Product specifications
Table Of Contents
- Table of Contents
- 1 Introduction
- 2 Feature Overview
- 3 Step-by-Step Cluster Setup and MPI Usage Checklists
- 4 InfiniPath Cluster Setup and Administration
- Introduction
- Installed Layout
- Memory Footprint
- BIOS Settings
- InfiniPath and OpenFabrics Driver Overview
- OpenFabrics Drivers and Services Configuration and Startup
- Other Configuration: Changing the MTU Size
- Managing the InfiniPath Driver
- More Information on Configuring and Loading Drivers
- Performance Settings and Management Tips
- Host Environment Setup for MPI
- Checking Cluster and Software Status
- 5 Using QLogic MPI
- Introduction
- Getting Started with MPI
- QLogic MPI Details
- Use Wrapper Scripts for Compiling and Linking
- Configuring MPI Programs for QLogic MPI
- To Use Another Compiler
- Process Allocation
- mpihosts File Details
- Using mpirun
- Console I/O in MPI Programs
- Environment for Node Programs
- Environment Variables
- Running Multiple Versions of InfiniPath or MPI
- Job Blocking in Case of Temporary InfiniBand Link Failures
- Performance Tuning
- MPD
- QLogic MPI and Hybrid MPI/OpenMP Applications
- Debugging MPI Programs
- QLogic MPI Limitations
- 6 Using Other MPIs
- A mpirun Options Summary
- B Benchmark Programs
- C Integration with a Batch Queuing System
- D Troubleshooting
- Using LEDs to Check the State of the Adapter
- BIOS Settings
- Kernel and Initialization Issues
- OpenFabrics and InfiniPath Issues
- Stop OpenSM Before Stopping/Restarting InfiniPath
- Manual Shutdown or Restart May Hang if NFS in Use
- Load and Configure IPoIB Before Loading SDP
- Set $IBPATH for OpenFabrics Scripts
- ifconfig Does Not Display Hardware Address Properly on RHEL4
- SDP Module Not Loading
- ibsrpdm Command Hangs when Two Host Channel Adapters are Installed but Only Unit 1 is Connected to the Switch
- Outdated ipath_ether Configuration Setup Generates Error
- System Administration Troubleshooting
- Performance Issues
- QLogic MPI Troubleshooting
- Mixed Releases of MPI RPMs
- Missing mpirun Executable
- Resolving Hostname with Multi-Homed Head Node
- Cross-Compilation Issues
- Compiler/Linker Mismatch
- Compiler Cannot Find Include, Module, or Library Files
- Problem with Shell Special Characters and Wrapper Scripts
- Run Time Errors with Different MPI Implementations
- Process Limitation with ssh
- Number of Processes Exceeds ulimit for Number of Open Files
- Using MPI.mod Files
- Extending MPI Modules
- Lock Enough Memory on Nodes When Using a Batch Queuing System
- Error Creating Shared Memory Object
- gdb Gets SIG32 Signal Under mpirun -debug with the PSM Receive Progress Thread Enabled
- General Error Messages
- Error Messages Generated by mpirun
- MPI Stats
- E Write Combining
- F Useful Programs and Files
- G Recommended Reading
- Glossary
- Index
IB6054601-00 H Glossary-1
Glossary
A glossary is provided for technical terms used
in the documentation. Italicized terms in the
definitions are defined in the glossary. If you
are viewing this document as a PDF file, the
blue terms are linked to the corresponding
definition.
bandwidth
The rate at which data can be transmitted.
This represents the capacity of the
network connection. Theoretical peak
bandwidth is fixed, but the effective
bandwidth, the ideal rate, is modified by
overhead in hardware and the computer
operating system. Usually measured in
bits/megabits or bytes/megabytes per
second. Bandwidth is related to latency.
BIOS
Stands for Basic Input/Output System. It
typically contains code for initial hardware
setup and bootstrapping.
build node
A machine on which source code,
examples, or benchmarks can be
compiled.
compute node
A machine used to run a job.
connected mode
IPoIB runs in either connected mode
(IPOIB-CM) or unreliable datagram
(IPoIB-UD) mode. Connected mode uses
the Reliable Connected (RC) protocol.
IPoIB in connected mode achieves higher
bandwidth because the RC protocol
supports a larger MTU (typically at least
4MB) than the UD protocol (limited to the
InfiniBand MTU).
context sharing
A method that allows MPI node programs
to share QLogic InfiniPath hardware
resources (contexts). With context sharing,
up to four node programs (in the same MPI
job) can share each available context.
DAPL
Stands for Direct Access Provider Library.
This is the reference implementation for
RDMA transports. Consists of both kernel
mode (kDAPL) and user mode (uDAPL)
versions.
development node
Same as build node
DHCP
Stands for Dynamic Host Configuration
Protocol, a communications protocol for
allocating IP addresses. DHCP also
provides other basic networking informa-
tion, such as router addresses and name
servers.