SGI® Altix® XE Clusters Quick Reference Guide 007-5474-003
2 007-5474-0031: SGI Altix XE Cluster Quick-referenceinstance, one application may run on 16 processors in the cluster while another application runs
Site Plan Verification007-5474-003 3not connect directly to the “outside world” because mixing external and internal cluster network traffic could imp
4 007-5474-0031: SGI Altix XE Cluster Quick-referenceSGI Altix XE250 and XE270 Node Front Controls and IndicatorsThe front control panel on the SGI A
Booting the XE Cluster007-5474-003 5Altix XE320 or XE340 Compute Node Controls and IndicatorsPowerRESETRESET21Overheat/Fan fail LEDNIC 2 activity LEDN
6 007-5474-0031: SGI Altix XE Cluster Quick-referenceCluster Configuration OverviewThe following figures are intended to represent some of the genera
1U slide outconsoleRemote workstationmonitorCustomer EthernetBase Gigabit Ethernetswitch for Admin.StandardRJ-45twisted-paircableHead NodeCompute Node
1U slide outconsoleRemote workstationmonitorCustomer EthernetBase Gigabit Ethernetswitch for Admin.StandardRJ-45twisted-paircableHead NodeCompute Node
1U slide outconsoleRemote workstationmonitorCustomer EthernetBase Gigabit Ethernetswitch for Admin.StandardRJ-45twisted-paircableHead NodeCompute Node
Base Gigabit Ethernetswitch for Admin.InfiniBandswitch (MPI)1U slide outconsoleRemote workstationmonitorStandardRJ-45twisted-paircableInfiniBand cable
Base Gigabit Ethernetswitch for Admin.InfiniBandswitch (MPI)1U slide outconsoleRemote workstationmonitorStandardRJ-45twisted-paircableInfiniBand cable
COPYRIGHT© 2008-2009 SGI. All rights reserved; provided portions may be copyright in third parties, as indicated elsewhere herein. No permission is gr
Base Gigabit Ethernetswitch for Admin.InfiniBandswitch (MPI)1U slide outconsoleRemote workstationmonitorStandardRJ-45twisted-paircableInfiniBand cable
Power Down the Cluster007-5474-003 13Power Down the ClusterNote: You can also use the baseboard management controller (BMC) interface to perform powe
14 007-5474-0031: SGI Altix XE Cluster Quick-referencePowering Off ManuallyTo power off your cluster system manually, follow these steps:Caution: !I
Ethernet Network Interface Card (NIC) Guidelines007-5474-003 15Ethernet Network Interface Card (NIC) GuidelinesWhile Ethernet ports are potentially va
Table 1-2 Head Node Ethernet Address Listings Head node numberInternal management IP address nic2(GigEnet) MPI NAS/SAN option nic3 Infiniband IP addr
Changing the NIC1 (Customer Domain) IP Address007-5474-003 17– Click the “IP Address” box for device eth0 and change the IP address – Click the “Subne
18 007-5474-0031: SGI Altix XE Cluster Quick-referenceCluster Compute Node IP Addresses The cluster system can have multiple compute nodes that each
Switch Connect and IP Address007-5474-003 19Web or Telnet Access to Maintenance Port on the Gigabit Ethernet SwitchYour switch(s) setup is configured
20 007-5474-0031: SGI Altix XE Cluster Quick-referenceSerial Access to the SMC SwitchUse of a serial interface to the switch should only be needed if
InfiniBand Switch Connect and IP Address007-5474-003 21InfiniBand Switch Connect and IP AddressThe subsection “Web or Telnet Access to the InfiniBand
007-5474-003 iiiRecord of RevisionVersion Description-001 March 2008First publication. Note that substantial content included in this document was ori
22 007-5474-0031: SGI Altix XE Cluster Quick-referenceSerial Access to the SwitchYou should connect a Voltaire serial cable (either DV-9 to DB-9 or D
Using the 1U Console Option007-5474-003 233. Set up the network for your InfiniBand switch cluster configuration using the following information and t
24 007-5474-0031: SGI Altix XE Cluster Quick-referenceInstalling or Updating SoftwarePlatform Manage offers a mechanism to upload and install softwar
Platform Manage Troubleshooting Tips007-5474-003 25Note: The DEL key and F2 key work only if the proper ACSII terminal settings are in place. Many Li
26 007-5474-0031: SGI Altix XE Cluster Quick-referencehave trouble that is more hardware related, see “Customer Service and Removing Parts” on page 3
Related Publications007-5474-003 27Related PublicationsThe following SGI system documents may be useful or necessary while configuring and operating y
28 007-5474-0031: SGI Altix XE Cluster Quick-reference• SGI Altix® Systems Dual-Port Gigabit Ethernet Board User's Guide, Publication Number 007
Third-Party Clustering Documents007-5474-003 29Third-Party Clustering DocumentsThe SGI Altix XE Cluster is provided in different configurations and no
30 007-5474-0031: SGI Altix XE Cluster Quick-reference• SMC® TigerStack™ II Gigabit Ethernet Switch Management Guide Use this guide to manage the ope
Customer Service and Removing Parts007-5474-003 31Customer Service and Removing PartsIf you are experiencing trouble with the cluster and determine th
32 007-5474-0031: SGI Altix XE Cluster Quick-referenceLatin America +55 11.5185.2860Europe +44 118.912.7500Japan +81 3.5488.1811Asia Pacific +1 650.9
007-5474-003 33Chapter 22. Administrative Tips and Adding a NodeThis chapter provides general administrative information as well as basic instructions
34 007-5474-0032: Administrative Tips and Adding a NodeAdministrative TipsRoot password and administrative information includes:• Root password = sgis
Administrative Tips007-5474-003 35The Platform Manage installer directory (/usr/local/Platform###) is the location of the code used to install Platfor
36 007-5474-0032: Administrative Tips and Adding a NodeStart the Platform Manager GUILogin to the Platform Manager interface as root, the factory pass
Head Node Information Screen007-5474-003 37Head Node Information ScreenYou can view and confirm the head node information from the main GUI screen. Cl
38 007-5474-0032: Administrative Tips and Adding a NodeAdding a Node Starting from the Main GUI ScreenAdd a node when you need to upgrade. To add a cl
Adding a Cluster Compute Node007-5474-003 39Adding a Cluster Compute NodeThese steps should only be taken if the cluster needs to be upgraded or re-cr
40 007-5474-0032: Administrative Tips and Adding a NodeSelecting the Server TypeClick on “Edit” to bring up the “Node Hardware Configuration” network
Network BMC Configuration007-5474-003 41Network BMC ConfigurationClick on the “Edit” button. Assign the new BMC IP address, stepping and BMC host name
007-5474-003 vContents1. SGI Altix XE Cluster Quick-reference . . . . . . . . . . . . . .
42 007-5474-0032: Administrative Tips and Adding a NodeSelect Preferred Operating SystemSelect the option to provision the new node’s operating system
Node Network Configuration Screen007-5474-003 43Node Network Configuration ScreenUse this screen to assign Ethernet 0 (eth0) as your network interface
44 007-5474-0032: Administrative Tips and Adding a NodeEnter the default gateway information (refer to the example in Figure 2-9) and select “Next” to
DNS and NTP Configuration Screen007-5474-003 45DNS and NTP Configuration ScreenThis screen extracts the name server numbers for use with the system co
46 007-5474-0032: Administrative Tips and Adding a NodeNIS Configuration ScreenThis screen allows you to specify, enable or disable a Network Informat
Platform Manager Options Screen007-5474-003 47Platform Manager Options ScreenThis screen provides the options shown, including installation of MPI, yo
48 007-5474-0032: Administrative Tips and Adding a NodeConfiguration Setup Complete ScreenThis screen allows you to install the operating system and P
Checking the Log File Entries (Optional)007-5474-003 49Checking the Log File Entries (Optional)You can check the log file entries during configuration
50 007-5474-0032: Administrative Tips and Adding a NodeSetting a Node Failure Alarm on Platform ManageThis section shows how to create an alarm using
Setting a Node Failure Alarm on Platform Manage007-5474-003 516. At this time you must enter the criteria that trigger the alarm. Click on “Add Criter
vi 007-5474-003ContentsThird-Party Clustering Documents . . . . . . . . . . . . . . . .
52 007-5474-0032: Administrative Tips and Adding a NodeFigure 2-17 Define Chart Data Popup Example (Filter Selected) Next we need to choose the priori
Setting a Node Failure Alarm on Platform Manage007-5474-003 53example, Platform Manager can send an email to a system administrator or e-mail alias. Y
54 007-5474-0032: Administrative Tips and Adding a NodeTo illustrate how an alarm makes it’s appearance we have intentionally brought down the node. A
Setting a Node Failure Alarm on Platform Manage007-5474-003 55Figure 2-20 Node Down Alarm Screen Example
007-5474-003 57Chapter 33. IPMI Commands OverviewThis chapter provides a set of example IPMI commands, and is not meant to be a comprehensive guide in
58 007-5474-0033: IPMI Commands OverviewUser Administration BMC Supports multiple users, username/password is required for remote connections. The clu
Serial-over-lan Commands007-5474-003 59ipmitool <opts> lan set 1 netmask x.x.x.x impitool <opts> lan set 1 arp respond on impitool <o
60 007-5474-0033: IPMI Commands OverviewConnecting to Node Console via SOLipmitool <opts> sol activateDeactivating an SOL ConnectionIn certain c
Chassis Commands007-5474-003 61Chassis CommandsUse the following chassis commands to administer the cluster. Note that you can also use the BMC inter
Contents007-5474-003 viiSerial-over-lan Commands . . . . . . . . . . . . . . . . .
007-5474-003 1Chapter 11. SGI Altix XE Cluster Quick-referenceOverviewYour SGI® Altix® XE cluster system ships with a variety of hardware and software
Comments to this Manuals