先試後買

A. When creating a user or user group quota for a namespace (i.e. Dtree is left empty), the namespace for which the quota needs to be created is empty

B. Before creating a quota, you need to complete the creation of a namespace or Dtree.

C. When creating a quota for a specified user or user group, you need to create the user or user group first.

D. When creating a quota for a specified Dtree, the selected Dtree must be empty.

A. SPC-2 is used to test the SAN array performance in the OLPT, database, and email application scenarios.

B. SPC-2 applies to scenarios of sequential large I/O access.

C. SPC-2 applies to scenarios where random small I/Os are quickly responded.

D. SPC-1 applies to scenarios of sequential large I/O access.

A. LUN

B. target ID

C. bus ID

D. initiator ID

A. The maximum cache capacity supported by this product is 512GB

B. Backend port type can support SAS3.0/100Gb RDMA

C. Front-end ports can support 1/10/25/40/100G Ethernet and 25G NoF (NVe over RoCE)

D. The height of the controller box is 20 and it can support up to 16 controllers

A. When the heterogeneous storage is a third-party storage system, the local storage system needs a license file to authorize its use

B. When the heterogeneous storage system is a Huawei storage system (excluding OE storage systems), the heterogeneous storage system needs to obtain a license to use the SmartVirtualization feature.

C. When the administrator configures offline takeover of a third-party heterogeneous storage system, there is no need to first establish a logical connection between the local storage system and the heterogeneous storage system.

D. After the administrator configures this feature, when the application server sends a request to read data from an external LUN, the eDevLUN of the local storage system receives the data read request and reads the data from the external LUN of the heterogeneous storage system.

A. By default, when creating a HyperMetro pair, the local storage system is set as the preferred site and the remote storage system as the non-preferred site.

B. The data, logs, and change information of a medium- or large-sized database are stored on different LUNs of a storage system. Placing these LUNs in a protection group (PG) and then creating a HyperMetro consistency group for that PG can ensure data consistency in replication.

C. If the quorum server is inaccessible, HyperMetro works in static priority mode. When an arbitration occurs, the preferred site wins the arbitration and provides services.

D. In cross DC HyperMetro deployments involving long-distance transmission (a minimum of 25 km for IP networking and 80 km for Fibre Channel networking), dense wavelength division multiplexing devices must be used to ensure a low transmission latency.

A. The engineer can check whether the optical module is faulty. If yes, the engineer can rectify the fault according to the alarm indicating that the optical module of the front-end port is faulty.

B. The engineer can check whether the rates of ports are consistent. If no, the engineer can contact Huawei engineers for onsite handling.

C. The engineer can check whether the optical module type matches between the ports. If no, the engineer can rectify the fault according to the alarm indicating that the optical module type does not match.

D. The engineer can check whether the optical module rate matches between the ports. If no, the engineer can rectify the fault according to the alarm indicating that the optical module rate does not match.

A. FALSE

B. TRUE

A. Generally, the high watermark of the cache should not be too high, and the difference between the high and low watermarks should not be too small.

B. If the high watermark of the cache is too high, the write performance will be affected. If the low watermark of the cache is too low, read performance will be affected.

C. If the high watermark of the cache is too high, I/O fluctuation and latency increase will occur usually when the current I/O traffic bursts.

D. Generally, the low watermark for storing dirty data in the cache should not be too low. The recommended value is 20%.

A. FALSE

B. TRUE

A. If there are multiple CPUs, allocate load among CPUs properly.

B. Replace the current CPU with a more advanced CPU.

C. Add system modules, such as front-end processors and parallel servers.

D. Reduce the data block buffer size.

A. Chunks from different disks form chunk groups based on the user-defined RAID policy.

B. During LUN creation, grains are mapped to LUNs for refined management of storage resources.

C. Generally, a chunk group is divided into 32 KB grains.

D. The storage system divides each disk in a storage pool into chunks of a fixed size. The size of a chunk is usually 8 MB.

A. Round robin algorithm

B. Minimum queue depth algorithm

C. Maximum path first algorithm

D. Maximum task algorithm

A. FALSE

B. TRUE