In an example, Virtual Machine (VM) attributes of a first VM are received. First VM data corresponding to the first VM is to be moved from a first cluster of nodes in which the first VM data is stored. Further, a movement value of the first VM is determined based on a movement model and the VM attributes of the first VM. The movement model indicates a dependence of movement value of a VM on VM attributes of the VM. The movement value of the first VM is indicative of a rank order for movement of the first VM data among a plurality of sets of VM data to be moved, where each set of VM data corresponds to a VM.

A method for taking a storage device offline, performed by a storage system, is provided. The method includes reserving the storage device and stopping direction to increment a heartbeat counter of the storage device. The method includes verifying the heartbeat counter of the storage device is static, and performing, responsive to the verifying, an operation on the storage device, with the storage device offline relative to the storage system I/O of user data.

A method and an apparatus for data processing are provided. Interface circuitry connects the apparatus in a distributed system. Processing circuitry generates first hash values for a first group of virtual nodes to be added into a hash ring. The first group of virtual nodes represents a first newly added device in the distributed system. Then, the processing circuitry distributes the data in the distributed system according to the hash ring that includes the first group of virtual nodes. When the distribution of the data in the distribution system fails to satisfy a uniformity requirement, the processing circuitry identifies a specific virtual node in the first group of virtual nodes that causes the distribution to fail the uniformity requirement. Then, a hash value for a makeup virtual node to replace the specific virtual node is generated to update the distribution of the data in the distributed system.

Techniques perform data input and output. For example, there is provided a technique for updating a data pattern table. The technique involves: during a time period for establishing or updating the data pattern table: receiving a data write request; making statistics of a frequency of data being written; and writing data frequently used and an identification of the data to a corresponding entry of the data pattern table based on a result of the statistics. Using such a technique, a data pattern table can be customized and data storage efficiency can be improved.

A shared storage architecture persistently stores database files in non-volatile memories (NVMs) of a plurality of computing nodes of a multi-node DBMS. The computing nodes of the multi-node DBMS store data blocks in NVM and each computing node of the DBMS stores copies of each data block stored on the plurality of computing nodes. A computing node that disconnects and subsequently rejoins the DBMS employs an on-demand approach to resilvering stale data blocks that have been updated in other computing nodes in the DBMS while the computing node was offline. A data block may be resilvered on-demand based on an I/O request for a specific data block from a workload running on the reconnected computing node. Stale data blocks on the reconnected computing node are not resilvered unless they are accessed by the workload.

One or more embodiments provide techniques for accessing a memory page of a virtual machine for which loading might have been deferred, according to an embodiment of the invention, includes the steps of examining metadata of the memory page and determining that a flag in the metadata for indicating that the contents of the memory page needs to be updated is set, and updating the contents of the memory page.

A computer-implemented method, according to one embodiment, includes: detecting that a configuration change for a virtualization layer in a multi-tiered data storage system has been received, and determining whether the configuration change corresponds to a tiering adjustment. In response to determining that the configuration change corresponds to a tiering adjustment, I/O pattern changes are predicted. Moreover, a determination is made as to whether each specific extent is stored in a tier of the multi-tiered data storage system suitable to satisfy the predicted I/O pattern changes which correspond to the respective specific extent. In response to determining that one of the specific extents is stored in a tier of the multi-tiered data storage system which is not suitable to satisfy the predicted I/O pattern changes, the at least one specific extent is migrated to another one of the tiers determined as being suitable to satisfy the predicted I/O pattern changes.

A method for oversubscribing a host memory of a host running a virtual machine monitor (VMM), comprising, examining a virtual machine (VM) memory for a VM for metadata associated with the VM memory, the metadata maintained by a guest OS running on the VM, collecting the metadata for the VM memory, and managing the VM memory using the metadata for oversubscribing a host memory.

Methods and systems for improving the performance of a primary system that is running one or more virtual machines and capturing snapshots of the one or more virtual machines over time are described. The performance penalty on the primary system when a hypervisor running the one or more virtual machines is used to capture the snapshots of the one or more virtual machines may be reduced by leveraging storage array snapshots to reduce the amount of time that the hypervisor must freeze virtual disks of the one or more virtual machines. In this case, changed block tracking information for changed data blocks associated with the snapshots may be acquired from the hypervisor and the changed data blocks themselves may be pulled from the storage array snapshots without requiring the hypervisor to keep the virtual disks of the one or more virtual machines in a frozen state.

Methods and systems for improving the performance of a primary system that is running one or more virtual machines and capturing snapshots of the one or more virtual machines over time are described. The performance penalty on the primary system when a hypervisor running the one or more virtual machines is used to capture the snapshots of the one or more virtual machines may be reduced by leveraging storage array snapshots to reduce the amount of time that the hypervisor must freeze virtual disks of the one or more virtual machines. In this case, changed block tracking information for changed data blocks associated with the snapshots may be acquired from the hypervisor and the changed data blocks themselves may be pulled from the storage array snapshots without requiring the hypervisor to keep the virtual disks of the one or more virtual machines in a frozen state.