Huawei HCIP-Storage key terms

Huawei's block-virtualization RAID scheme: disks are divided into small chunks that are pooled into extents, and each volume is spread across many disks, so a rebuild copies only the used data on a failed disk.

A logical container built from member disks or extents, from which LUNs, file systems, or objects are provisioned; it separates the physical disk layout from the volumes that hosts actually use.

Logical unit number: a block-level volume carved from a storage pool and presented to a host over FC, iSCSI, or NVMe, where the operating system sees it as a raw disk.

Assigning a large logical size to a volume while consuming pool capacity only as data is actually written, which improves utilization but requires monitoring so the pool does not fill unexpectedly.

Huawei's fully meshed controller interconnection in OceanStor systems, where every controller reaches every disk through redundant paths, so a controller or link failure does not interrupt host I/O.

Writing each host write into the cache of both controllers in a dual-controller array before acknowledging it, so that a single controller failure does not lose acknowledged data or drop the session.

Huawei's all-flash storage family for mission-critical block workloads, built on NVMe flash, a redundant SmartMatrix controller design, and RAID 2.0+ to deliver consistently low latency.

Huawei's scale-out storage family that serves file, object, and HDFS/big-data access from one cluster, targeting unstructured data, analytics, and large-capacity cloud-era workloads.

Huawei's midrange and high-end arrays (for example the 5000 and 6000 V6 series) that mix SSD and HDD in one system with tiering, balancing cost per gigabyte against flash-level performance for hot data.

Huawei's automated tiering feature, which monitors data access frequency and moves hot data to SSD tiers and cold data to HDD tiers inside a pool, without changing host configuration.

A Huawei feature that reserves part of the SSDs in a hybrid array as a read cache in front of HDD capacity, accelerating repeated reads without permanently relocating the data.

Huawei's heterogeneous virtualization feature that presents LUNs owned by third-party arrays through an OceanStor system, so existing data can use Huawei snapshots, replication, and migration.

Huawei's backup and recovery storage product line, including backup appliances and scale-out backup storage, with features such as source-side deduplication, instant recovery, and anti-ransomware protection.

A standard that carries SCSI block commands over TCP/IP, letting hosts reach storage through ordinary Ethernet NICs and switches instead of a dedicated Fibre Channel fabric.

A fabric rule set that permits communication only between selected switch ports; single-initiator, single-target zones limiting each host HBA to its intended array ports are the usual practice.

An extension of NVMe that carries storage commands over RDMA or Fibre Channel networks, removing SCSI overhead and cutting latency for all-flash arrays shared by many hosts.

A file-sharing protocol used mainly by Linux and UNIX clients, which mount a remote export from a NAS system and read or write files as if they were on a local disk.

The file-sharing protocol used by Windows and mixed clients to access NAS shares; modern SMB versions add encryption, multichannel, and continuous availability for clustered storage.

Huawei's host-side multipathing driver for OceanStor arrays; it spreads I/O across all available FC or iSCSI paths, removes failed paths automatically, and sets the preferred array in active-active pairs.

Mapping a LUN to a specific host or host group so that only the initiators in that group can see and write to it, complementing fabric zoning in a SAN.

Huawei's snapshot feature, creating a pointer-based point-in-time image of a LUN or file system almost instantly, used for rollback, test copies, and as a backup source.

Huawei's clone feature that makes a complete, independent physical copy of a source LUN, so testing or reporting can modify the copy without any risk to production data.

Huawei's block-level LUN copy function that duplicates data between volumes on the same or different arrays without host involvement, commonly used for migrations and offline copies.

Detecting identical data blocks and storing only one copy, with the rest replaced by references; Huawei implements it as SmartDedupe, mainly saving space on clone-heavy, VDI, and backup workloads.

Huawei's active-active feature that mirrors a LUN between two arrays in different data centers and lets both serve I/O, giving near-zero RPO and RTO if one site fails.

Huawei's remote replication feature that copies volume data to a remote array synchronously or asynchronously over IP or FC, providing a disaster recovery copy with a defined RPO.

Huawei's local active-active mirroring, keeping two copies of a LUN on different storage engines or arrays inside one data center and presenting one image to the host.

A small standalone server used in HyperMetro and stretched-cluster designs; when the inter-site link fails, it casts the deciding vote on which site keeps serving I/O.

A condition where mirrored sites lose contact and each assumes it should continue serving writes, which can create divergent data; arbitration or a pre-set preferred site resolves it.

A group of LUNs whose snapshots or replicas are created at the same instant, so a multi-volume application such as a database recovers to one consistent point in time.

Recovery point objective: the amount of data, usually expressed in time, that a business can afford to lose in an outage; it determines replication frequency and DR architecture.

Recovery time objective: the maximum acceptable time to bring a service back after a failure, which decides whether the design needs active-active, standby, or restore-from-backup.

A three-data-center disaster recovery layout with two local sites in active-active or active-passive mode plus a distant site receiving asynchronous copies, covering both local and regional failures.

Huawei's backup software for virtual machines, databases, and file systems; it runs full, incremental, and log backups and writes to OceanProtect or OceanStor targets.

The built-in web management interface of an OceanStor array, used for creating pools and LUNs, mapping hosts, configuring snapshots and replication, and reading alarms and performance data.

Huawei's cloud-based operations platform for storage: it gathers array telemetry, performs predictive health and capacity analysis, and gives support staff a single view of many arrays.

A Huawei feature that caps IOPS or bandwidth for chosen LUNs, hosts, or applications so one demanding workload cannot consume the array's performance and starve the others.

Huawei's feature for relocating a LUN and its data between pools or arrays while hosts keep reading and writing it, used for capacity rebalancing and hardware refreshes.

Huawei's disaster recovery management software that orchestrates protection policies, drills, and failover across arrays and hosts, covering HyperMetro, remote replication, and backup workflows.