Search Results for "git:/git.code.sf.net/p/docfetcher/code"

The experimental source code for the paper, "A Novel Recommendation Approach Based on Chronological Cohesive Units in Content Consuming"

A collection of library code and tools for application execution profiling and performance testing. You can create stopwatches to time select portions of your code. You can measure differences (often to sub-millisecond accuracy) between clocks on different machines. You can log application events in a .csv format for subsequent analysis. You can also generate CPU loading logs in a .csv format.

BEMAP (BEnchMarks for Automatic Parallelizer) is a benchmark used to measure performance for an automatic parallelizer. All OpenCL code benchmarks covered in this project are done step-by-step along with hand-tunning. Each tuning step executional time are measured in details with a comprehensive user interface and help option. The exact implementation in native code (C++) is also provided in each project folder for reference. By using these benchmarks, one may analyze: 1. ...

A framework that compares performance of different pieces of code. Available in four languages: c#.net (SharpKinoko), java (JKinoko), c (CKinoko) and c++ (CppKinoko)

Note: this tool has moved to github, this repository is not maintained any more. See https://github.com/turdusmerula/ftrace GScopeLog is a tool for instrumenting c++ code through gcc. The main purpose is to trace entry/exit point of functions. A status file may be generated to give overview of functions calls and timing informations with minimal impact on performances.

News: since 2015 we continue all related developments within Collective Knowledge Framework: http://github.com/ctuning/ck/wiki Alchemist plugin is a collection of plugins for GCC/LLVM for external and fine-grain code analysis and tuning. It is intended to to extract program properties for machine learning based optimization (see MILEPOST GCC); optimize programs at fine-grain level (such as unrolling, tiling, prefetching, etc); tune default optimization heuristic; gradually decompose program and detect performance or other anomalies; generate benchmarks particularly useful to train ML-based compilers. ...