Top
Interstage Big DataParallel Processing ServerV1.0.0 User's Guide
Interstage

2.2.1 Designing the System Configuration

The system configuration design items shown below are required when building this product.

  1. Design the server configuration

  2. System redundancy configuration design

  3. DFS file system design


Design the server configuration

This product scales out slave servers, and through the addition of slave servers, scalability is improved. It is recommended to make an estimate of how many servers are required by performing prototype tests before using the system in business, as the time required for processing depends on such factors as the number of slave servers used, the Hadoop applications, and the volume and characteristics of data to be processed. On top of this, determine the maximum number of slave servers, including any future expansion.


System redundancy configuration design

A system having higher reliability can be built by installing related products in addition to this product.

The systems that can be built using this product and related products are as follows.


Table 2.3 Supported redundancy configurations

Supported configuration

Supported server type

Means of support

Server repliation

Master server

Supported with just this product

HA cluster configuration 1:1 active standby

Cluster interface repliation / non-repliation

Master server

Supported with just this product

Public LAN repliation / non-repliation

Master server

Slave server

Supported with just this product

Master server: NIC switching method (logical IP address inheritance)

Slave server: NIC switching method (physical IP address inheritance)

Storage system connection repliation / non-repliation

Master server

Slave server

Development server

Collaboration server

Supported if a related product, the " ETERNUS Multipath Driver ", is installed

Design the system configuration by selecting from the "Redundancy configuration targets" (parts outlined in red) the configurations that suit the customer requirements.


Figure 2.1 Redundancy configuration targets


DFS file system design

A file system design, which should be investigated in advance, is required for using a DFS file system.

Refer to "D.1 File System Design" in "Appendix D DFS Environment Construction" for information on file system design.

This manual uses the following configuration as an example to explain file system configurations:

  • Management partition:

/dev/disk/by-id/scsi-1FUJITSU_300000370105

  • Representative partition:

/dev/disk/by-id/scsi-1FUJITSU_300000370106

  • File data partition:

/dev/disk/by-id/scsi-1FUJITSU_300000370107
/dev/disk/by-id/scsi-1FUJITSU_300000370108

  • Master server (primary):

master1

  • Master server (secondary):

master2


Point

A by-id name generated by the udev function is used for shared disk device names. The by-id name is a device name generated from the unique identification information set in the hard disk.

Use of the by-id names enables each server to always use the same device name to access a specific disk.

Check whether each server can recognize the disk partitions having by-id names.

  • Red Hat Enterprise Linux 5

    # udevinfo -q symlink -n /dev/sdb <Enter>
    disk/by-id/scsi-1FUJITSU_300000370105
    # udevinfo -q symlink -n /dev/sdc <Enter>
    disk/by-id/scsi-1FUJITSU_300000370106
    # udevinfo -q symlink -n /dev/sdd <Enter>
    disk/by-id/scsi-1FUJITSU_300000370107
    # udevinfo -q symlink -n /dev/sde <Enter>
    disk/by-id/scsi-1FUJITSU_300000370108
  • Red Hat Enterprise Linux 6

    # udevadm info -q symlink -n /dev/sdb <Enter>
    block/8:48 disk/by-id/ scsi-1FUJITSU_300000370105
    # udevadm info -q symlink -n /dev/sdc <Enter>
    block/8:48 disk /by-id/ scsi-1FUJITSU_300000370106
    # udevadm info -q symlink -n /dev/sdd <Enter>
    block/8:48 disk /by-id/ scsi-1FUJITSU_300000370107
    # udevadm info -q symlink -n /dev/sde <Enter>
    block/8:48 disk /by-id/ scsi-1FUJITSU_300000370108