MAQAO (Modular Assembly Code Quality Analyzer and Optimizer) is developed by UVSQ. See Link.

Within the H4H project, UVSQ will study enhancements of the MAQAO tool for support of GPU and for integration into the SILC measurement system. Actively support the application developers from WP4 in analyzing, porting and optimizing their codes.


- PAS2P, developed by UAB-CAOS, is a toolset to automatically extract the most significant behaviour (phases) of parallel applications, the Parallel Application Signature, that by its execution on different parallel computers, lets us predict the applications' performance. See Link.

Within H4H, UAB/CAOS will develop, implement and evaluate the enhancements of the performance prediction tool PAS2P (Parallel Applications Signature for Performance Prediction).


Scalasca, developed by JSC: is a toolset to analyze the performance behavior of parallel applications and to identify opportunities for optimization. See Link.

Within H4H, JSC will develop, evaluate and document the enhancements related to the analysis tool Scalasca. JSC will also assist ZIH in developing additional support for measurements on GPU and Cell accelerators in the SILC measurement system.


STEP, developed by TSP, stands for "Transformation System Parallel Execution". It is a tool transforming OpenMP programs into MPI programs. It is included in a compilation framework providing array region analysis. (See Link.)

In the H4H project, TSP will provide a transformation for C programs.

TSP will also transform data distribution directives to generation MPI communications.


ThreadSpotter, developed by RogueWave Software AB, performs automatic analyses of the whereabouts of a binary and suggest hands-on changes at the source level to a programmer. See Link.

Within the H4H project, Rogue Wave will develop heuristic to find optimal binding of threads to cores and memory to sockets. It will also propose and demonstrate, together with Nema Labs, a method for uncovering concurrency and memory locality.


Valgrind is a public instrumentation framework for building dynamic analysis tools. The distribution currently includes: a memory error detector, two thread error detectors, a cache and branch prediction profiler, a call-graph generating cache profiler, and a heap profiler. See Link.

Within H4H, HLRS will implement Valgrind extensions for runtime memory checks and GDB-Cuda-based debugging features for the target platforms.


VampirTrace, developed by ZIH is a run-time library and toolset for instrumentation and tracing of software applications using OTF. The traces can be visualized by the Vampir and Scalasca tools. See Link.

Within the H4H project, GWT will develop, evaluate and document the enhancements in the analysis tool Vampir.

Furthermore, ZIH will develop additional support for measurements on GPU and Cell accelerators in the performance measurement system. Further improve the memory and data movement analysis and visualization in Vampir. It will also extend the Marmot tool to identify correctness problems in accelerator based programming models.