Tag: Performance Engineering

SimGrid is a framework for developing simulators of distributed applications that executed on distributed platforms, which can in turn be used to prototype, evaluate and compare relevant platform configurations, system designs, and algorithmic approaches.

SimGrid provides ready to use models and APIs to simulate popular distributed computing platforms (commodity clusters, wide-area and local-area networks, peers over DSL connections, data centers, etc.) As a result, SimGrid has served as the foundational technology for developing simulators and obtaining experimental results for a wide range of distributed computing domains: Grid computing, P2P computing, Cloud computing, Fog computing, Volunteer computing, HPC with MPI, MapReduce.

SimGrid is accurate, scalable, and usable

• Accurate: SimGrid’s simulation models have been theoretically and experimentally evaluated and validated
• Scalable: SimGrid’s simulation models and their implementations are fast and have low memory footprint, making is possible to run SimGrid simulations quickly on a single machine
• Usable: SimGrid is free software (LGPL license) available on Linux / Mac OS X / Windows, and allows users to write simulators in C++, C, Python, or Java.

PyPOP package is designed to make it easy to perform application performance analyses based on the POP methodology. The primary goals of PyPOP are:

• Easy calculation of POP metrics
• High quality figure generation
• Easy access to underlying data and statistics (using Pandas)
• Flexible and extensible design

MAQAO (Modular Assembly Quality Analyzer and Optimizer) is a performance analysis and optimization framework operating at binary level with a focus on core performance. Its main goal of is to guide application developpers along the optimization process through synthetic reports and hints.

MAQAO mixes both dynamic and static analyses based on its ability to reconstruct high level structures such as functions and loops from an application binary.

Since MAQAO operates at binary level, it is agnostic with regard to the language used in the source code and does not require recompiling the application to perform analyses. MAQAO has also been designed to concurrently support multiple architectures. Currently the Intel64 and Xeon Phi architectures are implemented.

The main modules of MAQAO are LProf, a sampling-based lightweight profiler offering results at both function and loop levels, CQA, a static analyser assessing the quality of the code generated by the compiler, and ONE View, a supervising module responsible for invoking the others and aggregating their results.

Other modules, currently in beta version, allow performing value profiling (VProf) and decremental analysis (DECAN).

TAU Performance System® is a portable profiling and tracing toolkit for performance analysis of parallel programs written in Fortran, C, C++, UPC, Java, Python.

TAU (Tuning and Analysis Utilities) is capable of gathering performance information through instrumentation of functions, methods, basic blocks, and statements as well as event-based sampling. All C++ language features are supported including templates and namespaces. The API also provides selection of profiling groups for organizing and controlling instrumentation. The instrumentation can be inserted in the source code using an automatic instrumentor tool based on the Program Database Toolkit (PDT), dynamically using DyninstAPI, at runtime in the Java Virtual Machine, or manually using the instrumentation API.

TAU’s profile visualization tool, paraprof, provides graphical displays of all the performance analysis results, in aggregate and single node/context/thread forms. The user can quickly identify sources of performance bottlenecks in the application using the graphical interface. In addition, TAU can generate event traces that can be displayed with the Vampir, Paraver or JumpShot trace visualization tools.

Score-P – Scalable Performance Measurement Infrastructure for Parallel Codes- The Score-P measurement infrastructure is a highly scalable and easy-to-use tool suite for profiling, event tracing, and online analysis of HPC applications. It has been created in the German BMBF project SILC and the US DOE project PRIMA and will be maintained and enhanced in a number of follow-up projects such as LMAC, Score-E, and HOPSA. Score-P is developed under a BSD 3-Clause License and governed by a meritocratic governance model.

Dimemas is a performance analysis tool for message-passing programs. It enables the user to develop and tune parallel applications on a workstation, while providing an accurate prediction of their performance on the parallel target machine. The Dimemas simulator reconstructs the time behavior of a parallel application on a machine modeled by a set of performance parameters. Thus, performance experiments can be done easily. The supported target architecture classes include networks of workstations, single and clustered SMPs, distributed memory parallel computers, and even heterogeneous systems.

For communication, a linear performance model is used, but some non-linear effects such as network conflicts are taken into account. The simulator allows specifying different task to node mappings.

Dimemas generates trace files that are suitable for Paraver enabling the user to conveniently examine any performance problems indicated by a simulator run. The analysis module performs critical path analysis reporting the total CPU usage of different code blocks, as well as their importance for the program execution time. Based on a statistical evaluation of synthetically perturbed traces and architectural parameters, the importance of different performance parameters and the benefits of particular code optimizations can be analyzed.

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.

Scalasca is a software tool that supports the performance optimization of parallel programs by measuring and analyzing their runtime behavior. The analysis identifies potential performance bottlenecks – in particular those concerning communication and synchronization – and offers guidance in exploring their causes.

Paraver is a very flexible data browser that is part of the CEPBA-Tools toolkit. Its analysis power is based on two main pillars. First, its trace format has no semantics; extending the tool to support new performance data or new programming models requires no changes to the visualizer, just to capture such data in a Paraver trace. The second pillar is that the metrics are not hardwired on the tool but programmed. To compute them, the tool offers a large set of time functions, a filter module, and a mechanism to combine two time lines. This approach allows displaying a huge number of metrics with the available data. To capture the experts knowledge, any view or set of views can be saved as a Paraver configuration file. After that, re-computing the view with new data is as simple as loading the saved file. The tool has been demonstrated to be very useful for performance analysis studies, giving much more details about the applications behaviour than most performance tools. Some Paraver features are the support for:

• Detailed quantitative analysis of program performance
• Concurrent comparative analysis of several traces
• Customizable semantics of the visualized information
• Cooperative work, sharing views of the tracefile
• Building of derived metrics

Extrae profiling tool, developed by BSC, can very quickly produce very large trace files, which can take several minutes to load into Paraver, the tool used to view the traces. These trace files can be kept to a more manageable size by using Extrae’s API to turn the tracing on and off as needed. For example, the user might only want to record data for two or three time steps. This API was previously only available for codes developed in C, C++ and Fortran but now it also supports Python codes using MPI.