Practice Free H13-624_V5.5-ENU Exam Online Questions
An industry customer purchased two Huawei OceanStorPacific distributed storages and deployed them in different computer rooms. Now the engineer needs to configure the Hyperletro active-active feature in the above equipment. It is known that the arbitration mechanism at this time is static priority mode, and computer room A is the priority site. When the link between the two storage systems is interrupted, please analyze what the active-active pair operation status is at this time and what is the result after arbitration?
- A . The active-active pair is running normally, and the arbitration result is that the volumes in computer room A and computer room B continue to run services.
- B . The active-active pair is in the waiting-for-synchronization state. The arbitration result is that the volume in computer room A continues to run services, while the volume in computer room B stops running services.
- C . The active-active pair is in the waiting-for-synchronization state. The arbitration result is that the volume in computer room A cannot be accessed and the volume in computer room B stops operating.
- D . The active-active pair is running normally. The arbitration result is that the volume in computer room A cannot be accessed, while the volume in computer room B continues to run services.
Huawei scale-out storage provides file services using the distributed parallel client (DPC) to meet the requirements of high-performance computing (HPC) scenarios.
Which of the following statements about the DPC is false?
- A . DPC provides high storage performance for HPC scenarios.
- B . DPC implements I/O-level load balancing, preventing service load forwarding between storage nodes.
- C . A single client can connect to multiple storage nodes, and the single-stream bandwidth is high.
- D . DPC does not support MPI-I/O. Therefore, DPC can provide better access performance for upper-layer services.
Each SCSI device on the host has a SCSI address.
Which of the following parts does the address consist of? (Multiple choice)
- A . initiator ID
- B . bus ID
- C . target ID
- D . LUN
When delivering a flash storage project, Huawei engineers focused on introducing the SmartDedupe & SmartCompression features of flash storage to customers.
Which of the following descriptions of this feature are wrong? (Multiple choice)
- A . Background similarity deduplication is similar to online deduplication. Both require adding the mapping relationship of the reference block’s fingerprint, persistent address, and reference count list to the fingerprint library, and then updating the mapping.
- B . Flash storage systems only support LUN data reduction, and the storage can provide an adaptive deduplication strategy that combines online deduplication and similar deduplication.
- C . During the online deduplication process, the system will check the fingerprint database to see if the fingerprint exists. If it exists, it is the old block. If it does not exist, it is the new block.
- D . The adaptive deduplication and compression process of flash storage does not involve SFP. According to the characteristics of business data, it identifies the data suitable for online deduplication and directly performs online deduplication processing.
An engineer is participating in the development and testing of a storage product system. When performing the compilation optimization phase of the IOPS performance optimization part, which of the following operations will he not perform?
- A . Improve icache hit rate based on FGO
- B . Manual prefetching improves dcache hit rate in loop
- C . Improve the spatial locality of instructions between functions and improve the icache hit rate by rearranging the function positions during the compilation and linking phase
- D . Improving instruction prefetch hit rate based on hot and cold partitions
Engineer A now needs to configure Trap notifications for multiple Huawei flash storages.
Which of the following descriptions of the configuration are wrong? (Multiple choice)
- A . When configuring Trap notifications, if you use the SNMPv2c protocol, you need to configure a USM user on the storage system to log in to the upper-level external network management system.
- B . When setting USM user information, the authentication algorithms mainly include "MD5", "SHA", "SHA224", "SHA256", "SHA384" and "SHA512"
- C . When configuring Trap notifications, if you use the SNMPv1 or SNMPv3 protocol, you need to configure the SNMP community word on the storage system, which is mainly used for connecting to a third-party network management system.
- D . To query or modify the security policy of the SNMP service, you can use the show Snmp safe-strategy command and the change snmp safe-strategy command in the CLI interface.
An industry customer has purchased several Huawei flash storage devices for service deployment. An engineer wants to configure the HyperSnap feature on the storage devices. However, during the configuration, the engineer discovers that snapshots cannot be created for an existing protection group. The possible cause is that the storage pool where the source LUNs in the protection group reside has no available capacity.
- A . TRUE
- B . FALSE
A financial customer purchased several Huawei flash memory devices for business deployment. Each device carries logs and database-related business data. During a maintenance process, engineers found that the storage system frequently freezes, and after monitoring, they found that the disk read and write rates and IOPS are too high. At this time, which of the following measures can engineers take to optimize performance? (Multiple choice)
- A . Engineers replace the database software with another type of database
- B . Engineers can place different databases in different storage pools for data carrying
- C . Engineers add front-end machines and set up parallel servers
- D . Engineers deploy log and database objects separately and distribute them on independent storage devices
Distributed storage systems can support cabinet-level security, that is, by configuring the EC redundancy mechanism, N data blocks and M check blocks are stored in different cabinets. It can allow M nodes or M hard disks to fail, and the system can still read and write data normally without interrupting business.
- A . True
- B . False