Software

Scalasca is a software tool that supports the performance optimization of parallel programs by measuring and analyzing their runtime behaviour. The analysis identifies potential performance bottlenecks – in particular those concerning communication and synchronization – and offers guidance in exploring their causes.  Scalasca supports the performance optimization of simulation codes on a wide range of current HPC platforms. [...]

The Open Trace Format Version 2 (OTF2) is a highly scalable, memory efficient event trace data format plus support library. It is the standard trace format for Scalasca, Vampir, and Tau and is open for other tools. [...]

The instrumentation and measurement framework Score-P, together with analysis tools build on top of its output formats, provides insight into massively parallel HPC applications, their communication, synchronization, I/O, and scaling behavior to pinpoint performance bottlenecks and their causes. Score-P is a highly scalable and easy-to-use tool suite for profiling (summarizing program execution) and event tracing (capturing events in chronological order) of HPC applications. [...]

OPARI2 is a source-to-source instrumentation tool for OpenMP and hybrid codes. It surrounds OpenMP directives and runtime library calls with calls to the POMP2 measurement interface. [...]

Focussing on bug fixes, modernisation and performance improvements. <br>The memory footprint has been reduced substantially <br><br><b>Major changes since v0.10</b><ul><li> Modernising the code base and infrastructure: </li><ul> <li>Begin adopting C++20</li> <li>Use the `ruff` linter / formatter </li> <li>Use CPM and simplify CMake.</li></ul><li>Fixing problems with diffusion</li><ul> <li>Update units and scaling</li> </li>Correct solver matrix coefficients</li></ul><li>:warning: Modcc used to mis-compile `-K^n` into `(-K)^n` due to a parser error </li><li>Allow probing of point mechanism state variables by tag.</li></ul><b>Breaking changes</b><ul><li>Stricter checks on network construction / simulation parameters.</li><ul> <li>connections will now throw errors when given negative / zero delays</li> <li>simulation will throw when given a network with minimum delay less than the timestep</li></ul><li>Discretisation now is a property of the `cable_cell` object, no longer the `decor`</li></ul><b>New tutorials and documentation</b><ul><li>Plasticity tutorial</li><li>Add connectivity tutorial</li></ul><b>Internal changes</b><ul><li>Auto-generate type stubs</li><li>Remove clock for std::chrono</li><li>added E_R to lif cell model</li><li>Performance</li><ul> <li>Fix embarrassingly quadratic bug in fvm-layout</li> <li>Faster sort spikes</li> <li>Faster event dispatch</li></ul><li>Memory footprint</li><ul> <li>Create less intermediate data from events</li> <li>Elide GPU allocations for unused arrays (`Xd`, `Xi`, ...)</li></ul><li>Clean-up catalogue extension.</li><li>Refactor discretization</li></ul>

Choosing the optimal solver for systems of ordinary differential equations (ODEs) is a critical step in dynamical systems simulation. ODE-toolbox is a Python package that assists in solver benchmarking, and recommends solvers on the basis of a set of user-configurable heuristics. [...]

Cube is a generic set of tools for description and analysis of 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. [...]

Cube is a generic set of tools for description and analysis of 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. [...]

Cube is a generic set of tools for description and analysis of 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. [...]