Dell PowerMax Operate flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

V-brick

A PowerMax building block containing two director nodes, an InfiniBand switch pair, and attached media; two V-bricks form one engine, and a two-engine system has four V-bricks.

Engine failure domain

Cache and directors are shared within an engine, so an engine failure affects only the two V-bricks in that engine while the other engine continues serving host I/O.

RAID protection per DAE

PowerMax applies RAID-5 (4+1 or 8+1) or RAID-6 (8+2 or 16+2) within each DAE; the RAID level determines usable capacity and rebuild behavior after a disk failure.

Formatted vs usable capacity

Raw capacity is the sum of disk sizes; formatted capacity is after RAID overhead; usable capacity is what thin pools and devices can consume after further overheads.

Unisphere REST API base path

The REST API is served under /univmax/restapi/<version>, for example /univmax/restapi/100/system/session for authentication and /univmax/restapi/100/sloprovision for provisioning.

symcli command pattern

Solutions Enabler commands follow sym<verb> with -sid <array id>, such as symcfg list, symdev list, symsg list, symrdf query, and symsnapvx list.

Unisphere Auditor role

The Auditor role grants read-only access including audit log visibility, while Monitor is read-only without audit log access; Admin and StorageAdmin can modify configuration.

symcfg discover

Refreshes the local Solutions Enabler database of arrays and devices; required after zoning or device changes before symdev and symsg commands report new devices.

Storage group purpose

A storage group is the unit of masking and replication; SRDF, SnapVX, and mobility operate on storage groups, and a volume can belong to only one storage group.

Masking view components

A masking view binds one storage group, one port group of FA director ports, and one initiator group of host HBAs; all three are required to present LUNs to a host.

Thin pool subscription ratio

Subscribed capacity divided by physical pool capacity; thin devices allocate on first write, so the ratio can exceed 1 and must be monitored to avoid pool exhaustion.

Host I/O limits

IOPS and MB/s limits applied to a storage group or port group; they are enforced by front-end directors and are not cumulative across nested limits.

Single-initiator zoning

Each FC zone contains one host HBA WWN and one or more array FA ports; this isolates host traffic and is the recommended zoning practice for PowerMax front-end connectivity.

PowerPath licensing effect

Without a valid PowerPath license, PowerPath falls back to basic failover without load balancing; with a license it provides multipath load balancing and path management.

Path count formula

Paths per volume equals FA ports in the port group multiplied by host HBA ports zoned to those ports; at least two independent paths are required for redundancy.

symaccess list view

CLI command that verifies masking by showing storage groups, port groups, initiator groups, and the devices presented through each masking view on an array.

SnapVX first snapshot

The first snapshot consumes metadata only; subsequent writes to the source allocate pool space for changed tracks, so pool usage grows with write activity.

Linked snapshot target

A linked SnapVX target is a writable copy-on-write clone that can be used for backup, test, or restore; terminating it removes the target device and frees pool space.

Snapshot policy

Automates snapshot creation with an interval and retention count; assigned to a storage group, snapshots are named with the policy prefix and a timestamp.

Snapshot restore requirement

Restoring a snapshot overwrites the source with the snapshot image; the source must be unmounted or quiesced by the host before the restore to avoid data corruption.

SRDF/S acknowledgement

In synchronous mode the host write is acknowledged only after the R2 acknowledges it, so RPO is zero but write latency includes the inter-site round trip.

SRDF/A delta sets

Asynchronous SRDF groups writes into delta sets transferred per cycle; a cycle time too small for the write rate causes cycle switches to fail and pairs to suspend.

SRDF pair states

Healthy states are Synchronized and Consistent; Suspended, Partitioned, Split, and Invalid indicate replication is not protecting data and require investigation.

SRDF/Metro witness

SRDF/Metro presents R2 as active/active; a witness array or bias arbitration prevents split brain when the inter-site link fails and both sides could become read/write.

Response time bottleneck analysis

High FA port busy points to front-end contention; high BE director busy points to back-end; high device response time with low array utilization points to host or path issues.

Thin pool used threshold

A pool above roughly 80 percent used should trigger expansion or cleanup; pool exhaustion invalidates snapshots and can block host writes depending on configuration.

Workload Planner

A Unisphere for PowerMax and CloudIQ capability that models adding hosts or volumes against historical performance data to predict impact before provisioning.

symevent list

CLI command that displays the array event log; events carry severity levels Info, Minor, Major, and Critical and should be triaged by severity and recurrence.

Call-home connectivity

Connectivity Hub or Secure Remote Services sends array alerts to support; its status should be verified in Unisphere and with symcfg to ensure alerts are delivered.

Disk rebuild state

A failed disk in a RAID group triggers a rebuild; the array reports degraded redundancy until the rebuild completes, and a second failure in the same RAID group can cause data loss.

LUN not visible checklist

Verify the masking view (storage group, port group, initiator group), FC zoning, host HBA driver, and multipathing; symaccess list view confirms the mapping on the array.

SRDF resume procedure

After correcting the cause of a suspension, such as an RDF link failure or cycle time issue, resume replication with symrdf resume and confirm pairs return to Synchronized or Consistent.