No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

FusionInsight HD 6.5.0 Product Description 02

Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
System Specifications

System Specifications

Table 5-1 lists the recommended FusionInsight HD system specifications.

Table 5-1 System specifications

Type

Indicator

Specifications

Description

Cluster

Maximum number of nodes

5000

Universal x86 servers (not limited to Huawei servers) or Huawei TaiShan servers

Number of tenants

512

Maximum number of tenants

Number of peer clusters supported by multi-cluster mutual trust

16

Number of peer mutual-trust clusters that can be configured for each cluster

HDFS

Number of NameServices

10

Maximum number of NameNodes supported by the system

Maximum number of files for each NameNode pair

100 million

-

Maximum number of blocks on a DataNode

5 million

-

Maximum number of blocks on a DataNode disk

500 thousands

-

Maximum number of file directories in a directory (excluding recursion)

1 million

Configuration parameter:

dfs.namenode.fs-limits.max-directory-items

Maximum number of blocks in each file

1 million

Configuration parameter:

dfs.namenode.fs-limits.max-blocks-per-file

Maximum length of a file path

8000

Configuration parameter:

dfs.namenode.fs-limits.max-component-length

Minimum block size

1048576

Configuration parameter:

dfs.namenode.fs-limits.min-block-size

Minimum number of normal disks allowed by a DataNode

1

Configuration parameter:

dfs.datanode.failed.volumes.tolerated

Yarn

Maximum memory allocated to a single NodeManager

Physical memory x 0.8

-

Maximum virtual cores allocated to a single NodeManager

Logic CPU x 0.6 to 1.5

-

HBase

Number of HBase RegionServers

512

Number of RegionServer instances of a single HBase service

Number of regions of a RegionServer instance

2000

Maximum number of regions supported by a RegionServer instance

Number of active regions supported by a single RegionServer

200

Maximum number of active regions supported by each RegionServer instance

Hive

Number of partitions supported by a single Hive table

10 thousands

Maximum number of partitions recommended for a single Hive table

Maximum number of files in a single table

1 million

Maximum number of files that can be stored in HDFS for a single Hive table

Maximum number of partitions

3 million

Maximum number of partitions recommended for each Hive service table

Maximum number of concurrent requests of a single HiveServer

500

Maximum number of concurrent requests supported by a HiveServer instance

Storm

Number of nodes in a Storm cluster

128

Universal x86 servers (not limited to Huawei servers) or Huawei TaiShan servers

Kafka

Number of nodes in a Kafka cluster

256

Universal x86 servers (not limited to Huawei servers) or Huawei TaiShan servers

Maximum length of topic names

200 bytes

Not greater than 200 bytes

Redis

Number of instances in a single Redis cluster

512

Number of Redis processes

Number of Redis clusters

512

Maximum number of Redis clusters

Solr

Number of instances in a Solr cluster

500

Number of Solr processes

Number of cores supported by a single SolrServer

200

-

Number of records supported by a single core

1 to 400 millions

-

Maximum memory configuration of a single SolrServer

31 GB

-

Optimal ratio of the memory and disk of a single SolrServer

1:20

-

Elasticsearch

Maximum number of EsMaster instances in an Elasticserach cluster

1024

At least three instances must be installed.

Maximum number of EsNode instances in an Elasticserach cluster

1024

At least two instances must be installed.

Maximum number of EsClient instances in an Elasticserach cluster

1024

Installation of EsClient is optional.

Maximum memory configuration of a single Elasticsearch instance

31 GB

-

Number of records supported by a single shard

1 to 400 millions

-

Maximum number of shards in a single EsNode instance

500

Each 1 GB memory can be used to manage 15 shards. It is recommended that 200 to 300 shards be configured in each EsNode instance.

Maximum storage capacity in a single EsNode instance

15 TB

5 TB to 10 TB data is recommended.

Optimal ratio of the memory and disk of a single EsNode instance

1:50

-

ZooKeeper

Number of instances in a ZooKeeper cluster

5

Maximum number of instances in a ZooKeeper cluster

Maximum number of connections supported an IP address for each ZooKeeper instance

2000

-

Maximum number of connections supported a ZooKeeper instance

20000

-

Maximum number of ZNodes in the case of default parameter configurations

400000

-

Size of a single ZNode

4 MB

-

Flume

Maximum number of Flume instances in a cluster

128

Maximum number of Flume instances

GraphBase

Maximum number of connections that can be enabled in a GraphServer instance at the same time

4096

Configuration parameter:

MAX_CONNECTIONS_PERSERVER

Download
Updated: 2019-05-17

Document ID: EDOC1100074548

Views: 3229

Downloads: 36

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next