index(1).html

来自「Pthread lib库完整说明文档」· HTML 代码 · 共 402 行 · 第 1/2 页

differ from other resources. It covers available compilers, libraries, and 
tools. It describes how to set up and run parallel batch jobs. The tutorial 
demonstrates how to run with Digital MPI (DMPI) and MPICH. The lecture is 
followed by a lab exercise. There will be a description of common errors and 
ways to troubleshoot MPI jobs.  A description of performance analysis 
tools and further documentation will also be presented.
<P>
<I>Level/Prerequisites: </I>Ideal for new users of the Compaq clusters. A 
basic understanding of message passing tools is assumed.
<P>
NOTE: This tutorial is no longer being maintained as of 12/04 since the 
remaining LC Compaq clusters are scheduled for decommission in the near future.
</UL>
-------------------------------------------------------------------------->

<A NAME=lcrm> </A>
<A NAME=dpcs> </A>
<A HREF=../lcrm>
<B>Livermore Computing Resource Management System (LCRM)</B></A> (EC3515)
<UL>
The Livermore Computing Resource Management System (LCRM) is a product of LLNL
Livermore Computing Center. Its primary purpose is to allocate computer
resources, according to resource delivery goals, for UNIX-based
production computer systems. It is the batch system that most LC
users use to submit, monitor, and interact with their production
computing jobs.
<P>
This tutorial begins with a brief overview of LCRM and its two primary
functional components, the Resource Allocation and Control System and
the Production Workload Scheduler. Each of these components is then
further explored, with a practical focus on describing commands and
utilities that are provided for the user's interaction with LCRM. Building job
command scripts, running parallel jobs, and job scheduling policies
are also included. The lecture is followed by a lab exercise.
<P>
<I>Level/Prerequisites: </I> Beginner. The material covered by the following
tutorials would also be helpful.
<BR><A HREF=#lc_resources>EC3501: Introduction to Livermore Computing 
Resources</A>
<BR><A HREF=#ibm_sp>EC3503: IBM SP Systems Overview</A> 
<BR><A HREF=#linux_clusters>EC3516: IA32 Linux Clusters Overview</A>
</UL>

<A NAME=mpi> </A>
<A HREF=../mpi>
<B>Message Passing Interface (MPI)</B></A> (EC3505)
<UL>
The Message Passing Interface Standard (MPI) is a message passing library
standard based on the consensus of the MPI Forum, which has over 40 
participating organizations, including vendors, researchers, software library
developers, and users. The goal of the Message Passing Interface is to
establish a portable, efficient, and flexible standard for message passing
that will be widely used for writing message passing programs. As such, MPI
is the first standardized, vendor independent, message passing library. The
advantages of developing message passing software using MPI closely match the
design goals of portability, efficiency, and flexibility. 
<P>
This tutorial will provide a means for those interested in exploring these 
advantages to become familiar with MPI and also to learn the basics of
developing MPI programs. The primary topics that are presented focus on those
which are the most useful for beginning MPI programmers. The tutorial begins
with an introduction, background, and basic information for getting started
with MPI. This is followed by a detailed look at the MPI routines that are
most useful for new MPI programmers, including MPI Environment Management,
Point to Point Communications, and Collective Communications routines. 
Numerous examples in both C and Fortran are provided, as well as a lab
exercise. 
<P>
<I>Level/Prerequisites: </I> Ideal for those who are new to parallel programming
with MPI. A basic understanding of parallel programming in C or Fortran is 
assumed.
</UL>

<A NAME=pthreads> </A>
<A HREF=../pthreads>
<B>POSIX Threads Programming</B></A> (EC3506)
<UL>
In shared memory multiprocessor architectures, such as SMPs, threads can be
used to implement parallelism. Historically, hardware vendors have
implemented their own proprietary versions of threads, making portability a
concern for software developers. For UNIX systems, a standardized C 
language threads programming interface has been specified by the IEEE 
POSIX 1003.1c standard. Implementations that adhere to this standard are
referred to as POSIX threads, or Pthreads. 
<P>
The tutorial begins with an introduction to concepts, motivations, and design
considerations for using Pthreads. Each of the three major classes of 
routines in the Pthreads API are then covered: Thread Management, Mutex
Variables, and Condition Variables. Example codes are used throughout to
demonstrate how to use most of the Pthreads routines needed by a new Pthreads
programmer. The tutorial concludes with a discussion and examples of how to
develop hybrid MPI/Pthreads programs in an IBM SMP environment. A lab
exercise, with numerous example codes (C Language) is also included. 
<P>
<I>Level/Prerequisites: </I> Ideal for those who are new to parallel 
programming with  threads. A basic understanding of parallel programming in 
C is assumed. For those who are unfamiliar with Parallel Programming in 
general, the material covered in <A HREF=#parallel_comp>EC3500: Introduction 
to Parallel Computing</A> would be helpful. 
</UL>

<A NAME=openMP> </A>
<A HREF=../openMP>
<B>OpenMP</B></A> (EC3507)
<UL>
OpenMP is an Application Program Interface (API), jointly defined by a group
of major computer hardware and software vendors. OpenMP provides a portable,
scalable model for developers of shared memory parallel applications. The API
supports C/C++ and Fortran on multiple architectures, including UNIX & Windows
NT. This tutorial covers most of the major features of OpenMP, including its
various constructs and directives for specifying parallel regions, work
sharing, synchronization and data environment. Runtime library functions
and  environment variables are also covered. This tutorial includes both C
and Fortran example codes and a lab exercise.
<P>
<I>Level/Prerequisites: </I> Geared to those who are new to parallel programming
with OpenMP. Basic understanding of parallel programming in C or Fortran 
assumed.  For those who are unfamiliar with Parallel Programming in
general, the material covered in <A HREF=#parallel_comp>EC3500: Introduction
to Parallel Computing</A> would be helpful.
</UL>

<A NAME=totalview> </A>
<A HREF=../totalview>
<B>TotalView Debugger</B></A> (EC3508)
<UL>
The TotalView debugger is part of a suite of software development tools from
Etnus, Inc., that debug, analyze, and tune program performance. TotalView
provides source level debugging for serial, parallel, multiprocess and
multithreaded codes, and can be used in a variety of UNIX environments,
including those with distributed, clustered, stand-alone and SMP machines. 
TotalView is easy to use because of its completely graphical, X Windows GUI. 
TotalView has been selected as the Department of Energy's ASCI debugger. 
<P>
This tutorial has three parts, each of which includes a lab exercise. 
Part 1 begins with an overview of TotalView and then provides detailed 
instructions on how to set up and use its basic functions. Part 2 continues 
by introducing a number of new functions and also providing a more in-depth 
look at some of the basic functions. Part 3 covers parallel debugging, 
including threads, MPI, OpenMP and hybrid programs. Part 3 concludes with a 
discussion on debugging in batch mode. 
<P>
<I>Level/Prerequisites: </I> Intended for those who are new to TotalView. 
A basic understanding of parallel programming in C or Fortran is assumed.
The material covered in the following tutorials
would also be beneficial for those who are unfamiliar with parallel 
programming in MPI, OpenMP and/or POSIX threads:
<BR><A HREF=#mpi>EC3505: MPI</A>
<BR><A HREF=#pthreads>EC3506: POSIX Threads</A>
<BR><A HREF=#openMP>EC3507: OpenMP</A>
</UL>

<A NAME=perf_analysis> </A>
<B>Performance Analysis Tools And Topics For LC 'S IBM ASCI Systems</B>
(EC3509)
<BR>Part 1: <A HREF=../mpi_performance>MPI Performance Topics</A>
<BR>Part 2: <A HREF=../performance_tools>Performance Analysis Tools for 
the IBM SP Environment</A>
<UL>
Part 1 assumes a previous knowledge and use parallel computing
with MPI.  It begins with a brief review of basic MPI terminology,
message passing routines, and factors that affect an MPI program's
performance. A more in-depth examination of a number of specific factors
that affect performance is pursued. The topics covered include message
buffering, message passing protocols, synchronization issues, message size
factors, collective communications, derived datatypes, communication
contention and more. Comparisons between various options are made and
suggestions for how to improve program performance are offered. The topics
covered apply to MPI programs in general, though the example performance
results cited are derived from program executions on the IBM SP platform.
<P>
<I>Level/Prerequisites: </I> Intermediate to experienced MPI programmers.
</UL>

<UL>
Part 2: An essential prerequisite for optimizing an application is to first
understand its execution characteristics. A number of tools are available
for the application developer to accomplish this, ranging from simple
shell utilities, timers and profilers, to sophisticated graphical
tools. This tutorial investigates, in varying depths, a number of tools
that can be used to analyze an application's performance towards the goals
of optimization and trouble-shooting. A lab exercise featuring
a subset of these tools is provided.
<P>
<I>Level/Prerequisites: </I> A basic understanding of parallel programming 
in C or Fortran is assumed. 
</UL>

<!-------------------------------------------------------------------------->

<SCRIPT LANGUAGE="JavaScript">PrintFooter("UCRL-MI-133316")</SCRIPT>

<BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR>
<BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR>
<BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR>
<BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR><BR>

</BODY>
</HTML>