Skip to content

linearregression/scalasca

 
 

Repository files navigation

/****************************************************************************
**  SCALASCA    http://www.scalasca.org/                                   **
*****************************************************************************
**  Copyright (c) 1998-2013                                                **
**  Forschungszentrum Juelich GmbH, Juelich Supercomputing Centre          **
**                                                                         **
**  Copyright (c) 2009-2013                                                **
**  German Research School for Simulation Sciences GmbH                    **
**                                                                         **
**  Copyright (c) 2003-2008                                                **
**  University of Tennessee, Innovative Computing Laboratory               **
**                                                                         **
**  See the file COPYRIGHT in the package base directory for details       **
****************************************************************************/

The SCALASCA project (Scalable performance Analysis of Large-Scale parallel
Applications) is the successor of the KOJAK project (Kit for Objective
Judgment and Knowledge-based detection of performance bottlenecks).

The projects aim to develop a generic automatic performance analysis
environment for parallel applications.  Performance problems are specified
in terms of execution patterns that represent situations of inefficient
behavior.  These patterns are input for an analysis process that recognizes
and quantifies the inefficient behavior in event traces.  Mechanisms that
hide the complex relationships within event pattern specifications allow a
simple description of inefficient behavior on a high level of abstraction.


== Tool Install/Administration Documentation ==

* Installation instructions for Scalasca (INSTALL) & CUBE (CubeInstall.pdf):

The CUBE GUI and tools may be installed with Scalasca, though a separate
installation is generally recommended (often on a separate system).
Installation requirements (such as package dependencies) and options are
documented for supported platforms.


== Generic User Documentation ==

Scalasca user documentation for a specific installation is installed into the
$SCALASCA_ROOT/doc subdirectory. Documentation for the latest release can also
be browsed from the website http://www.scalasca.org/download/documentation
(where a flyer and technical papers are also available).

* Scalasca User Guide (manuals/UserGuide.pdf): 

A short introduction to parallel performance analysis and Scalasca; the basic
instrumentation, measurement, analysis and examination steps and Scalasca
commands demonstrated with an example application; and more in-depth reference
material.  Beginners are encouraged to start with the first two chapters,
and consult the later chapters as they require more advanced capabilities.
The tools for processing and GUI for interactively examining Scalasca
analysis reports are covered by the separate CUBE User Guide.

* CUBE User Guide (manuals/CubeGuide.pdf):

Reference for using the CUBE GUI and associated report algebra utilities and
tools, as well as developer chapters covering the CUBE file format and API. 
On-line help, as well as context-sensitive help, are provided within the GUI
itself, and should assist beginners and more advanced users alike.

* Scalasca analysis report performance properties (scalasca_patterns.html) &
  region descriptions (scalasca_regions.html):

Scalasca-specific analysis documentation, generally accessed via the "Online
description" context menus of the Scalasca analysis report explorer (CUBE).

* Scalasca Quick Reference (manuals/QuickReference.pdf):

6-page review of the main steps and commands for Scalasca usage.  Generally
also accessible via the "scalasca -h" command from a Scalasca installation.
After basic familiarisation with Scalasca usage, this short documentation
should provide brief reminders of common commands and options.

* Scalasca open issues & limitations (OPEN_ISSUES):

Compendium of known limitations and unimplemented features that may affect
Scalasca usability on various platforms or with certain applications, with
suggested workarounds where appropriate.  Although issues may be uncommon
and/or quite technically complex, users are encouraged to briefly review
this document, particularly if they believe they have encountered a problem
when using Scalasca, prior to contacting user support.


== (Tool) Developer Documentation ==

* Manuals for Scalasca components (e.g., CUBE)

The CUBE User Guide includes sections for tools developers and various other
Scalasca components also come with separate manuals that are either installed
or available with the Scalasca installation package.  Source-code documentation
is also provided.  Anyone interested in using or developing Scalasca components
is recommended to contact the Scalasca development team.


== Example/Test/Tutorial Material ==

* Scalasca examples ($SCALASCA_ROOT/example)

This subdirectory of the Scalasca installation contains several files
which can be used to test Scalasca and/or explore different features.
See the associated README.example for details.

* VI-HPS tutorial (http://www.vi-hps.org/training/material#tutorials)

Tutorials using the NPB example codes (primarily BT) are available on
the VI-HPS ISO disk image (Linux Live DVD/USB) or for separate download
covering use of Scalasca and other tools.



* For further more detailed information see the project webpages

  http://www.scalasca.org/

* For help, reporting bugs, and making comments or suggestions send e-mail

  mailto:scalasca@fz-juelich.de


Releases

No releases published

Packages

No packages published

Languages

  • C++ 58.7%
  • C 32.2%
  • Shell 3.5%
  • Makefile 2.2%
  • HTML 1.8%
  • Yacc 0.8%
  • Other 0.8%