Cube 4.x series

Download Build requirements Documentation ChangeLog

 

Cube, which is used as performance report explorer for Scalasca and Score-P, is a generic tool for displaying a multi-dimensional performance space consisting of the dimensions (i) performance metric, (ii) call path, and (iii) system resource. Each dimension can be represented as a tree, where non-leaf nodes of the tree can be collapsed or expanded to achieve the desired level of granularity. In addition, Cube can display multi-dimensional Cartesian process topologies.

The Cube 4.x series report explorer and the associated Cube4 data format is provided for Cube files produced with the Score-P performance instrumentation and measurement infrastructure or the Scalasca version 2.x trace analyzer (and other compatible tools). However, for backwards compatibility, Cube 4.x can also read and display Cube 3.x data.

This software is free but copyrighted

© 1998-2022 Forschungszentrum Jülich GmbH, Germany
© 2009-2015 German Research School for Simulation Sciences GmbH, Jülich/Aachen, Germany

By downloading and using this software you automatically agree to comply with the regulations as described in the Cube v4 license agreement.

Cube Packages

ATTENTION: Since version 4.4 Cube has been split in its components and is distributed in form of four separate packages: CubeW, CubeLib, CubeGUI and jCubeR.

VersionDateDescription
CubeW 4.7
12-Apr-2022
High performance C Writer library
Latest Release
MD5sum: a4468a84c7906d017f7d013b64f6849d
DOI

Includes:
  • Support for location type CUBE_LOCATION_TYPE_ACCELERATOR_STREAM and location group type CUBE_LOCATION_GROUP_TYPE_ACCELERATOR. CUBE_LOCATION_TYPE_ACCELERATOR_STREAM is an alias for CUBE_LOCATION_TYPE_GPU and has the same numerical value.
  • Added a call "bool cube_make_unique_name ( const char *, char * )" to perform the metric unique name correction to ensure one can use this metric in CubePL expression.
  • Build shared and static libraries as default.
  • Minor fixes.
CubeLib 4.7 12-Apr-2022
General purpose C++ library and tools
Latest Release
MD5sum: 1cfae536f0e0daa8c3afc5efada0437a
DOI

Includes:
  • Requires support for the C++11 for the build.
  • Voiding one metric voids its whole subtree. Important for remapping .spec files.
  • cube_dump also prints CUBE file meta data by section.
  • cube_remap2 uses only CPU threads to calculate SCALASCA metrics.
  • cube_cut accepts multiple regions for reroot and prune operations.
  • Fixed handling of cube files containing GPU locations.
  • CubePL operator "defined" can be used to test if metric is defined.
  • Support for location type CUBE_LOCATION_TYPE_ACCELERATOR_STREAM and location group type CUBE_LOCATION_GROUP_TYPE_ACCELERATOR. CUBE_LOCATION_TYPE_ACCELERATOR_STREAM is an alias for CUBE_LOCATION_TYPE_GPU and has the same numerical value.
  • Added a call "bool cube_make_unique_name ( const string&, string& )" to perform the metric unique name correction to ensure one can use this metric in CubePL expression.
  • Various bug fixes in library.
CubeGUI 4.7 12-Apr-2022
Graphical explorer
Latest Release
MD5sum: c99da41343348d13abab0832503f24b4
DOI

Includes:
  • Requires support for the C++11 for the build.
  • Build system resolves target path to the Qt's libraries if it is setup via symlinks.
  • Looking for moc, uic and rcc tools in libexec as well.
  • Added in SourceCodeViewer the ability to reset the previously selected user defined path to the source file.
  • Added metric "Processor frequency" to the standard collection.
  • Improved performance of synchronized cubegui instances.
  • Fixed summation of derived metrics in value views, selections and for hiding.
  • Added option to disable QWebengine at runtime. QWebengine renders. documentation nicely, but doesn't work with some graphic drivers.
  • System Topology Plugin: added toolbar widget to directly choose a topology, if more than three topologies are available.
  • Improved boxplot presentation.
  • TreeItemMarker plugin shows relation between location group and its creator by marking both items if one of them is selected.
  • Sucessfully loaded plugins automatically notify other plugins by setting a global value.
  • Various bug fixes in GUI.

Requires : Installed CubeLib 4.7.x
jCubeR 4.7 12-Apr-2022
Java reader library
Latest Release
MD5sum: ee45dc40e284dc647c2d692ab50132ec
DOI

Includes:
  • Support for location type ACCELERATOR_STREAM and location group type ACCELERATOR. ACCELERATOR_STREAM is an alias for GPU and has the same numerical value.
Supplementary packages for download (Comfort zone)
Bundle 4.7
MD5sum : 9c7adec35336e2ab9cddff57677e8f23
Version 4.7 contains CubeLib 4.7, CubeGUI 4.7 and CubeW 4.7
win32 binary
MD5sum : c90316de95e178cdd3b5d2e2a605e3a6
Contains CubeLib 4.7, CubeGUI 4.7 and CubeW 4.7. It contains JengaFett, Blade and CallGraph plugins, which can be install optionally.
Mac OS .dmg
MD5sum : b53547c0984d62d0361cb7c995f2f83a
Contains CubeLib 4.7, CubeGUI 4.7 and CubeW 4.7
Cube Java Reader (JAR)
MD5sum :2786184cc20c0872c4127e35f4d31197
Contains jCubeR 4.7
Latest master (Developers zone, NO WARRANTY )
CubeW All packages for downloads
CubeLib All packages for downloads
CubeGUI All packages for downloads
jCubeR All packages for downloads
CubeGUI i18n
  • Contains *.ts files of actual CubeGUI/trunk and might be incomplete.
  • Use Qt Linguist to add or correct translations.
  • Send updated *.ts files to Scalasca development team and we will be happy to include it to the list of supported languages.

 

 

Additional Cube Plugins

With 4.7 version of Cube we provide a set of additional plugins, which can be build and installed additionally to Cube GUI/Library 4.7. Note that they have additional prerequisites, which need to be installed before.

ATTENTION: Plugins require CubeLib 4.7.x or younger.

Please note that it is an EXPERIMENTAL development and we do not provide any warranty.
Cube Plugins were tested only on Linux Ubuntu with Qt 5.9.1 and have no documentation so far. Requirements (Score-P, OTF2, CubeGUI or similar) need to be available via PATH.

Plugins are updated to build and work with CubeGUI 4.7.x even if there weren't any development commits.

Plugin NameDateDescription
ScorePion 0.4 12-Apr-2022
Score-P/Intel compiler filter creation plugin
Latest Release
MD5sum: 1c411df7dbb4f0319624095d72e7df66


Allows to build/edit a filter file while exploring an experiment and selecting regions or files to be included or excluded in the filtering process. Allows to store the created filter file in Score-P format (.flt) or as an intel compiler filter file (.ifl)

Requires : Installed Score-P >=v7.0 and CubeLib 4.7.x

Blade 0.4 12-Apr-2022
Simple OTF2 trace explorer
Latest Release
MD5sum: 747d7abe952f05e1693b4f168903b6c4


If experimental directory contains an OTF2 trace, this plugin allows to browse it and study communication patterns. Suitable only for small traces.

Requires : Installed Score-P >=v7.0 or OTF2 >=v3.0 and CubeLib 4.7.x

JengaFett 0.4 12-Apr-2022
Metrics correlation explorer
Latest Release
MD5sum: cacd2cbae3dc4bb21fe9f61e52f8dd5d


This plugin allows to display correlation between metrics as a parallel bar charts distribution across the system tree.

Requires : Installed CubeLib 4.7.x


CallGraph 0.3 12-Apr-2022
Call graph display
Latest Release
MD5sum: fb77d3eda1a20407d6736981a89055be


This plugin builds a call graph based on call tree and assigns the metric values to the edges. Allows to detect a critical calls in the program execution.

Requires : Installed Graphviz and CubeLib 4.7.x

Tau Value View 0.2 20-Apr-2021
Tau Value display
Latest Release
MD5sum: fb77d3eda1a20407d6736981a89055be


This plugin displays statistical data of the metric, colelcted using Tau Format as a small barplot in the tree view. Allows to get an overview over the values distribution.

Requires : Installed CubeLib 4.7.x

Supported Platforms

The Cube tool set has been successfully tested on the following platforms:

  • Linux (x86, x86_64, ia64, ppc64le, ARM)
  • IBM BlueGene/Q
  • AIX (Power)
  • Cray (x86_64, AArch64)
  • Mac OS X (x86_64) (10.8 and younger)
  • Windows 10 ( experimental )

In addition, the provided configure options (see installation instructions for Writer, Library and GUI) may provide a good basis for building and testing the toolset on other systems. Please report success/failure on other platforms to the Scalasca development team.