01. A payment-clearing application runs in one data center. The business sets an RPO of zero and an RTO of four hours, and a second data center 40 km away is available. Running the application in both sites at the same time is not required, and management wants the lowest-cost design that meets both targets.
Which design should the architect propose?
a) Active/standby DR with synchronous remote replication to the second site
b) Active/standby DR with asynchronous remote replication to the second site
c) Active-active DR with HyperMetro and a quorum server at a third site
d) Daily full backups replicated to backup storage in the second site
02. An auditor asks how the backup solution would survive an attacker who has stolen a backup administrator's credentials and tries to delete every backup copy before encrypting production.
Which two measures directly defend against that attack? (Choose two.)
a) Encrypt every backup copy with keys held in an external key manager outside the backup system
b) Lock backup copies with a WORM retention period that no administrator can shorten
c) Increase backup frequency so the newest copy is never more than an hour old
d) Keep an isolated copy in a vault whose network link opens only during transfers
03. A customer runs file systems on a Huawei OceanStor Dorado array and owns a second compatible Huawei array in another building. They want several independent point-in-time backup copies of each file system on the second array, so that the copies survive the loss of the first array, and they do not want to buy or run backup software, backup servers or media servers.
Which Huawei feature fits this requirement?
a) HyperReplication/A
b) HyperCDP
c) HyperVault
d) HyperClone
04. Two arrays in the same city are protected by synchronous remote replication. The replication link fails during business hours while the production hosts keep writing, and it is repaired three hours later.
Which statement describes the protection status during those three hours?
a) Host writes are held until the link returns, so the secondary copy stays identical to the primary
b) Writes continue and are logged as differences; a disaster now would lose them
c) The secondary LUN becomes writable automatically so the DR site can take over if production fails
d) RPO stays at zero, with the primary array caching each write until the secondary acknowledges it
05. In a parallel 3DC design, DC-A replicates synchronously to DC-B in the same city and asynchronously to DC-C in a remote region. DC-A is destroyed.
Which two statements are correct? (Choose two.)
a) Services can resume at DC-B with no data loss
b) DC-C now holds newer data than DC-B, since asynchronous cycles skip the wait for acknowledgment
c) DC-C holds data only up to its last completed asynchronous cycle
d) DC-B and DC-C both lose their replicated data until DC-A returns
06. A HyperMetro pair runs in quorum server mode with DC-A configured as the preferred site. At 10:00 the quorum server at the third site fails while the heartbeat between the arrays stays normal. At 11:00 DC-A loses power completely.
What happens to HyperMetro service at DC-B?
a) It continues, because DC-B wins arbitration as the only array still running
b) It continues, using the quorum decision DC-B cached before the server failed at 10:00
c) It stops for writes, while reads continue from the local cache at DC-B
d) It stops, because the pair fell back to static priority and DC-B is not the preferred site
07. A non-disruptive software upgrade is planned for a production array, and hosts must keep running throughout.
Which three tasks belong in the preparation before the upgrade starts? (Choose three.)
a) Run a pre-upgrade inspection and clear any faults it reports
b) Clear the current alarm list so that new alarms during the upgrade stand out
c) Confirm host multipathing and HBA versions are compatible with the target software
d) Stop all host I/O, because every controller restarts during the upgrade
e) Export the current configuration and document how to roll back
08. A server's backup schedule is a full backup on Sunday night, a differential backup on Wednesday night, and incremental backups on Monday, Tuesday, Thursday and Friday nights. Each incremental captures the changes since the most recent backup of any type. The server fails on Saturday morning.
Which backups restore the most recent data, in order?
a) Sunday full, followed by the Monday, Tuesday, Thursday and Friday incrementals in order
b) Sunday full, Wednesday differential, Thursday incremental, Friday incremental
c) Sunday full, Wednesday differential, Friday incremental
d) Wednesday differential, Thursday incremental, Friday incremental
09. An architect is sizing the replication link for an asynchronous DR solution with an RPO of 15 minutes.
Which two items of information must be collected to size the link for steady-state operation? (Choose two.)
a) The total provisioned capacity of all protected LUNs and file systems at the production site
b) The peak read IOPS of the protected applications
c) The volume of data written to the protected LUNs in each interval across a business day, including the peak
d) The number of host paths configured to each LUN
e) The RPO, which limits how long changed data may wait before reaching the DR site
10. A storage pool has 200 TB of usable capacity, of which 140 TB is used, and usage grows by a steady 4 TB per month. Operations policy requires expansion to be complete no later than the point at which the pool reaches 80% used, and new disks take 3 months from order to installation.
By when must the expansion be ordered?
a) Within 5 months
b) Within 12 months
c) Immediately, with the pool already above 70%
d) Within 2 months