an electronic device, a storage system and a computer program product. The method includes: expanding a second number of existing disks with a first number of expanded disks to rebuild a redundant array of independent disks (RAID) having a predetermined extent width, the RAID being built based on the existing disks, and the predetermined extent width indicating a number of disk extents from different disks that form an RAID extent. The method further includes: determining a third number of RAID extents contained in the rebuilt RAID. The method further includes: determining a count of data movements to be performed between the expanding disks and the existing disks during the rebuilding based on the first number, the second number, the third number and the predetermined extent width. With this method, time for estimating count of data movements may be reduced.

Abstract: A storage system comprises a plurality of enclosures and a storage controller. Each enclosure comprises at least one processing device and a plurality of drives configured in accordance with a redundant array of independent disks (RAID) arrangement. The storage controller obtains data pages associated with at least one input-output request and provides the obtained data pages to the at least one processing device of a given enclosure. The storage controller issues a command to the at least one processing device of the given enclosure to perform at least one operation based at least in part on the obtained data pages. The at least one processing device of the given enclosure receives the obtained data pages from the storage controller, calculates RAID parities based at least in part on the received data pages, and stores the data pages and RAID parities on the plurality of drives according to the RAID arrangement.

Abstract: Methods, systems, and other aspects for reconstructing data and rebuilding a failed storage device in a storage system using one or more functioning compute resources and/or storage resources of the failed storage device. For example, a method may include, responsive to a detection of a failed storage device in a storage system, locating data and redundancy information in functioning storage device(s) in the storage system for reconstructing data of the failed storage device; issuing peer-to-peer commands to the functioning storage device(s) to obtain the data and the redundancy information from the functioning storage device(s); and reconstructing the data of the failed storage device based on the data and the redundancy information obtained from the functioning storage device(s), wherein a functioning compute resource of the failed computing device at least partially performs one or more of the locating, issuing, and reconstructing.

Abstract: Techniques involve receiving a write request for a first storage block in a source storage area during copying first data from the source storage area to a target storage area. The techniques further involve writing second data associated with the write request into a second storage block of the target storage area, wherein the second storage block corresponds to the first storage block. The techniques further involve updating an indicator associated with the first storage block, wherein the indicator indicates that the second data is written by the write request during the copying. Such techniques enable the data to be written into the target storage area directly without being written into the source storage area by setting an indicator indicating whether a storage block receives a write request during data copying. Accordingly, such techniques can improve write performance during data copying, and extend source storage area lifetime.

Abstract: Methods and apparatus for allocating logical sectors and bands to store data on interlaced magnetic recording tracks. The systems and methods include formatting a data storage medium to include a plurality of bands, each band of the plurality of bands including a plurality of tracks, the plurality of tracks including a subset of top tracks interlaced with a subset of bottom tracks, and each track of the plurality of tracks including a number of sectors, formatting a first band of the plurality of bands, determining an isolation region of the first band, and formatting a second band of the plurality of bands responsive to determining the isolation region of the first band.

Abstract: A method for execution by a dispersed storage and task (DST) processing unit includes determining a first optimal slice size requirement is determined for a first independent data element (IDE). The first IDE is split into a first plurality of IDEs based on determining the first IDE compares unfavorably to the first optimal slice size requirement. An error coding function is performed on the first plurality of IDEs to produce a set of encoded slices for each of the first plurality of IDEs for transmission to storage units. A subset of additional IDEs are selected to be merged into a merged IDE based on a second optimal slice size requirement, and continuous content of the subset of additional IDEs is concatenated to produce the merged IDE. The error coding function is performed on the merged IDE to produce a second set of encoded slices for transmission to storage units.

Abstract: A semiconductor apparatus may include technology to identify two or more types of storage controller traffic, direct a first identified type of storage controller traffic along a fixed processing path, and direct a second type of storage controller traffic along a programmable processing path. Other embodiments are disclosed and claimed.

Abstract: The technology disclosed herein enables a group of clients to concurrently access data of a distributed storage system over multiple paths without including a client portion of the distributed storage system. An example method may include: determining, by a processing device, a portal group comprising a plurality of network portals for accessing a storage unit; transmitting data of the portal group to a first client and to a second client, wherein data transmitted to the first client indicates a first network portal is preferred and wherein data transmitted to the second client indicates a second network portal is preferred; and providing access for the first client to the storage unit using a storage session, the storage session providing the first client multiple paths to access the storage unit, wherein one of the multiple paths comprises the first network portal.

Abstract: The presently disclosed subject matter includes various inventive aspects, which are directed to protecting data stored in a data storage system using a RAID protection scheme, which enables, responsive to a storage device failure, to execute a rebuild process that uses a total number of storage devices which is greater than the number of segments in a RAID stripe.

Abstract: A distributed storage system includes multiple partitions. A block received for storage is striped across the partitions if it meets predetermined criteria. Multiple tables are maintained for indexing the blocks and the subblocks of blocks that are partitioned. The tables for the subblocks and the tables for the corresponding blocks of the subblocks are updated to include metadata for the subblocks for retrieving the subblocks in response to a request for a corresponding stored block.

Abstract: A virtual disk is instantiated as a representation of a storage volume. The virtual disk is configured with metadata corresponding to the storage volume. A storage stack is instantiated that is operable to provide an active I/O path to the storage volume. The storage stack is modified by adding an alternative I/O path to the virtual disk. The alternative I/O path includes a layout driver configured to manage a converted storage layout for the storage volume. The storage volume is encapsulated to the virtual disk. The active I/O path is closed and the alternative I/O path is enabled for I/O requests to the virtual disk. The storage layout of the storage device is converted without taking the storage volume offline.

Abstract: Embodiments of the present application provide a method for reading and writing video data in a NAS device. The NAS device is mounted to a block interface of a storage server and includes a plurality of data files, and the correspondence between disk block numbers of block devices and filenames of the data files in the NAS device is stored in the index server. In the present application, a reading or writing request sent by the client device is received, a filename of a data file which data is to be read from or written into is located, and then a reading or writing operation is performed in the data file corresponding to the determined filename in the NAS device. In the present application, a reading or writing operation on a block device is transformed into a reading or writing operation on a data file in a NAS device so that a video cloud storage system can support the NAS device.

Abstract: A computer-implemented method for writing data to a first media using a set of data structures to reduce potential errors when reading the data from the first media is described. The method includes writing, user data to a set of memory cells in the first media; and storing, in response to writing the user data to the set of memory cells, a first set of parity bits associated with the user data in a first buffer that is held within a second media separate from the first media and is a different type than the first media, wherein the first set of parity bits provide error correction information for correcting errors introduced to the user data while stored in the set of memory cells or read from the set of memory cells.

Abstract: A method and a manager for managing a storage system including a manager and a storage device. The storage device includes a data region and a metadata region. The data region is divided into data blocks. The metadata region stores metadata describing zeroing states of the data blocks. The method comprises allocating a metadata cache in a memory of the manager. The metadata cache includes respective zeroing indication bits indicative of the zeroing states of the corresponding data blocks. The allocating procedure comprises allocating a user data cache for reading or writing user data and allocating a background zeroing cache for a background zeroing operation of the storage device. The method further comprises, in response to receiving an I/O request for the storage system, processing the I/O request with the metadata cache.

Abstract: Implementations disclosed herein include a storage system including a plurality of storage devices, wherein each of the plurality of storage devices includes compute resources, memory resources, and a storage device controller configured to perform application-specific data management operations using the compute resources and the memory resources of the storage device, and a storage system controller configured to distribute a workload across the plurality of storage devices based on a capability of each of the plurality of storage devices during an IDLE state. The capability of each of the plurality of storage devices may be specifications, current availability, and performance history of each of the plurality of storage devices. In some implementations, each of the plurality of storage devices communicate with each other via a peer-to-peer networking communications protocol (e.g., NVme, NVMof, PCIe, Ethernet, etc.).

Abstract: A “snapshot-control media agent” (“snap-MA”) is configured on a secondary storage computing device that operates apart from client computing devices. A “forwarder” media agent operates on each client computing device that uses a storage array, yet lacks command channels to the storage array. The forwarder media agent forwards commands directed at the storage array to the snap-MA, which receives and processes commands directed at the storage array that were forwarded by the forwarder media agents. Responses from the storage array are transmitted to the respective forwarder media agent. The snap-MA pools storage-array-command devices to avoid capacity limitations in regard to communications channels at the storage array.

Abstract: The disclosure generally describes computer-implemented methods, software, and systems for accessing volumes of data records structured to include sets dimensions, each dimension labelled in a manner specific to respective entities; identifying candidates data records keyed by managed keys that span a subset of dimensions even though at least one dimension from the subset of dimensions is labelled differently between the different volumes; comparing the candidate data records from the different volumes to determine whether a particular managed key is valid based on contents of the candidate data records from the different volumes; in response to determining that the particular managed key is valid, combining the candidate data records keyed by the valid managed key to be merged and accessible as one continuous entry; and in response to determining that the particular managed key is invalid, combining the candidate data records from the different volumes as separate entries.

Abstract: A technique for operating a file system in a NAS (network attached storage) cluster deploys a file system on a thin LUN in block storage and unilaterally provisions storage space to the file system without extending provisioning activities to the LUN. Rather, the file system proceeds as if the storage space is available from the LUN without reserving space or confirming availability.

Abstract: In one embodiment, virtual storage drives are allocated to RAID arrays so that no two virtual storage drives of a RAID array are mapped to the same physical storage drive. In another aspect, error handling routines are limited to virtual storage drives impacted by an error in a physical storage drive so that virtual storage drives of the physical storage drive not impacted by the error are bypassed. In yet another aspect, cache operations to a target virtual storage drive may be throttled as a function of both a limit imposed on cache operations directed to the RAID array to which the virtual storage drive is allocated, and a separate limit on cache operations directed to a group of virtual storage drives which are mapped to the same physical storage drive as the target virtual storage drive. Other features and aspects may be realized, depending upon the particular application.

Abstract: A system and method for dynamic RAID geometries. A computer system comprises client computers and data storage arrays coupled to one another via a network. A data storage array utilizes solid-state drives and Flash memory cells for data storage. A storage controller within a data storage array is configured to configure a first subset of the storage devices for use in a first RAID layout, the first RAID layout including a first set of redundant data. The controller further configures a second subset of the storage devices for use in a second RAID layout, the second RAID layout including a second set of redundant data. Additionally, the controller configure an additional device not included in either the first subset or the second subset to store redundant data for both the first RAID layout and the second RAID layout.