Resource Sharing (RIB and RIP Frameworks)

We maintain two generic benchmarking frameworks for investigating performance related effects due to sharing of resources such as the processor core, memory architecture (memory caches, translation buffers) or file system among multiple software components.

These are described in more detail in the SPEC Benchmarking Workshop 2009 Paper and Q-ImPrESS Deliverable 3.3.

JPMF: Java Performance Measurement Framework

We develop an extensible framework for performance measurement that allows aggregating configurable amount of data from configurable sources during configurable events.

A short overview can be found in the Q-ImPrESS Tool Manual, more details including initial design can be found in the Ph.D. Thesis of Lubomír Bulej.

Performance Prediction Validation on Random Systems (RPG Framework)

We develop a framework for generating applications randomly from predefined components. The purpose of the tool is to provide many implementations of software systems together with their models, intended for more thorough validation of performance prediction models. More details can be found on the tool's homepage.

Regression Benchmarking

BEEN logo

BEEN is a framework for automating performance experiments in heterogeneous distributed environment.

JOLT: Java Object Lifetime Tracer

JOLT allows collecting lifetime traces of objects on Java heap. The tool works in two phases - first the profiler collects a trace of object allocations and dealloactions (by forced early garbage collections to get lifetimes with required granularity) and then analysis tools can transform the trace to lifetime traces. The tool measures "time" in allocations, and therefore also lifetimes are in number of allocations in between allocation and deallocation.

  • The tool is currently available for Linux and can be downloaded here. README
Logo of Faculty of Mathematics and Physics
  • Phone: +420 951 554 267, +420 951 554 236
  • Email: info<at-sign>
  • How to find us?
Modified on 2016-02-29