Index

ace-extras directory
Extra Software: SGI SystemImager

add an execution node
Adding an Execution Node to the Cluster

assumptions
Assumptions

boot messages from Lconsole
Send Boot Messages to the Remote Administration Node

building products from source RPMs
Building Products from Source RPMs

capability
Terminology

C/C++ and exporting performance data
Monitoring Specific Applications and Processes

Cicso switch traffic and PCP visualization tools
Using PCP to Visualize Cisco Switch Traffic

clock
synchronization
Syncronizing Clocks in the Cluster

clock(1) command
Syncronizing Clocks in the Cluster

cluster
Terminology

clustervis
Using PCP Visualization Tools

clustervis tool
Configuring PCP for Remote Display

collection daemon for PCP (PMCD)
Configuring PCP for Remote Display

compressed tar files
Building Products from Source RPMs

confidence script
Using confidence
Using confidence

configuration files
gm-mpich
Configure gm-mpich
installace
Batch Mode Instructions
PBS
New Execution Node

connectivity
Verify TCP/IP Connectivity

date(1) command
Syncronizing Clocks in the Cluster

delete an execution node
Remove an Execution Node

dependencies
Installation of Additional Base OS Packages

DISPLAY variable and Lconsole
Operations

dkvis tool
Configuring your X Server for 16-Bit Display

driftfile
Syncronizing Clocks in the Cluster

driver for Myrinet
Installing the Myrinet Driver (gm)

Errors
MPICH
Permission Denied
PBS
PBS Problems
PCP
PCP Errors

/etc/ace/nodes file
Installing ACE Software
Configuring PCP for Remote Display

/etc/conf.modules file
/etc/conf.modules
/etc/conf.modules

/etc/HOSTNAME file
/etc/HOSTNAME

/etc/hosts.equiv
Interactive Mode Instructions

/etc/hosts.equiv file
/etc/hosts.equiv or $HOME/.rhosts

/etc/inetd.conf file
Configure the Remote Administration Node

/etc/lilo.conf file
Send Boot Messages to the Remote Administration Node

/etc/ntp.conf file
Syncronizing Clocks in the Cluster

/etc/pam.d/rlogin
Problems with PAM

/etc/pcp.conf file
Performance Co-Pilot (PCP)

/etc/resolv.conf file
/etc/resolv.conf
/etc/resolv.conf

/etc/sysconfig/network file
/etc/sysconfig/network

/etc/sysconfig/network-scripts/ifcfg-eth0 file
/etc/sysconfig/network-scripts/ifcfg-eth0
/etc/sysconfig/network-scripts/ifcfg-eth0

ethernet package
Interactive Mode Instructions

Examples
add an execution node
batch mode
Modified install.conf File
interactive mode
Interactive Mode Example
installace
batch mode
Modified install.conf File
interactive mode
Interactive Mode Example

execution node
Terminology
addition
Adding an Execution Node to the Cluster
remove
Remove an Execution Node

expander
Lconsole Command-line Serial Console
Lconsole Utility

experience level required to use this guide
Assumptions

Extra software
Extra Software: SGI SystemImager

Fortran and exporting performance data
Monitoring Specific Applications and Processes

gateway
/etc/sysconfig/network

gm driver and kernel rebuild
Rebuilding the Kernel when using Myrinet without SGI ProPack

gm package
Interactive Mode Instructions

GM_INSTALL utility
Steps on All Nodes

gm-mpich configuration
Configure gm-mpich

gm-mpich package
Interactive Mode Instructions

.gmpi/conf file
Configure gm-mpich

hardware
clock
Syncronizing Clocks in the Cluster
configuration
Topology
reset with Lconsole
Operations

head and execution node
Topology

head node
Terminology

$HOME/.gmpi/conf file
Configure gm-mpich
Remove an Execution Node

$HOME/.rhosts file
/etc/hosts.equiv or $HOME/.rhosts

hostname
Assumptions

HOSTNAME file
/etc/HOSTNAME

hosts.equiv file
/etc/hosts.equiv or $HOME/.rhosts

inetd.conf file
Configure the Remote Administration Node

installace utility
Installing ACE Software

installation tasks
batch mode
example
Modified install.conf File
instructions
Batch Mode Instructions
interactive mode
example
Interactive Mode Example
instructions
Interactive Mode Instructions
verification
What confidence Verifies

installation verification
Using confidence

install.conf file
Batch Mode Instructions

introduction
Introduction

Java and exporting performance data
Monitoring Specific Applications and Processes

kernel rebuild
Rebuilding the Kernel when using Myrinet without SGI ProPack

lclogin(8) command
Lconsole Utility

lconsole package
Interactive Mode Instructions

Lconsole utility
add Lconsole users
Add Lconsole Users and Change Passwords
configure the remote administration node
Configure the Remote Administration Node
configure the serial console
Configure the Serial Console
connect to Lconsole
Connect to Lconsole
connect to the node's console
Operations
delete users
Delete Lconsole Users
exit
Exit
features
Use Lconsole Features
operations
Operations
passwords
Add Lconsole Users and Change Passwords
permit login on the serial port
Permit Login on the Serial Port
power down the node
Operations
power up the node
Operations
reset hardware
Operations
send messages to the remote node
Send Boot Messages to the Remote Administration Node
steal an occupied line
Operations

lcpasswd(8) command
Lconsole Utility
Delete Lconsole Users

lcpassword(8) command
Add Lconsole Users and Change Passwords

lctelnetd(8) command
Lconsole Utility

logs directory for Lconsole
Lconsole Utility

machines.LINUX file
Remove an Execution Node

make(1) file
Building Products from Source RPMs

messages file
Syncronizing Clocks in the Cluster

mom_priv/config file
Jobs Fail to Start

monitoring with PCP
Performance Co-Pilot (PCP)

mount directory
Interactive Mode Instructions

MPI and PCP visualization tools
Using PCP to Visualize MPI Jobs

MPI verification
What confidence Verifies

MPICH
errors
Permission Denied

mpich package
Interactive Mode Instructions

MPICH programming environment
MPICH for Ethernet

mpirun
Permission Denied

mpivis
Using PCP Visualization Tools

mpivis tool
Configuring your X Server for 16-Bit Display

multiplexer
Lconsole Command-line Serial Console
Lconsole Utility

Myrinet driver installation
Installing the Myrinet Driver (gm)

Myrinet MPICH configuration
Configure gm-mpich

myrinet package
Interactive Mode Instructions

Myrinet switch traffic and PCP visualization tools
Using PCP to Visualize Myrinet Switch Traffic

network configuration problems
/etc/conf.modules file
/etc/conf.modules
/etc/HOSTNAME file
/etc/HOSTNAME
/etc/resolv.conf file
/etc/resolv.conf
/etc/sysconfig/network file
/etc/sysconfig/network
/etc/sysconfig/network-scripts/ifcfg-eth0 file
/etc/sysconfig/network-scripts/ifcfg-eth0

network file
/etc/sysconfig/network

Networks
Topology

NIS
/etc/sysconfig/network

node
Terminology

osvis tool
Configuring your X Server for 16-Bit Display
Configuring your X Server for 16-Bit Display

package requirements
Installation of Additional Base OS Packages

packages
Interactive Mode Instructions

passwords for Lconsole
Lconsole Utility

PBS
PBS Batch Scheduling Tool
errors
PBS Problems
verification
Using confidence

pbs_mom
Jobs Fail to Start

pbs-client package
Interactive Mode Instructions

pbsnodes
Using confidence

pbs-server package
Interactive Mode Instructions

PCP
Performance Co-Pilot (PCP)
enhancing the use of
Configuring PCP for Remote Display
errors
PCP Errors
IRIX workstation differences
Configuring PCP for Remote Display
pcp package
Interactive Mode Instructions
Configuring PCP for Remote Display
pcp packages
Configuring PCP for Remote Display
pcp-ace package
Interactive Mode Instructions
Configuring PCP for Remote Display
pcp-pro package
Interactive Mode Instructions
Configuring PCP for Remote Display
visualization tools
Using PCP Visualization Tools

pcp package
Remote Execution for root

pcp-ace package
Remote Execution for root

pcp-pro package
Remote Execution for root

Performance Co-Pilot
Performance Co-Pilot (PCP)
See PCP

ping(1) command
Verify TCP/IP Connectivity

pluggable authentication modules (PAM)
Problems with PAM

pmafm
Using PCP Visualization Tools

pmcd
Using and Enhancing Performance Co-Pilot (PCP)

pmdatrace
Monitoring Specific Applications and Processes

pmgadgets
Using PCP Visualization Tools

pmgcluster
Using PCP Visualization Tools

pmgcluster tool
Configuring PCP for Remote Display

pmgsys
Using PCP Visualization Tools

pmlogger
Using PCP Visualization Tools

pmview
Customizing PCP Visualization Tools

power cycle with Lconsole
Operations

proc metrics
Monitoring Specific Applications and Processes

products in ACE
Products
Space Requirements

rcp access
Remote Execution for root
Verify Remote Access for root

rebuilding the kernel
Rebuilding the Kernel when using Myrinet without SGI ProPack

remote access
Remote Execution for root
Verify Remote Access for root

remove an execution node
Remove an Execution Node

reset hardware with Lconsole
Operations

rhost.auth.so
Problems with PAM

.rhosts errors
.rhosts Errors in /var/log/messages

rpm(8) command
Configuring PCP for Remote Display
Building Products from Source RPMs

RPMs and building source
Building Products from Source RPMs

rsh access
Remote Execution for root

serial console
Configure the Serial Console

serial multiplexer
Lconsole Command-line Serial Console
Lconsole Utility

setclock(8)
Syncronizing Clocks in the Cluster

SGI 1400 and SGI 1200 systems
Topology

SGI SystemImager
Extra Software: SGI SystemImager

sgi-acedocs package
Interactive Mode Instructions

sgi-acedocs-print package
Interactive Mode Instructions

sgi-ace-tools package
Interactive Mode Instructions

source RPMs
Building Products from Source RPMs

SOURCES directory
Building Products from Source RPMs

space requirements
Space Requirements

steal console lines
Operations

switch traffic and PCP visualization tools
Using PCP to Visualize Myrinet Switch Traffic

synchronize clocks
Syncronizing Clocks in the Cluster

system clock
Syncronizing Clocks in the Cluster

SystemImager
Extra Software: SGI SystemImager

tar(1) command
Building Products from Source RPMs

tarballs
Building Products from Source RPMs

TCP/IP connectivity
Verify TCP/IP Connectivity

telnet(1) and Lconsole
Connect to Lconsole

terminology
Terminology

thread-safe API
Monitoring Specific Applications and Processes

throughput
Terminology

timezone
Syncronizing Clocks in the Cluster

/tmp requirement
Space Requirements

topology
Topology

touch(1) command
Syncronizing Clocks in the Cluster

troubleshooting
MPICH problems
Permission Denied
network configuration problems
/etc/conf.modules file
/etc/conf.modules
/etc/HOSTNAME file
/etc/HOSTNAME
/etc/resolv.conf file
/etc/resolv.conf
/etc/sysconfig/network file
/etc/sysconfig/network
/etc/sysconfig/network-scripts/ifcfg-eth0 file
/etc/sysconfig/network-scripts/ifcfg-eth0
no rsh access
No rsh Access
node access problems
Remote Execution for root
PAM problems
Problems with PAM
PBS problems
PBS Problems
PCP errors
PCP Errors
.rhosts errors
.rhosts Errors in /var/log/messages

tstmachines(1) command
What confidence Verifies
Permission Denied

tzselect(8) command
Syncronizing Clocks in the Cluster

/usr requirement
Space Requirements

/usr/local/bin directory
PBS Batch Scheduling Tool

/usr/local/gm directory
Steps on All Nodes

/usr/local/sbin directory
PBS Batch Scheduling Tool

/usr/share/pcp/bin
Performance Co-Pilot (PCP)

/usr/spool/pbs/mom_priv/config
New Execution Node

/usr/spool/pbs/server_priv/nodes file
Head Node
Remove an Execution Node

/usr/src/redhat/SOURCES directory
Building Products from Source RPMs

/usr/util/machines/machines.LINUX file
Remove an Execution Node

usurp Lconsole line
Operations

/var/Lconsole/logs directory
Lconsole Utility

/var/log/messages file
Syncronizing Clocks in the Cluster

/var/log/pcp
Performance Co-Pilot (PCP)

/var/logs/messages
.rhosts Errors in /var/log/messages

/var/pcp
Performance Co-Pilot (PCP)

weblogvis tool
Configuring your X Server for 16-Bit Display

webvis tool
Configuring your X Server for 16-Bit Display

xdpyinfo tool
Configuring your X Server for 16-Bit Display

xntpd(8) daemon
Syncronizing Clocks in the Cluster