Abstract: Load balancing using predictive VM-based analytics is disclosed, including: determining a plurality of storage device specific predicted metric data structures corresponding to respective ones of a plurality of storage devices; and combining the plurality of storage device specific predicted metric data structures corresponding to respective ones of the plurality of storage devices into a combined predicted metric data structure.

Abstract: Dynamic block optimization for space and performance is disclosed, including: determining that a data pattern associated with a data block included in a write request matches a promoted data pattern; and performing the write request by associating the data block with a previously stored copy of the data block without updating an associated reference count. Dynamic block optimization for space and performance further includes determining that a data pattern associated with a data block included in a write request matches a predetermined data pattern; and performing the write request by storing a static representation associated with the data block and not storing the data block, wherein the static representation is usable to generate the data block.

Abstract: Modeling space consumption of a migrated VM is disclosed, including: obtaining aggregated effective historical space usage data associated with a plurality of VMs at a storage device; using the aggregated effective historical space usage data to generate one or more predicted metrics of space usage at the storage device; and using the one or more predicted metrics of space usage to potentially migrate a VM with respect to the storage device.

Abstract: Access rate prediction in a hybrid storage device is disclosed, including: using a first tier type of storage access counts to model an access distribution across a first tier type of storage and a second tier type of storage at a storage device; and generating a predicted first tier type of storage hit rate that is predicted to result from moving an existing VM off of the storage device using access statistical information associated with the existing VM and the access distribution at the storage device.

Abstract: Detecting and pinpointing data corruption is disclosed, including: storing an object-level metadata structure corresponding to a stored object, wherein the stored object comprises a plurality of blocks; and determining for a block included in the plurality of blocks, based at least in part on a piece of identifying information of the block, two or more locations in the object-level metadata structure at which to store a value computed based at least in part on data comprising the block.

Abstract: Storage system accelerated techniques to distribute and run containerized applications are disclosed. In various embodiments, an indication to run a container is received. A snapshot identifier comprising an image data associated with the container is used to create a clone based on a snapshot with which the snapshot identifier is associated. The clone is used to store data associated with at least a topmost layer of the container.

Abstract: Dynamic RAID expansion is disclosed, including: incrementally expanding a redundant storage system comprising a plurality of storage drives by adding one or more storage drives to the plurality of storage drives; and storing, for each stripe included in the redundant storage system, metadata indicating which storage drives in the redundant storage system store valid data for the stripe.

Abstract: Techniques to satisfy quality of service (QoS) requirements on a per virtual machine basis natively in a storage system are disclosed. In various embodiments, for each of a plurality of virtual machines a corresponding input/output (I/O) request queue is stored on the storage system. Requests are scheduled to be pulled from the respective request queues and added to a corresponding storage operation pipeline at the storage system in an order determined based at least in part on a per virtual machine quality of service parameter.

Abstract: Efficient deletion of a shared snapshot and other workflows are described herein, including: determining to delete a shared snapshot associated with a first container, wherein the shared snapshot shares with a second container at least a subset of data values that are stored by the shared snapshot, the determination is based at least in part on the shared snapshot sharing data values with not more than a prescribed number of containers other than the second container and an occurrence of a deletion triggering event; and causing zero or more data values that are stored by the shared snapshot and shared with the second container to be associated with the second container prior to deleting the shared snapshot.

Abstract: Techniques to automatically allocate resources among storage system resource consumers are disclosed. In various embodiments, for each of a plurality of heterogeneous categories of resource consumer of the storage system a corresponding allocated portion of storage system resources to be available for use to perform operations associated with that category of resource consumer is determined. A storage system is configured automatically to provide access to resources of the storage system based at least in part on the respective portions of storage system resources allocated to each category of resource consumer, including by ensuring availability to workloads associated with each category at any given time at least a minimum amount of storage system resources corresponding to the portion of storage system resources allocated to that category.