36 results about "Volume (computing)" patented technology

The invention provides a method and a device for quickly verifying the consistency of multi-duplicate storage. The method includes S1, judging data writing volumes of head files when systems are in idle states, creating snapshot files by the aid of I/O (input/output) redirection technologies when the volumes of data written into the head files reach 30% of the volumes of integral spare files at least, utilizing new head files with continuously written data as null comparison sparse files, computing hash values of the snapshot files and storing the hash values of the snapshot files as extended attributes; S2, carrying out snapshot operation on the head files once at intervals of preset time and comparing hash values of the new head files to the hash values of the snapshot files when the consistency of multi-duplicate data need to be timely verified. The method and the device have the advantages that characteristics of distributed block storage are combined with one another, the method for quickly detecting the consistency of multiple duplicates on the basis of snapshot technologies is proposed, hash value computation tasks are divided to be carried out in multiple time slots, accordingly, influence on normal reading and writing tasks of the systems can be prevented, and the multi-duplicate storage consistency verification speeds can be increased.

A file system executing on a computing system is configured to efficiently manage and store data on a tiered storage volume using hybrid media that includes a shingled magnetic recording (SMR) data store, such as an SMR disk drive, and a randomly-writable data store, such as a conventional magnetic recording (CMR) disk or solid-state drive (SSD) using flash memory. Write operations in the SMR diskare performed in a sequential manner to optimize storage density on the disk. The file system utilizes logic incorporated in an allocator that monitors I/O operations on the computing system, for example, write requests from applications executing on the system - to determine if a request is associated with file data or metadata. If the request is for metadata, then the allocator allocates spaceon the SSD/CMR for the metadata, and if the request is for file data, then space is allocated on the SMR disk.

A method, computer program product, and computer system for creating, by a computing device, a child leaf of a snapshot associated with a parent leaf of a volume. At least one entry associated with one of the parent leaf and the child leaf may be added to a delta log. Only delta changes are written to the child leaf. One of a first destaging technique and a second destaging technique may be used for the child leaf.

The invention discloses a memory snapshot method and a device of a cloud computing platform virtual machine based on a storage system, and a computer readable storage medium. A memory synchronizationinterface matched with each virtual machine operating system is constructed in advance; wherein the memory synchronization interface is used for synchronizing memory data at a preset moment to a corresponding storage disk, and the method can comprise the following steps: when a target virtual machine snapshot creation operation instruction issued by a cloud platform is received, obtaining storageresources which are mounted on a target virtual machine and comprise a disk and a storage volume corresponding to the disk on a storage system; sending an instruction of interrupting the storage resource read-write request, and meanwhile, calling a memory synchronization interface matched with the target virtual machine to synchronize the memory data into the disk; based on the corresponding relation between the disk and the storage system, calling the snapshot interface of the storage system to carry out snapshot operation on the disk, so that the memory snapshot of the virtual machine is efficiently realized on the basis of not occupying IO resources of the virtual machine and IO resources of a cloud platform.

Virtual computing instance data that are stored across multiple storage volumes are replicated in a manner such that the write order is maintained. The frequency of the replication is set so that the recovery point objective defined for the VM data can be satisfied. The replication includes the steps of determining a set of logical storage volumes across which the virtual computing instance issues dependent write IOs, issuing a first command to the virtual computing instance to block new IOs and to block receipt of IO acknowledgements, issuing a command to create replicas of all the logical storage volumes in the set, and then issuing a second command to the virtual computing instance to unblock new IOs and unblock receipt of IO acknowledgements.

The invention discloses a container-based distributed computing method and device, and the method comprises the following steps in a Shuffle stage: calling a bottom driver to initialize a persistent memory connected to a host machine, determining an equipment application mode for the persistent memory, and creating an area and a namespace on the host machine; creating a data volume with a file system for the persistent memory based on the region and the namespace, and mounting the data volume to a host machine to allow the container to access the file system through a container storage interface; and monitoring the Shuffle management interface, determining Shuffle data according to a dependency relationship between the elastic distributed data sets output by the management interface, and accessing the data volume from the container through the container storage interface so as to overwrite and/or cache the Shuffle data to the file system. According to the method, the processing amountof the Shuffle data can be reduced, the reading and writing cache speed of the Shuffle data in the container can be increased, and then the Shuffle efficiency of distributed computing is improved.

The invention discloses a tgt-based cluster handle management method, system and device and a computer readable storage medium, and the method comprises the steps: adding one to a cluster reference count in a cluster handle structure body when the cluster handle structure body of a storage cluster is created; when the storage pool structure body of the storage pool corresponding to the LUN creation information is created, adding one to the reference count of the storage pool in the storage pool structure body; when the volume structure body of the volume corresponding to the LUN creation information is created, adding one to the volume reference count in the volume structure body. According to the invention, the corresponding reference count is added in the structure body and used for recording the number of LUNs corresponding to a storage cluster, a storage pool and a volume; the number of LUNs is counted through reference counting, the LUNs are managed through the handle informationand the context information, one tgt thread can manage all the LUNs only by establishing one cluster handle structure body, the load capacity of a cluster is improved, and consumption of computing resources is reduced.

The invention discloses a theater data accurate query optimization method based on HBase. The theater data accurate query optimization method comprises the following steps: 1) accessing the HBase by using C + + through Thrift; (2) generating a file writing interface by using Thrift, and putting an HBase. Thritf file in an HBase source code under a main directory of a root user; (3) accessing the HBase in the program; the method comprises the following steps: starting the service of HBASETHRIF2service at a server side, calling an API (Application Program Interface) to write an interface program by referring to a function definition in a generated file, applying for accessing a database from the server by a client side, and calling a Thrift function to open the connection of the database; 4) adopting a PARTITION algorithm to carry out direct connection operation, and reducing data processing time; and 5) performing computation-intensive and large-data-volume operation by using a Spark distributed computing platform, and realizing efficient query of data through the iterative computation advantage of the Spark platform.

The invention discloses a method for dynamically creating a local storage volume, relates to the technical field of cloud computing, and solves the technical problem that a plurality of storage pools cannot be supported to perform local storage according to different nodes and different storage performance. The method comprises the following steps: step 1, creating a local storage provider to provide local storage using each node for a k8s user, it is guaranteed that a pod copy is operated on each of part of or all of the nodes; step 2, a service account (ServiceAccount), a role binding (ClusterRole Binding) and a role specification (ClusterRole) of the provisor are generated, and the service account (ServiceAccount), the role binding (ClusterRole) and the role specification (ClusterRole) of the provisor are generated; step 3, creating a hostpath storage class StorageClass to dynamically generate a storage volume, configuring a provider parameter of the storage class to be example.io/hostpath, and configuring a VolumeBindingMode binding mode of the storage class to be a WaitForFirstConsumer, and configuring a VolumeBindingMode binding mode of the hostpath to be a VolumeBindingMode binding mode of the hostpath to be a VolumeBindingMode binding mode of the hostpath to be a VolumeBindingMode binding mode of the hostpath to be a VolumeBindingMode binding mode of the hostpath; and step 4, creating a test-pvc and binding a hostpath storage class, and requesting a specific storage class by the pvc through the Storage ClassName. The invention further discloses a local storage volume dynamic creation device.

The invention discloses an imported file checking method, a checking system and computing equipment. The method is executed in an operating system, a shared data volume and a container connected with the shared data volume are arranged on the operating system, and the method comprises the steps of: detecting whether a target file imported from the outside is newly added in the shared data volume or not through the container; if yes, checking the target file, generating a checking result file, and storing the checking result file into the shared data volume; determining whether the target file passes the check according to the check result file; and if the check is passed, storing the target file and the check result file in the shared data volume in a predetermined directory of the operating system. According to the technical scheme, the function of checking the externally imported file is achieved based on the operating system, and the safety and integrity of the file imported into the system from the outside are ensured.

The invention relates to the technical field of cloud computing, and discloses a cloud disk data backup method and device, electronic equipment and a readable storage medium. The method comprises the steps of obtaining a to-be-backed-up cloud disk and to-be-backed-up data corresponding to the to-be-backed-up cloud disk; detecting whether the to-be-backed-up data is valid; and when the to-be-backed-up data is valid, writing the to-be-backed-up data into the object storage corresponding to the to-be-backed-up cloud disk for data backup. By implementing the method and the device, the backup of invalid data is avoided, the space occupation of the invalid data in object storage is reduced, the backup data volume is reduced, and the backup time is saved, so that the backup efficiency is improved.

The invention relates to the field of cloud computing, in particular to a cloud loading method of a physical server, which comprises the following steps: a cloud migration step of migrating data in a disk space of a source end to a target end; and a service switching step of, after the cloud migration step is completed, carrying out service switching between a source end and a target end. The method further comprises an incremental migration step executed together with the cloud migration step, and incremental data generated by a source end is migrated to a target end in multiple times from the time point when the cloud migration step is started. According to the automatic cloud loading method of the physical server, the user can perform migration operation while not closing the service, so that the service of the user is ensured to be affected to the minimum extent. When a system service is online, the whole disk is copied through full-volume migration at a fixed point, full-volume data is transmitted through a network, then the data generated in the full-volume migration process is copied into the target VM again in an increment mode, and the consistency of the data is ensured.

The invention relates to a novel software and hardware collaborative RAID improvement system, and belongs to the field of hard disk arrays. The RAID software computing module and the monitoring moduleare added to the computer host end; the data path 2 and the counting module are added to the RAID board card end, the current CPU load is monitored through the monitoring module, when the CPU load islow, the RAID software computing module is used for RAID computing processing, and the computing result is transmitted to the cache module through the data path 2; in addition, whether the data volume in the cache module is larger than or equal to the total data volume (L + 1) or not is calculated through the counting module, if yes, calculation is stopped, and repeated calculation is prevented.Due to the fact that calculation is conducted through the data calculation module and the RAID software calculation module of the RAID board card at the same time, the RAID calculation processing speed is increased, and remaining CPU resources are utilized. According to the invention, the RAID board card is matched with the software, so that faster RAID performance is realized, and CPU (Central Processing Unit) resource shortage of a computer end cannot be caused.