Question 1

Neither Spark nor Hadoop is suitable for iterative computing scenarios.

ATrue

BFalse

Answer : B

Question 2

The Fusioninsight HD management node provides a dual-machine HA mechanism. After the master node fails, the master and backup nodes need to implement failover manually.

ATrue

BFalse

Answer : B

Question 3

In the FusionlnsightHD system, which disk partition does not belong to the cluster node?

AOS partition

BData partitioning

CMetadata partitioning

Dmanagement partition

Answer : D

Question 4

The order in which the YARN scheduler allocates resource requests, which of the following descriptions is correct.

AAn arbitrary machine-> same rack-> Mochi resources

BAny machine-> local resources-> same rack

CLocal resources-> Same rack-> Any machine

DSame rack-> Any machine-> Local Huangyuan

Answer : C

Question 5

A user needs to build a 350-node Fusioninsight HD cluster. Which plan is the best?

AManagement nodes, control nodes, and data nodes are deployed together in a two-tier network

Bmanagement node and control node are deployed together, data nodes are deployed independently, and layer 2 networking

C1 management node, control node, and data node are all deployed independently, with three-layer networking (correct plan)

DManagement nodes and data nodes are deployed together, control nodes are deployed independently, and layer 2 networking