This section shows design examples of the PRIMECLUSTER system. Refer to these examples when designing the PRIMECLUSTER system.
System configuration
Figure A.1 System configuration example
Note: The NTP server should be set to one of the following operation mode;
Maintains time that is reliable for special hardware devices such as time units and GPS units.
Uses another server that maintains time that is reliable for special hardware devices such as time units and GPS units, and uses NTP for time synchronization.
Used hardware
PRIMEQUEST x 2
FC-Switch x 1 unit
LAN card x 8
Fibre channel card x 4
Disk array unit x 2
Used software
Red Hat Enterprise Linux AS
PRIMECLUSTER 4.3A00
Multipath driver
Netscape Communicator 4.7x (installed in the operation management personal computer)
Java(TM) 2 Platform Standard Edition Runtime Environment Version 5.0 (installed in the operation management personal computer)
IBM(R) Runtime Environment for Linux(R), Java(TM)2 Technology Edition, Version 1.3.1, 32-bit version (Installed when Linux(R) is used as the operation management client.)
System disk settings
Node name | Physical disk name | Slice number | Slice name | Size (MB) |
primecl01 | sda | 1 | / | 15374 |
2 | /boot/efi | 256 | ||
3 | swap | 2048 | ||
primecl02 | sda | 1 | / | 15374 |
2 | /boot/efi | 256 | ||
3 | swap | 2048 |
Note
When GDS is used to mirror the system disk, it uses one slice number and at least 32 MB of unused area (or swap area) on the system disk. If system disk mirroring is to be executed, consider the resources used by GDS when designing the system.
Shared disk array settings
Device name: ETERNUS3000 | Number of LU | Number of RAID groups | RAID level | Size |
4 | 1 | 5 | 248.2GB |
