H13-624 Practice Exam Questions with Answers HCIP-Storage V5.0 Certification

True

False

The data block is transferred from the source end to the target end, and the deduplication operation is performed on the target end.

Do not occupy the resources of the source.

Target end deduplication is equivalent to post-processing deduplication (Post-processig).

Can not save transmission bandwidth.

Oracle RAC application

SQL Server application

DB2 application

FusionSphere application

When creating a sub-client, you can select global deduplication as the storage strategy.

You can modify the existing storage policy to global deduplication.

The global deduplication strategy can be referenced by multiple deduplication strategies.

Global deduplication and parallel deduplication cannot be used at the same time.

A.49.28MB/s

B.61.6MB/s

C.98.56MB/s

D.123.2MB/s

GE-WAN link

FC link

10GE link

GE private network link

A.1 is wrong and 2 is right.

B.1 wrong 2 wrong.

C. 1 pair 2 pairs.

D.1 is right and 2 is wrong.

True

Wrong

DataStore is based on SAN environment

Need to back up directly to tape

A separate backup network can be used to transmit backup data

Physical MA can be used for other backup tasks, and can also be used for AuxCopy

It is suitable for scenarios that do not require high backup speed.

Applicable to scenarios that require high backup capacity.

Suitable for backup scenarios that do not need to be frequently erased.

Data can be stored for a long time.

True

False

Optional LAN-Base networking

The required bandwidth is 61.6MB/s

The required bandwidth is 71.7MB/s

Must use LAN-Free networking\

CPU usage

User Group

Number of node connections

Node throughput

Volume

Variety-Diversity

Value-value density

Velocity-speed

Business-centric Scale-Out architecture

Data-centric Scale-Up architecture

Data-centric Scale-Out architecture

Computing-centric Scale-Up architecture

True

False

Support

The medium supports multiple interfaces

The FTL algorithm is built into the memory chip

The chip directly confirms the information such as the reading and writing position

After the Guest is checked, the user does not need to enter the user name and password when accessing the shared directory on the client.

After Notify is enabled, a client can not be accessed by other clients for changes in a directory.

The client-side awareness of this directory and its parent directory.

When Oplock (Opportunisticlocking) is enabled, the file will be advanced before being sent to the shared storage

Local caching of rows helps improve client access efficiency.

After the mandatory signature is checked, the client will be forced to use the signature function. If the client is not set to enable state, then

OceanStor 9000 will reject this client's data access request.

The active-active replication network in the two-machine room simultaneously carries data synchronization and heartbeat information between storage systems

Sync and other various data

Computer room A and computer room B need to be in the same fault domain

The network types from the same business host to two active-active storage devices can be different

The arbitration server must use the host in computer room A or computer room B

True

False

As the root directory of the entire system, the namespace cannot be shared or accessed. Only the subdirectories below it can be shared and accessed.

The namespace is automatically created when 0ceanStor 9000 is started, and its name is consistent with the system naming.

A unified name space provides users with a unified storage resource pool to achieve centralized management and allocation of storage space.

Each file sharing directory is a subdirectory of the namespace

Multiple bonded ports can be configured on 1 VLAN

Multiple bonding ports can be configured on one Ethernet port

Ethernet port. Multiple VLANs can be configured on it

Multiple logical ports can be configured on one bonded port

Object size

Data protection level

MD5 value

Creation time

When online takeover uses the multipath software of other vendors, the camouflage attribute of eDevLUN is "basic camouflage takeover"

When taking over offline, the camouflage attribute of eDevLUN is "no camouflage"

Online takeover includes configuration of basic camouflage takeover, configuration of extended camouflage takeover, and configuration of third-party takeover

When offline takeover, application server business needs to be suspended

Mail service reading business

Web server log service

Database analysis business

OLTP data query business

Configure an accessible IP address

Configure account policy

Configure user account audit

Configure login policy

True

False

Throughput by node

Round Robin

Number of connections by node

By CPU usage

True

False

Fast failover, the failure of a single node in the cluster does not affect the business

Multi-node cluster, load balancing, global cache, providing strong performance

The engine cluster is followed by SAN storage, and the SAN storage can be heterogeneous storage

Scalable scale-out engine architecture

POP3

HTML

HTPS

HTTP

HyperReplication: Disaster recovery protection between data centers

SmartMigration: incremental local data protection

HyperSnap: Local data protection for complete copies

HhyperCopy: Data protection between devices

True

False

OceanStor 9000

OceanStor Dorado 5600 V6

OceanStor 6800 V5

OceanStor 100OD

OceanStor Dorado 5600 V6

OceanStor 9000

OceanStor 6800V5

OceanStor 18500V5

The application server is directly connected to the storage controllers A and B through two optical fibers, and the application server is connected to the LUN

The path belonging to the controller is the optimal path

In the dual-switching networking scenario, each application server has multiple paths to the storage, providing link redundancy and avoiding

Single point of failure

In a single-switch networking scenario, the data transmission distance can be extended through the access of the switch

In a single-switch networking scenario, the switch becomes a single point of failure, and network reliability is at risk

True

False

Does not occupy the resources of the source.

Target end deduplication is equivalent to post-processing deduplication (Post-procesig).

Can not save transmission bandwidth.

The data block is transferred from the source end to the target end, and the deduplication operation is performed on the target end.

True

False

Integrated management of hardware, software, and resources

One-click operation and maintenance

Factory pre-integrated

Single node high performance

When the bucket capacity reaches the configured bucket quota, no more write operations can be performed on the bucket.

Bucket resources available for statistics include the size of the bucket, the number of buckets owned, and the number of objects in the bucket.

Account resources available for statistics include account quota, number of buckets owned, number of objects, and total capacity.

When the total bucket capacity of the account reaches the configured account quota, the account can no longer perform write operations.

The remote replication consistency group and its members must have the same attributes

The master and slave luns of the members of the remote replication consistency group can be in any controller.

All members are synchronized together, split, and master-slave switching

If there is a member replication link failure in the consistency group, the data replication of other members will also be

2

3

4

5

power module

Replace the hard disk online

cable (or optical module)

IO module

Alarm management

SLA management

Service Orchestration

Service release

True

wrong

Disk utilization

CPU utilization

Memory utilization

Node throughput

Enterprise ERP

Medical HIS database

Educational video on demand

Enterprise financial system

Capacity quota

File quota

Calculate quota

Mandatory quota

Cancel write protection from the end

Adjust the recovery strategy to manual

Split

Switch directly without any operation

True

False

Data transmission based on TCP protocol

Data transmission based on UDP protocol

Linux platform file sharing agreement

Windows platform file sharing agreement

When data is transmitted from the source end to the target end, the source end first performs a hash comparison on the transmitted data block and removes duplicates.

Compared with target end deduplication, it can save transmission bandwidth.

Occupy source end resources for deduplication processing.

Only the data of the new data block needs to be transferred to the target end, no information related to the old data block needs to be transferred

The security layer provides security protection functions such as FW and IDS of the data center through the Huawei USC series.

The storage layer optimizes the standard 1 write I/O 2 round trips to 1 round trip through the FastWrite function to improve write performance.

The computing layer adopts virtualization platforms such as Huawei FusionSphere and VMware to provide cross-data center large cluster technology to meet the needs of enterprises A variety of key business dual-active requirements.

The transmission layer uses dispersion compensation and other tuning methods to ensure the minimum delay of the transmission layer.

When an inter-array link failure occurs, or a non-priority site fails, the LUN on the priority site continues to provide services, and the non-priority site

The point LUN stops providing services.

When the priority site array fails, the non-priority site will automatically take over the active-active service.

When the priority site array is actively powered off for maintenance, the non-priority site array immediately takes over all active-active services and the services will not be interrupted.

The static priority mode has higher reliability than the arbitration server mode.

Cabinet-->Switch-->KVM--->Node

Cabinet-->KVM-->Switch--->Node

Cabinet-->Switch-->Node--->KVM

Cabinet-->Node-->KVM--->Switch

4+3:1

6+3

4+2

12+3:1

SAN

NAS

NBD.

Hot-Add

NTFS

HDFS

WushanFS

EXT3

capacity frequency allocation

file quota

calculate quota

Mandatory Quota Answer

All members must synchronize, split, disconnect and switch between master and slave together.

It is used to maintain the time consistency of mirroring data between multiple LUNs.

A consistency group can have both synchronous remote replication and asynchronous remote replication.

For remote replication in the same consistency group of a Huawei array, the primary end must be from the same array, and the secondary end must also be from the same array. answer

Bandwidth and network design,

Backup system migration design

Data and application recovery strategy design

Capacity and storage strategy design

8+1

6+3

6+3: 1

16+3:1

Backup management node

Backup media

Backup business node

Client

Quota can be set based on users

quota is set based on user group

Support setting quotas for file directories

Support setting quotas on files

3320GB.

3504GB.

4380GB.

5256GB.

When repairing a bad block, you must first apply for read permission and confirm the local read permission of the active-active pair.

When reading the local LUN of the array, if the bad block is read, if the error cannot be repaired, the host read error message will be returned.

When reading the local LUN of the array, after reading the bad block, if the error is repairable, return the read data to the host to ensure

Fast return of host response.

When reading the local LUN of the array, after reading the bad block, if the error cannot be repaired, the system will

The data is repaired.

61

71

81

91

Project Backbroom

Customer needs

Live network environment

Component selection

True

False

large number of digging

file data

small data

connection data block

True

False

Measure the frequency of data backup based on the importance of the data.

Measure the task execution time based on the amount of change.

Ensure that the network bandwidth used for backup in the same time period is sufficient in the backup window.

Measure the data retention period based on the amount of data.

Remote replication

Snapshot

NDMP backup

HtyperMetro

OBS

MDS

PVS

CA

power module

Replace the hard disk online

cable

IO module

Compared with disaster recovery, the recovery granularity of backup is finer, files can be used as the recovery level, and multiple recovery versions can be made.

If the format of the source data and the backup data are the same, the backup data can be directly read and written by the host.

The backup software can back up data to tape for offline archiving.

True

False

True

False

True

False

Outline design stage

Conceptual design stage

Research phase

Physical design stage

Business disaster recovery

Data distribution

Data access from the end

Filter data that is not highly repetitive through the opportunity table

The fingerprint of the data block enters the opportunity table count first, and only the same data block is written to the threshold multiple times before it can enter the fingerprint table to perform deduplication

The distributed storage space is huge. In order to reduce the memory space consumption of the fingerprint table, the mechanism of the fingerprint opportunity table is introduced

The fingerprint table has a small memory footprint, which can support the deduplication of a larger capacity system

True

False

It is recommended to use SS hard disks for the high-performance tier

The capacity layer uses SATA HDD hard disks

The performance layer uses SAS hard drives

Enterprises have different requirements for storage performance. Considering the cost issue, it is recommended to use tiered storage

Resource allocation

Automatic recommendation of equipment placement in the cabinet

Generate offline configuration file

Equipment networking

The static priority mode has higher reliability than the quorum server mode.

When an inter-array link failure occurs, or a non-priority site fails, the LUN on the priority site continues

Services will continue to be provided, and LUNs at non-priority sites will cease to provide services.

When the priority site array is powered off for maintenance, the non-priority site array immediately takes over all active-active businesses

The business will not be interrupted.

When the priority site array fails, the non-priority site will automatically take over the active-active service.

N FS uses TCP protocol to send and receive data, appropriately increasing the buffer size can speed up TCP transmission

Enabling Jumbo Frames can reduce TCP loss and improve performance. The prerequisite is to confirm that the customer

All network devices between end interfaces must support jumbo frames

In order to support jumbo frames, all nodes of the corresponding Oceanstor 9000 need to be set accordingly

Oceanstor9000 supports jumbo frames, it is necessary to set the MTU of all front-end and back-end business network ports

Pair split occurs, the possible cause is temporary link abnormality or network bandwidth expansion

At this time, the master-slave relationship of the two directories specified by Pair remains unchanged

After the Pair is resynchronized, the difference data and the synchronized amount that have not been copied to the secondary directory due to the split operation in the last synchronization need to be

Seek synchronization

119

72

131

107

87

CIFS transfer block size refers to the size of CIFS file sharing in a single service request between the IP client and Oceanstor 9000.

The size of the data block

Analyze the file size ratio that needs to be stored in Oceanstor 9000 to obtain the storage file model, and then select an appropriate size Numerical value

To adjust the transmission block size, you need to adjust the client and Oceanstor9000 at the same time. Small value shall prevail

Since the parameter value in Oceanstor 9000 has been set to the minimum value by default, it is only necessary to adjust the parameter value of the client to the target value.

Nominal value

The system provides an admin (super administrator) account by default, and the administrator account is not allowed to be deleted

Only super administrator users can execute the create system user command to create users

To view the information of all accounts, the command cat/etc/passwd can be used.

ps-ef command to check TCP service/port

O BS(Obiect Based Store)

MDS (MetaData Service)

Disk K-V Store

CA (Client Agent)

PVS(Protocol and Value-added Service)

Global Cache in wushanFS will all storage servers

The data in the cache of a node cannot be hit by the read and write services of other nodes

Global cache technology helps to improve node memory resource sharing

wushanFS uses distributed locks to realize global cache data management. The same business data is only cached in one node. When other nodes

When you need to access the data, obtain the cached data by applying for a lock

Ground bearing capacity of the machine room

Cabinet parameters

Cabinet power system

If the cluster has 6 nodes, you need to apply for 6 P addresses

Need to specify which node to use.-An IP address

After the software installation is completed, these addresses will be deleted automatically

It does not matter if these temporarily planned back-end storage network IP addresses overlap with the actual planned back-end storage network IP addresses of the nodes.

The switches used in the configuration of the Oceanstor 9000 storage system include CE6810, S5752/S5352 and IS5023.

IS5023 switch provides IB access

S5700/S5352C switches provide GE access

The CE6810 switch provides 10GE/GE access.

Support up to 32 10GE SFP+ Ethernet optical ports

Support 4 40GE QSFP+ Ethernet ports

The 10GE interface on the switch is used for the interconnection between downstream storage nodes and also for the connection between storage nodes and upstream services

40G E interface is used for stacking between switches

True

False

FAT32

NTFS

NFS

CIFS

After Notify is enabled, a client’s change operation in a directory cannot be accessed by other users who are accessing this directory and its parent.

Directory client awareness

When oplock (Opportunistic locking) is enabled, the file will be cached locally before being sent to the shared storage, which is helpful

To improve client access efficiency

After Guest is checked, users do not need to enter a user name and password when accessing the shared directory on the client

After the mandatory signature is checked, the client will be forced to use the signature function. If the client is not set to start state, then Oceanstor9000 will reject the client's data access request

The metadata of Oceanstor 9000 uses mirroring for redundancy, and COW technology is used for the metadata.

InfoStamaper feature provides snapshots based on file level

Oceanstor 9000 InfoStamaper adopts ROW technology

The advantage of using ROW is that it has less impact on system performance.